osch D-Jetronic TPS and Miss at Cruise (hereafter referred to as MaC) Investigation
DRAFT WORKING COPY!
10/2015 R. Kwas [Authors Comments/Annotations and editing highlights present: Yellow are areas needing attention/confirmation of info etc., Red indicates material to be possibly removed
Related Link: http://www.sw-em.com/bosch_d-jetronic_injection.htm Some duplication of info with this page occurs, and needs to be cleanup up, but this page has complete Kerfoot circuit drawings of the D-Jetronic injection, as well as the Volvo installation specifics.
TPS is Misnamed!
Close Visual Inspection of a sample TPS
Electrical and Functional Inspection
Power Up Tests
Bosch D-Jetronic Throttle Position Switch Recommended Maintenance
Resting voltages and Comb Circuit Currents
Throttle Position Switch (TPS) Function
Reproduction Parts for D-Jetronic
TPS Working Diagram
TPS PCB Detail
Bad MaC Explanation No 1
The Edge Triggered Flip-Flop as used in the Bosch-D-Jet
Distributor Contacts of the Bosch D-Jet
TPS - 5 Terminal, Non-Volvo Application
Info, Experience and Thread Excerpts
Note: The information here applies to PN 028 120 039 (Volvo application) specifically, and other PNs generally.
Introduction: As the first widely installed electronic injection system, the Bosch D-Jetronic Injection system was well engineered and was highly reliable, for which drivers of all such equipped vehicles can thank Bosch. It is essentially the grand-daddy of the modern Bosch Motronic fuel control systems. The Electronic Control Unit (ECU) of the D-Jet is strictly a collection of analog signal processing circuits strategically combining inputs (...OK, I suppose one could use the term: Computer...), but there is no digital Central Processing Unit crunching numbers away at light speed as in more modern ECUs. Unfortunately this also means there is no data, error or sensor out-of-range logging implemented, so the D-Jet ECU has no memory which can be interrogated later. One cannot simply plug in a terminal and ask the computer to remember and tell want hurts as on modern digital ECUs! This makes it particularly tricky to verify MaCs and their cause directly. Indirectly, if MaCs occur, one can simply unplug its' connector, take a (gently accelerated) test drive and note if MaCs still occur. Because, yes, the D-Jet System will work just fine without the TPS connected(!), and allow car to be driven...the Manifold Absolute Pressure (MAP) sensor provides ECU with the necessary info to allow metering the correct fuel. So what, one might ask, does the TPS do anyway? Answer: It allows ECU to add enrichment pulses to enable better Throttle Response at times of transition from steady state RPM to something higher (definition of Acceleration).
TPS is Misnamed! The TPS does not even, as the name suggests, give throttle position information to the ECU!...it does recognize the Idle Position specifically, but otherwise, it does not know, and the ECU does not care about absolute position during operation! TPS does sense the Direction of throttle shaft rotation...it must, so that it can call for enrichment pulses upon throttle opening only ...but clearly, no enrichment pulses are required (or desired) on throttle closing...so the TPS is really incorrectly named!...it would more accurately and correctly be called an Idle / Acceleration Sensor!
If after taking a test drive with TPS unplugged and experiencing no MaCs, then plugging the connector back on, and MaCs occur again, that rather points the finger of suspicion squarely at the TPS, but it would be nice to know exactly what part of the assembly...and knowing the exact cause of the MaC, service, adjustment, and maintenance suggestions could be made (rather than the typical "Replace TPS" recommendation, which is just not enough for inquisitive minds).
Acknowledgments and Thanks: The author has performed a detailed analysis and the test presented here, but this was very much helped by others. The extensive work, understanding and explanation of function and theory of operation of the Bosch D-Jetronic System of Frank Kerfoot and Brad Anders, as well as their republished info and material are hereby gratefully acknowledged. Additional thanks are hereby given for collaboration in various forms and input by: Bill Arey, Dave Farrington, Cameron Louvre, Magnus from the Volvoniacs Forum, Dr. D-Jet from the Oldtimer Tips Forum.
Background: The Throttle Position Switch (TPS) of the D-Jetronic injection system is used to give driver throttle input information to the Electronic Control Unit (ECU). There is a lot happening in this component(!)...it contains an Idle Switch, a direction sensitive "Drag Switch", plus, position sensitive sliding contacts on precision PCB Contacts.
The TPS is often blamed for the MaC events, although this symptom can be certainly be caused by any number of other things, including a momentary dropout of ignition, or contaminated fuel, or poor injector spray pattern, or even problems with the Injection Sensing Contacts in the Distributor (however, these are used in conjunction with an edge triggered Flip-Flop within the ECU, similar to the TPS, which is highly reliable and has excellent immunity to false triggers or drop-outs, more on this below!), but MaC indirectly confirmed to be due to TPS, is a well known (and despised!) issue...and in those cases where it is confirmed to be caused by TPS (because it went away after TPS replacement), it would sure be nice to know exactly what in the TPS was the cause, because first, TPSs are not an inexpensive item to replace [actually they must be rebuilt at this point, as they are No Longer Available (NLA). See: Rebuilding TPS, although using a 5 Terminal TPS PN 028 120 047 is an option for the Volvo application!], and second, with a proper and full understanding of the fault mechanism, there may be things which can be recommended to be performed short of replacement (like for instance lubing or preventative maintenance on the internal contacts) which could restore or prolong their trouble-free service.
To make any recommendations, a thorough understanding of the cause of the MaC symptom is necessary, and this is where much goes wrong. The author spent much time critically reviewing and devouring virtually every scrap of info the search engines could find, as well as consulting with all sources of information and first-hand (and maybe second-hand) experience I could find in the automotive world (not only of the Volvo world!)...unfounded recommendations, conjecture and misinformation abound! Advice such as the simple "when a MaC caused by TPS occurs, they are not repairable, and must be replaced", or the somewhat more creative and highly exaggerated "when arcing occurs at worn circuit board tracks...replace TPS" (BAAAD Info, paraphrased) is simply not good enough for me...I need a little...no, a whole lot, more!
If you have experienced a MaC (attack) while driving your D-Jet equipped Volvo (or Jaguar, Mercedes, Porsche, Renault, or VW for that matter!) only after you have taken reasonable steps to check for and eliminate other possible causes by ignition or fuel, may you start thinking that your TPS might be at fault.
As noted earlier, the TPS is not actually required for correct (driveable) function of the D-Jet system, because the Manifold Absolute Pressure (MAP) sensor actually also senses throttle position (indirectly by way of manifold vacuum), except with a modest throttle response only...the TPS was added in the design, to enable some enriching of the mixture to allow a faster, more responsive acceleration/RPM increase. Knowing this (and other Practically Observed Operational Characteristics of the D-Jet System) is actually a troubleshooting advantage in a way, because if a MaC occurs, we can disconnect the TPS to see if it goes away (under the principle of: Something that isn't present/connected, can't hurt you!).
If, while disconnected, MaCs still occur, we obviously need to look elsewhere...but if MaCs have been eliminated by disconnection, we can indeed focus our attention on the TPS, and knowing specifically what exactly to check and repair or service (if possible) and where, is the goal of this investigation. Unfortunately, and this is a disadvantage, the occurrences of MaCs is typically rare and intermittent, so a confirmed association of symptoms and conditions is often difficult to establish.
Close Visual Inspection of a sample TPS:
A very low mileage/virtually new TPS (PN: 0 280 120 039) was opened for inspection and test. After careful removal of the snapped-on cover (thereby confirming this is a late production switch, as early versions had the cover held in place by screws), switch innards can be subjected to a detailed inspection. See also: TPS Versions
|Terminal numbers are cast into plastic connector housing, at Red. TPSs used in the Volvo application have four terminals: 9, 20, 12/14, 17. These number correspond to the Kerfoot drawings.||A five terminal TPS, as used on other D-Jet Applications, See: TPS - 5 Terminal, Non-Volvo Application|
A gold plated circuit board with 3 Sliding Contacts, (gold on gold, with wiping action!) as well as a Single Pole Double Throw (SPDT) Drag Switch are evident. The circuit board is secured to a metal baseplate with one rivet (not involved electrically), and one alignment dimple, and makes connections to the main 4 pin connector by way of 4 tin plated copper Pressure Contacts. These 4 spring contacts are measured to exert a significant preload (about 1 pound) to each respective circuit board pad, by way of a formed dimple (tin on gold, no wiping action, but very good thermal and vibration compensation).
Pressure Contacts (qty. 4).
Sliding Contacts (qty. 3). Low angle edge view showing riveted contact beads.
Sliding Contacts are measured to exert about 8 ounces of preload on the board each. By inspection, all Sliding Contacts are the same size, and it is observed, that the sliding Comb Wiper Contact is not wide enough to bridge the two Interlaced Comb Conductors (confirming that contact to Interlaced Comb Conductors is alternate not bridging). Actual Sliding Contacts are gold (plated?) and riveted onto the Spring copper finger (gold on copper, and gold on gold, with wiping action to PCB).
The Drag Switch consists of a spring copper contact riveted with a gold/brass(?) rivet, to the fixed end of Wiping Contact No1 (TPS term 14), which contacts one of two gold posts, each riveted with a gold/brass(?) rivet, to the fixed end of Wiping Contacts 2, and 3 respectively (copper on gold, no wiping action!). Separation of Drag Switch contact in the post area is smaller than the separation of contact Posts, resulting in a dead zone where Drag Switch contact is between the two and not contacting either. PLACEHOLDER: for angular measurement of degrees of Throttle shaft rotation associated with breaking contact to one of the Comb Contacts, to contact to the next.
Top view of Drag Switch contact and Contact Posts. Throttle shaft is shown relaxed, so Drag Switch moving contact is in neutral center position not contacting either post.
Drag Switch Conditions as a Function of Shaft Rotation:
Throttle Opening = Counter Clock-wise (CCW)
Throttle Steady = No Rotation
Throttle Closing = Clockwise (CW)
Throttle Opening / Shaft Rotation CCW. Comb Contact of Drag Switch is closed, connecting chassis connection of moving Contact to Comb Wiper, which in-turn can connect to either Comb Conductor (or neither, if in between conductors). In order to supply inputs to the RS Flip-Flop which in turn leads to enriching pulses being generated, the Comb Contact must also be closed.
Throttle opening position of Drag Switch (during CCW rotation). Select contact is in contact with Comb Contact Post.
Steady Throttle / No Shaft Rotation. Moving Contact of Drag Switch is between end states (unconnected).
Drag Switch Select Contact - shown in central position, not connected to either post!
Throttle Closing / Shaft Rotation CW. Throttle shaft is in Idle Position, contacting Post, but Terminal 17 is connected to chassis ONLY when at end stop due to Conductor position on PCB and break in conductor (Reference: PCB Detail below).
Drag Switch shown in Idle Position with Drag Switch Frame up against mechanical stop (Green). When installed on the Manifold, it would be positively held there by spring force.
Side view of (gold plated) Drag Switch Posts showing riveted connection to Wiper Contacts 2 and 3. Last digits (...120 039) of Bosch PN are visible on underside of baseplate.
A Spring Washer between PCB and Drag Switch frame creates a Friction Clutch to hold Drag Switch frame with a slight constant friction and prevent Drag Switch frame motion on direction changes, until motion transfer post is contacted for the other direction. This causes about 2degrees of intentional lost motion on direction changes between Throttle shaft and Drag Switch, resulting in 2degrees of mechanical Hysterisis. Because of the Drag Switch, TPS is able to have direction dependent functions.
From top to bottom: Drag Switch frame, Clutch Washer (Red, which assures a level of friction between the two moving parts), PCB.
Electrical and Functional Inspection:
Results of a detailed inspection of a TPS:
Link to PCB detail confirming Term 17 can only be grounded at Idle (fully CW position): PCB_Detail
Power Up Tests:
A simple test fixture with current limiting resistors and LED indicators for the contacts was constructed to allow power up tests with TPS:
TPS on Test: LEDs simultaneously display condition of Terminals 9, 20, 17, as Drag Switch hub is turned (LED ON = connection to ground is complete). Little red and white strips are thin insulators used between contacts, to isolate connections for tests and simulating contact failures.
Observations: The alternate switching action of terminals 9 and 20 confirms, Comb slider makes contact one at a time to provide alternate negative trigger edges to ECU to trigger SET and RESET states of Flop. Term 9 is the first to be grounded when TPS comes away from the Idle (fully CW) position, so the author's convention will be to call this the Set Input of the Flop..
Idle terminal (17) is grounded ONLY while shaft is in the fully CW position, and up against the mechanical stop, because of PCB layout. This means it is unlikely...NO, ABSOLUTELY IMPOSSIBLE, that Idle being somehow inadvertently selected at part throttle/cruise can be the cause of a MaC (this is a favorite explanation often heard, but herewith inarguably disproven / discounted! Call this Noteworthy Conclusion No 1 of this Investigation.). (More on this here! LINK: Bad MaC Explanation No 1)
Function of Enrichment Flip-Flop: Shown below are circuit conditions for RESET (Shaft fully CCW, and also the alternate condition subsequently, as Comb Wiper sweeps up the Comb), and SET (first state-change when shaft turn CW, also alternate state as Comb Wiper sweeps up the Interlaced Comb Conductors).
Interaction of TPS with ECU: The test fixture was modified to replicate the ECU function by adding a two transistor SR Flip-Flop in the same configuration of the ECU (T904, T905, from the Kerfoot drawings, with component values being supplied "Dr. D-Jet"). This allowed the observation of the latching function (Q and Q/ outputs) of the enrichment FF, beyond just the immediate TPS output conditions.
TPS Investigation Conclusions:
Tests are on-going and not yet complete, so I have yet to reach final conclusions, and unfortunately, no subtle little flaw has jumped out and made itself known as the cause of MaCs. What I am certain of already is that the TPS design is sound, construction is robust, and it will only benefit from periodic contact cleaning and lubing as described in Recommended Maintenance below.
As a result of the preceding investigation and findings, the author presents the following:
Bosch D-Jetronic Throttle Position Switch Recommend Maintenance: This procedure can be applied with TPS in place, not affecting adjustment.
Note: This procedure has been put together with the utmost care, and I absolutely stand by it having successfully performed it on my own vehicle, with good result, but must be used by other individuals in conjunction with normal, cautious shop practice! I cannot be responsible for the specific actions of others!
TPS is a precision device! Once inside TPS, do not bend or otherwise damage or change sliding or closing contacts!
With Ignition OFF, Remove Cover. Early style TPS Cover is secured by four corner screws, later style TPS has a snap-on cover (remove by gently levering to defeat the two holding keys between base and plastic cover. Do this only when warm! Cold plastic is brittle!).
Once cover is removed, Clean Comb Contacts with a cotton swab saturated with a few drops of Deoxit D5 (Use nothing more abrasive! The microns thick gold plating you abrade off with a pencil eraser today (exposing copper which will corrode(!), makes it that much sooner you will have to replace TPS tomorrow!). Move throttle to allow access to all areas of Comb conductors. No further cleaning of comb or other wiping contacts is necessary. Remove any lint cotton Swab may have left behind.
To Clean Closing Contacts, of Drag Switch, saturate the corner of a piece of typical office paper with Deoxit D5, and run this paper (flat, and mildly abrasive) between closing contacts several times. When cleaning is complete, sweep Throttle through its complete operating range several times, and generally flush and lubricate the areas which have been cleaned with a light application from the spray can. Catch overspray and drops with a rag.
For maximum protection and effectiveness, and longevity without MaCs, Repeat this treatment every year, or 10kmiles. Also lube Throttle Bushing.
TPS Idle Checking and Adjustment:
1. With Ignition OFF and Throttle at Idle, disconnect TPS connector and verify continuity between TPS 17 and 14 with a low test current, continuity checking device such as a DVM. If open, adjustment to this condition is necessary! This connection must show OPEN any time Throttle position is not Idle. Replace TPS connector. Cleaning and preventatively treating connector male and female pins at this time with Deoxit cant hurt!
2. With Ignition ON, slowly sweep Throttle through its operating range. 20 Enrichment clicks should be heard as Injectors are activated. There is no adjustment of this function possible, but if 20 clicks are not heard, something is is not right! Start by performing TPS maintenance above.
Reference Information (or noteworthy info I ran across or developed during this investigation):
A excellent D-Jet site (in German) operated by Dr. D-Jet (Volker): https://oldtimer.tips/en/d-jetronic
From Volvoniacs: http://www.networksvolvoniacs.org/index.php/D-Jetronic_-_Beschreibung_und_Fehlersuche [Translation required]
Das Ohmmeter zwischen die Klemmen 14 und 17 anschließen. Der Wert soll 0 Ohm (Durchgang) betragen. [Check of Idle Contact being closed.]
Bei "unendlich" den Drosselklappenschalter neu einstellen, sonst gibt es kein Leerlaufsignal und damit keine Schubabschaltung. Wenn trotz korrekter Einstellung kein Leerlaufsignal kommt, muss der Drosselklappenschalter erneuert werden. Bei etwas Gasgeben muss das Leerlaufsignal weg (anzeige unendlich) sein.
Achtung! CO im Leerlauf kann nur eingestellt werden, wenn das Leerlaufsignal durchgeschaltet wird.
Beim Fehlen des Leerlaufsignals kommt es auch zu erhöhtem Kraftstoffverbrauch, da die Schubabschaltung nur mit Leerlaufsignal und über 1800 U/min arbeitet.
Drosselklappenschaltersignal 1 (9 1/2 Impulse) und Drosselklappenschaltersignal 2 (10 Impulse) können nur mit der "Blauen Bosch Kiste" geprüft werden. [Or SW-EM tester! ]
Ruckeln bei Konstantfahrt Konstantfahrruckeln liegt normalerweise am Drosselklappenschalter, alte Ausführung mit geschraubtem Deckel. Diesen dann erneuern. Es kann aber auch mit der neueren Ausführung (Deckel nur geclipst ) auftreten. In diesem Fall sind die Kontaktbahnen durch den Läufer und durch Spiel der Drosselklappenwelle verschlissen. Daher sind vor der Erneuerung des Drosselklappenschalters die Buchsen und unter Umständen die Drosselklappenwelle zu erneuern.
Baaaad, Misinformed, Superficial and Highly Speculative Info with no Basis in Fact ( my mother told me not to say anything if I didn't have anything good to say, but I just had to speak out against this baaad info...sorry!):
HARDLY! With Voltage and Current being highly limited to 12V/18mA, (See Calculations below, also: Resting voltages and Comb Circuit Currents. This value is a quarter of the current level at the Fuel Sender in the Tank where where Sparks would present an even bigger problem), there is thankfully no sparking and accompanying carbonization going to occur in the D-Jet TPS. (Reference: Fuel Sender and Gauge Electrical Calculations),
Calculations for TPS Sensor Wiper Contact:
Condition 1 (Comb Select Contact, as well as connection to one of Term 9 contacts, Closed):
I Comb Wiper= 12 / R tot = 12 / (390
+ 270 = 660 Ohms) = 18mA
Condition 2 (Comb Select Contact, as well as connection to one of Term 20 contacts, Closed):
I Comb Wiper= 12 / R tot = 12 / (390 + 270 = 660 Ohms) = 18mA (Same calculation as for Condition 1!)
Absolute worst case voltage possible on Wiper to Comb Contacts: 12V (Nominal system bus voltage)
Maximum Wiper Current occurring: 18mA
Conclusion: With both Voltage being low and and Current at the Comb Wiper being highly limited to these low levels, there exists no risk of spark erosion of PCB contacts.
One very mysterious intermittent fault caused by the throttle switch is an occasional "bucking" while driving at constant speed. This is caused by the switch cutting of the fuel and then turning it back on. This is certainly what it feels like, but without substantiating data or info must be considered to be unsubstantiated conjecture!
Resting voltages and Comb Circuit Currents:
Sheet 2 of ECU with Input / Output shown. Component values of Enrichment Flip Flop have been added to allow voltage and current characterization and calculations.
Throttle Position Switch (TPS)
Early type TPS easily recognized by the radial angle of comb contacts, and separate drag and Idle switches. If this was a picture of an actual switch, discrete wires between I/O connector and innards would be evident.
Translation (Credit?...looks like one of those on-line translators...lots of literal translation gobbledygook, and tough to understand...I should/will rework this sometime to improve understandability):
3.4 Throttle Position Switch
Fig. 7 shows the throttle valve switch in the plan view with cut cover. The throttle idle micro-switch of forming contacts a by the lever c connected rigidly with the throttle valve wave with closed throttle valve are operated. The disk b is concentrically arranged to the hub of the lever c. One wave feather/spring between the disk and the housing bottom provides for a defined friction opposite the housing. When opening the throttle valve (in the picture against the clockwise direction) the disk b is held by the friction first, the lever c moves against the contact record d drags switch and closes this. After a small overflow route for the contact record dashed a drawn driver of the lever c carries the disk forward b. Thus the two slide with the disk b firmly connected pickup shoe on the contact plate e. With closed contact record d the teeth of the two edge contacts in the change are connected electrically with the continuous pickup shoe path. Switching a comb on the other hand has the advantage in relation to easy switching on and off that during uncertain contacting or bouncing the contact no additional enrichment impulses in the controller are released.
Translation R. Kwas: PLACEHOLDER
Adjustment of the TPS
The following good info is copied from Volvoniacs, and still needs translation: http://www.networksvolvoniacs.org/index.php/D-Jetronic_-_Beschreibung_und_Fehlersuche
Er sitzt an der Drosselklappe und gibt ein Leerlaufsignal (für Schubabschaltung) und Drosselklappenstellungssignale über ein Potentiometer an das Steuergerät.gleichzeitig auch Drosselklappenstellungsgeber .ZV kontakt 1 (9 1/2 Impulse) .ZV kontakt 2 (10 Impulse)
Das Ohmmeter zwischen die Klemmen 14 und 17 anschließen. Der Wert soll 0 Ohm (Durchgang) betragen.
Bei "unendlich" den Drosselklappenschalter neu einstellen, sonst gibt es kein Leerlaufsignal und damit keine Schubabschaltung. Wenn trotz korrekter Einstellung kein Leerlaufsignal kommt, muss der Drosselklappenschalter erneuert werden. Bei etwas Gasgeben muss das Leerlaufsignal weg (anzeige unendlich) sein.
Achtung! CO im Leerlauf kann nur eingestellt werden, wenn das Leerlaufsignal durchgeschaltet wird.
Beim Fehlen des Leerlaufsignals kommt es auch zu erhöhtem Kraftstoffverbrauch, da die Schubabschaltung nur mit Leerlaufsignal und über 1800 U/min arbeitet. ???
Drosselklappenschaltersignal 1 (9 1/2 Impulse) und Drosselklappenschaltersignal 2 (10 Impulse) können nur mit der "Blauen Bosch Kiste" geprüft werden.
Der Schließwinkel der Auslösekontakte unten im Verteiler ebenfalls.
Als letztes CO einstellen.
Sollwert: 2% CO
Eingestellt wird ab Baujahr 1971 am Poti am Steuergerät, vorher (1800 E mj 1970)ist das Poti nicht vorhanden.
Das Poti ganz nach rechts und dann zwei Klicks nach links drehen, die Kontrollmessung muss jetzt 2% CO ergeben. Wenn nicht: Nach rechts = fett, nach links = mager.
B20: Die Kontermutter der konischen Schraube unter dem Lufteinlass vorne am Ansaugkrümmer unten aufkontern.
Schraube reindrehen = niedrigere Drehzahl
Schraube rausdrehen = höhere Drehzahl
Wenn die Einstellschraube bis zum Anschlag reingedreht ist und die Drehzahl trotzdem zu hoch ist, kann es folgende Ursachen haben:
Den Drosselklappenschalter lösen, die Anschlagschraube entlasten und die Drosselklappe schließen. Dann die Anschlagschraube beidrehen, bis die Drosselklappe anfängt,sich zu öffnen. Dann die Anschlagschraube eine HALBE Umdrehung reindrehen Dann den Drosselklappenschalter einstellen.
Ruckeln bei Konstantfahrt Konstantfahrruckeln liegt normalerweise am Drosselklappenschalter, alte Ausführung mit geschraubtem Deckel. Diesen dann erneuern. Es kann aber auch mit der neueren Ausführung (Deckel nur geclipst ) auftreten. In diesem Fall sind die Widerstandsbahnen [...this is poor use of terminology...Widerstand means Resistance, and there is no variable Resistance here like a Potentiometer, only Contact/No-Contact areas. A better term would be "Kontaktbahnen". ] durch den Läufer und durch Spiel der Drosselklappenwelle verschlissen. Vor der Erneuerung des Drosselklappenschalters die Buchsen und Drosselklappenlagerung erneuern.
More Links at the Volvoniacs on the TPS: http://www.networksvolvoniacs.org/index.php/Spezial:AWCforum/st/id1928/limit:0,15/#last
D-Jetronic DKS Drosseklappenschalter: http://www.networksvolvoniacs.org/index.php/Spezial:AWCforum/st/id4878/drosselklappenschalter.html
Reproduction Parts for D-Jetronic: (links/availability not checked as of 10/2015)
Neue Stecker und Gummitüllen für D jetronic B 20 E/F Robert Bissler bei repro-parts.de
Nachproduzierte Platine Drosselklappenschalter . bei office bei v 1800.org
|Throttle Switch||039 906 111 A||0 280 120 032||2.0L||1973 - 1976||No longer available new. If you go to http://www.914world.com , and look up user "davesprinkle", he's been fabricating a kit to replace worn TPS circuit boards, so that you can rebuild your TPS.|
Replacement PCBs for TPS from:
I have not had one of these in my little hands for a critical inspection, but they look good from where I'm sitting.
Lube with Deoxit D5 at installation to prolong life and reduce wear
Picture Source: http://members.rennlist.com/pbanders/TPS%20Contacts%2003.jpg
Here, although not perfectly clear,
at Red, damaged gold plating and metal displaced by continuous wiping of the
moving contact is evident. Lubrication would delay/reduce such
displacement! I'll try to get an even better picture:
Link to D. Farrington pictures: https://drive.google.com/folderview?id=0BwOnJ74Oi_8xY2dKV1FVYnBBMXc&usp=sharing
Throttle Position Sensing Switch
In exercising the search function, detailed info on the TPS is sparse, and in the case of this publication, even highly imaginative, totally speculative...incorrect, just plain BAAAD information, and not elementary my dear Watson! It is simply not possible for the sliding contact or alternate comb contacts to "begin to arc as the throttle is held at a steady position"...resistive voltage divider networks for T904, T905 (Throttle Flip-Flop, see: Resting voltages and Comb Circuit Currents above) limit current and voltage to a miniscule level (18mA) which would never be able to produce an arc! Lack of lubrication between the board conductors and non-ending motion of the sliding contact can cause wear (I'll call it abrading of the conductor material and microscopic level displacement of particles of conductive material to next to the original conductor), but no arcing or carbonization ever takes place!
TPS related Excerpts from the Anders circuit analysis ( http://members.rennlist.com/pbanders/ecu.htm ) my highlights/comments :
Acceleration Enrichment (AE) - Sheet 2
The AE circuit provides immediate and delayed enrichment when the throttle is opened for acceleration. Throttle opening is signaled by the TPS through alternating ground signals from two inter-digitated traces that are each connected to the inputs of a flip-flop (see sheet 2 of the schematic for a drawing of the switch details) [I call them Interlaced Comb Conductors, more consistent with Bosch's nomenclature, but we're clearly talking about the same thing]. A drag switch in the TPS prevents these signals from being sent when the throttle is closing [Correct!...and important to dispelling bad MaC explanation No 1!]. The outputs of the flip-flop are sent to two edge detectors, whose outputs are combined and sent to pulse shaping and narrowing circuits. The pulse shaper provides immediate injection pulses to the IL. The width of these pulses is independent of engine speed and load, and from Kerfoot's schematic, they are about 1.5 ms in duration.
The IM sets the mixture (adjustable with IA) during idle operation. The output of the IM is combined with the output of the SC to control the voltage threshold in the PL for turn-off of T201. When the throttle is open, the IM appears as an open circuit and the SC controls the PL threshold. Note that the IM is effective ONLY when the IS on the TPS is closed!! Adjustment of the IA has no effect on mixture when the throttle is open.
Throttle valve switch full-load contact (only for VW T3's from 1972)
d: TPS Idle switch, full-load switch (if present), and immediate accelerator track switches (3-way switch to accommodate all TPS types)
Adjustment procedure for TPS: http://www.pelicanparts.com/techarticles/914_cooley/throttle_position_calibrate_17-18.jpg
Numbers from Ebay, Dutch seller: peuver
Early Type: PCB Type-17 for Bosch: 0 280 120 018 / 0 280 120 021 / 0 280 120 039 [Comment: ...039 number seems to be incorrectly associated with early type...should be associated with late type. See: Rebuilding TPS ]
Later Type: Type-03 for Bosch: 0 280 120 007 /0 280 120 011 / 0 280 120 012 / 0 280 120 026
Questions for Anders:
What is difference between 9,20,14,17 connector on Volvo and 9,20,12/47,17,2/14 of VW and -,17,12,-,9 of Porsche when PCB is identical?
TPS adjustment procedure calls for Throttle to be fully closed, but it would seem that Idle switch closes any time shafts turns CW. Not true. Because of interruption in conductor, Idle Switch can ONLY have continuity to gnd at fully CW shft position.
Does he have drawing which shows the connections of WOT switch of VW versions? How does WOT switch interface to early ECU, or does it not?
Get resistor values for Enrichment Flop.
During deceleration, above 1500 RPM, throttle switch cuts fuel supply off and below 900 RPM, fuel supply is turned on.
E-mail to Anders:
I have seen your excellent work on the D-Jetronic (for years now) and have to compliment you on your dedication and work, and efforts to present it on your site!
I am involved with the Volvo clubs, and as you are undoubtedly aware, the D-Jet was used in several models from this manufacturer. I am also an EE (with heavy hands-on factor as yourself) and am currently very interested in understanding the TPS and really getting to the root-cause of its' well known "cough at cruise" symptom. I have some thoughts on this I would like to discuss with someone who I can really "talk turkey" with, and who knows the difference between a negative edge triggered Flip-Flop and a hole in the ground...so I took a chance and googled your name for Phoenix, and came up with a number which I left a message on this past Sunday...I don't know if you got the message, but out of the blue like that, I might not respond to a message from a non-Porsche guy either, but I ask again if you might be so kind to allocate a few minutes of your precious time to discuss the finer points of the TPS with me, a fellow vintage car guy and D-Jet friend, I would very much appreciate it...and you might just get something out of it yourself...with the Cough at cruise symptom being pervasive, I'm surprise at the lack of info and root-cause and corrective action info out there...
...and just plain BAD information doesn't help...I had a real laugh when I ran across this:
Link to Baaad Info
Having studied the switch at length (including the VW version with WOT contact), I actually don't think the comb contacts have much to do with the cough at cruise at all, because according to my initial inspection, the sliding contact which supplies the comb conductors with ground and thereby a Neg edge to trigger the Flop is smaller in OD than the insulated space between combs (comb-wear and conductor smearing notwithstanding)...and even IF an extra trigger occurred, the resulting mixture enriching is unlikely to cause a misfire...I believe it is instead the hysterisis (or lack thereof) of the drag-switch which causes the cough (I've seen "2degrees" shaft rotation) because of a momentary switching to Idle (which would result in a lean stumble). I think this is rather the root-cause (not "arcing" or "comb-conductor wear"), but again, this is what I would dearly like to discuss in depth with you. My goal is to generate the complete and definitive documentation on this issue, including tests, adjustment and maintenance recommendations. I think ALL D-Jet users, Volvo, Porsche, and VW alike could greatly benefit from such a documentation, and you are clearly the expert I would like to talk to...
Greets from Connecticut and Regards,
TPS Working Diagram, tying TPS and ECU functional info together:
Consideration of Comb/Idle Switch / Theory of Operation: There is a lot going on in the TPS, both mechanically and electrically! It should be noted that contact connections are a function of BOTH Shaft Rotation Direction AND Absolute Shaft Position. It should also be noted that although Drag Switch is in a SPDT configuration, which normally would have only two states, in this case, it has three states, because it has no fast acting (snap) mechanism to toggle between the two end-states, but is directly actuated by the slow/dead-slow/non-moving Throttle Shaft, the unconnected, in-between condition must also be electrically considered (because it can even exist continuously and intentionally), making a total of three possible switch states (and that is the reason the middle, unconnected center position is also included as NC in my drawing of the "Drag Switch, Select Contact"):
TPS PCB Detail:
Bad MaC Explanation No 1. This detail of the PCB dispels Bad MaC explanation No 1. and presumed cause of the MaC...that the Idle switch is somehow closing and changing fuel delivery to Idle Level while at cruise. It is clear from the picture that this in absolutely impossible...Drag Switch Idle contact may indeed close, but electrical path is open anywhere but at angular Idle position because of Slider on PCB being in series and only able to be closed when at angular Idle also...so: Idle Continuity is ONLY possible at Idle position!
A second hypothetical explanation for MaC considered: At transition from no shaft rotation to CCW shaft rotation...Comb Contact of Drag Switch is not closed when a slight acceleration has been called for by a slight increase in gentle throttle opening, (which, because it is so slight, might be recalled by driver to be "constant throttle"). Since the pressure of closing Comb Contact of Drag Switch was dead slow, it might not have made good contact and continuity...Result: A "missing" enrichment pulse(s) resulting in a possible lean stumble. But ...I don't give this explanation a lot of credence...as a counter-argument to this hypothesis, it is known that the Manifold Absolute Pressure (MAP) sensor is able to detect and enrichen mixture adequately for slow throttle open speeds (even without any TPS installed at all), so this is an unlikely explanation.
Dragswitch Conditions as a Function of Absolute Shaft Position:
Throttle Fully Closed = Fully CW
Throttle Mid Travel = Anywhere but Fully CW
Shaft Fully CW. Idle Position.
Shaft Mid Travel. Partial to full Throttle opening. Note: Contrary to misinformation out there, TPS or ECU has no way of counting the comb pulses, or tracking the absolute position, or knowing (or caring and being affected by!) when Throttle is fully open (Wide Open Throttle, WOT). Period! [The 5 terminal version of TPS [VW and Cadillac applications, do have a WOT feature. Reference: TPS - 5 Terminal, Non-Volvo Application (PN 028 120 047, but this does not apply to Volvo application, or even using a 5 Terminal TPS on a Volvo!].
Slider Contacts: Both contacts of Continuous Sliders (1, 2), and Comb Slider Contact (3) are continuously preloaded by their own (probably beryllium copper) spring to single ounces of force. Comb slider contact is less than Comb Contact width, therefore cannot bridge comb contacts, guaranteeing the alternate inputs required by enrichment Flip-Flop. It should be noted that the Enrichment Flip-Flop is by design absolutely immune to false trigger inputs. See Reference Info: Edge Triggered Flip-Flop
Second, direction of shaft turning must also be taken into consideration (mechanical Hysterisis).
|Contact||Direction of Shaft Rotation||Open||Closed||Failure mode and Notes|
|Comb||CCW||Normal Active Condition|
|Idle||CW||Normal Active Position|
My detailed consideration of the function of TPS in support of an explanation of the MaC symptom:
1. I don't think it is even related to the double comb and slider contact...in the first place,
Diameter of the slider contact which sweeps comb conductors is smaller than the spacing between comb conductive teeth...this means that slider can never make contact across the two combs, or even that it makes contact with either comb at any one time. Sliding contact can actually be between the two combs also, not making contact with either! This is also consistent with the negative edge triggered Flip-Flop inputs the two combs are connected to at the ECU, where the two inputs would be triggered alternately in operation (Note 1). Therefore, three contact conditions can exist: Contact with one comb (term 9), OR contact with the other comb(term 20), OR between the two, and not making contact with either comb(open). Resulting circuit function: Flip-Flop T904,T905 remains un-triggered, and in a stable condition, holding last state.
Note 1. As the Flip-Flop is an edge triggered circuit, the only time Flip-Flop changes state is during CCW (throttle opening) rotation AND when the next comb conductor is contacted.
2. Drag Switch (internal to TPS), provides ground connection to Comb Slider when shaft is turned CCW (only).
3. Idle Switch (grounds terminal 17, when Throttle shaft is fully closed CW)
Note 2. Since Dragswitch powering comb slider is in essentially a SPDT configuration with Idle Switch, providing ground to either is exclusive of the other. In other words, only one of the two circuits is enabled, at any one time, as a function of Throttle Shaft direction. Also there is a deadzone or Hysterisis between the connection of the two directions. When a direction change occurs, the contacts of the SPDT do not change instantaneously, because in the first place it is not a snap-action mechanism, and secondly the shaft moving them is turning dead slowly.
I therefore hypothesize that it is not the wearing of the PC board in the comb area which is responsible for the "cough" symptom, but normal ECU action in the deadzone during a (slow) switchover which is responsible. It might be as simple as the idle contact being engaged at cruise [IMPOSSIBLE! See: Bad MaC Explanation No 1 ] through failed mechanical hysterisis (cutting off fuel) then the idle contact opened again, the engine reaction might be a momentary lean stumble. My research will concentrate on trying to substantiating this, as well as supporting the use of Deoxit D5 as a lubricant and preserver of the internal contacts.
Recommendation for periodic cleaning and protection, lubing and preservation of all versions of the Bosch TPS:
I do recommend periodic cleaning and lubing, and protecting of the TPS, but with nothing more abrasive than a cotton swab whetted with Deoxit D5 (the micron thick gold plating you abrade off with a pencil eraser today makes it that much sooner you will wear through it and have to replace it tomorrow!) For Drag Switch contacts, wet the corner of a piece of office paper with a drop or two of Deoxit, and draw through between contacts. Finally flush with Deoxit.
The Edge Triggered Flip-Flop in the Enrichment Circuit. Using this circuit configuration is brilliant engineering by Bosch, and why it is therefore not possible for the comb contact to be causing the "Miss at Cruise" issue!
Bosch was indeed clever in their system design!...their use of an Edge Triggered Flip Flop (T904,T905 of the Kerfoot drawings) has complete and total immunity to false triggers from the Comb Contact! Looking at the Comb Contact specifically, and the fact that each side triggers one input of an edge triggered Flip-Flop, it is clear that a worn, and intermittent (or a bouncing multiple contact), CANNOT be the cause false inputs ...the author can say this with a high degree, no...absolute(!) certainty, because of the nature of an edge triggered Flop.
A Flip-Flop, also called a Latch, is a circuit configuration often used in control circuits as a bounce eliminator where electromechanical switches are used as inputs.
When electro-mechanical contacts are closed, they typically bounce microscopically, giving multiple inputs to the circuit, which, if fast enough to respond to these (electronics certainly are!), can cause errors, if not allowed for, and accommodated...and the Flip-Flop does this perfectly! By using this circuit configuration at the input, the downstream control circuit will see only the first contact closure, ignoring any and all subsequent closures, contact bounces or intermittents caused by PCB wear, because these are effectively "locked-out" and ignored, thus control errors are prevented.
Generic Flip-Flop circuit
Detail of Flip-Flop Operation as it applies to the TPS: Notice on circuit of Flip-Flop, because of the cross-coupling of the two transistors, once the FF has changed state as a result of the first valid low going edge to occur, presume closing switch at A (caused for instance in the Bosch D-Jet, by the first continuous connection through Comb Slider Contact, Comb Contact, Continuous Slider, and finally Continuous Pressure Contact), it doesn't care (or react to...is therefore totally insensitive and immune to!) additional triggers on the same input...because as soon as this first trigger causes QA to turn off and to assume that stable state (and results in one single enrichment pulse due to the function of the downstream circuit), it effectively locks out that input and renders any and all further negative edges on it as ignored (called a "Don't Care" operating condition). Once the door is closed, it cannot be closed again!
Therefore, the only, and next trigger which can result in a Flop state change must occur on the other Flop input (B, since even if it were to occur again at A, it would not be valid due to the ignored input condition)...and to get a valid input to the B Flop input would take a large physical shaft rotation to the next and alternate comb connection, compared to the small wear area that the moving contact is in at the time. Again...a brilliant design with superior tolerance to wear! Once the next valid triggering Edge comes at the B input, the FF changes to the other stable state, and this results in the next enrichment pulse due to the function of subsequent parts of the circuit...and so on and so forth up the 20 possible edges of the Comb Contact area.
Link to further Flip-Flip information on a popular reference site: https://en.wikipedia.org/wiki/Flip-flop_%28electronics%29
Interesting to note is that the Distributor Contacts of the Bosch D-Jet, Trigger connect to an identical edge triggered Flip-Flop (T251,T252) as the Enrichment Flip-Flop, so this ECU input is therefore equally robust in its immunity to false triggers. This circuit understandably has no provision for a Reset.
Extract from the Kerfoot ECU Wiring Diagram Sheet 1, showing a Flip-Flop input circuit for the Distributor Contacts.
TPS - 5
Terminal, Non-Volvo Application (PN 028 120 047):
(Cadillac application (need to confirm model and availability details!) info and picture provided by D. Farrington.)
The 5 Terminal version has Wide Open Throttle (WOT) sensing output added at Blue. An additional internal Slider Contact at Red makes contact and grounds the PCB conductor when in the fully CCW position (only), shown at Green, and this signal is brought out to Terminal 2/14.
The 5 Terminal TPS (PN 028 120 047) can be used, unmodified, in the Volvo application! The WOT sensing as well as Terminal 2/14 are strictly additional to, and only along for the ride in the TPS in terms of the Volvo application. Even the Terminal is located such that the 4 Terminal Connector found in a Volvo, can plug-in without issue! The only thing which should be observed when installing a 5 Terminal TPS into a Volvo is that the connector should be oriented to the left as shown above, so that the WOT Terminal 2/14 is not connected.
Just as an idea, Terminal 2/14 could conceivably be used in conjunction with additional circuitry in the Volvo application, to disable the Air Conditioning Magnetic Clutch at WOT, and shed its significant mechanical load on the engine (extra VROOOM when needed!).
TPS Functional drawing with 5th Terminal , WOT Option added.
Below, a TPS (PN 0 280 120 039) with WOT option on the PCB contact, but no WOT Slider or Terminal 2/14 to bring this contact out. This suggests the replacement PCB was installed at some point.
Again, this rebuilt 4 Terminal TPS would be suitable for use in a Volvo application.
PCB (only) with provisions for WOT sensing at Green, but no second WOT slider at Red or Terminal 2/14 at Yellow!
See also: TPS - 5 Terminal, Non-Volvo Application
Info, Experience and Thread Excerpts: [Yellows are addition Comments/Corrections which were not part of the original posting]:
[From www.Brickboard.com]... in response to: "'73 p1800 es throttle position switch 1800" posted by posted by "flaps" on Wed Feb 25 18:18 UTC 2004
On the earlier TPSs, the plastic cap just snapped on, and just as easily snaps off to reveal the innards...and IF you have on of those, go ahead and remove it from the throttle housing (make witness marks to help getting it back in the same place on reassembly), then carefully pop the top off, blast clean with spray solvent, and inspect:
1. PC board foils (especially the area where the slider sits almost forever at cruise throttle openings) if edges of conductors show excessive wear at the edges, that's the classic cause of hiccups at cruise. [My detailed research has shown this as just not true! The Bosch design is brilliant in its' use of an edge-triggered Flip-Flop! Reference: D-Jet Theory of operation. TPS. Interlaced comb conductors are inputs to an edge-triggered Flip-Flop which is totally immune to inadvertent retriggers!] If worn away, there's not much to repair for even the most ambitious Brickboarder...replacement is about the only option...[replacement TPSs are no longer available, but reproduced PBCs are, with which a TPS can be refurbed are, See Reference Info!] that's why I've considered making an optical (ZERO contact) replacement for those damned switches... (cost?...but it's got to be better than those damned things... who's interested? If it doesn't look too bad, lubricate liberally with Deoxid D5 *[I still maintain this as the best practice!], pop cap back on and reassemble, using witness marks as a guide (lubing the shaft couldn't hurt).
* see also: http://www.intelab.com/swem/gastight.htm
2. Mechanical switch contacts of idle sensing switch.
I have no experience with late, replacement TPS, which I understand it is not possible to open (maintenance free?...BS!...more like unmaintainable!...just like the car batteries)...somebody needs to be slapped for that!). Plastic cover of later switch (no screws) can indeed be carefully snapped off to gain access to innards for cleaning and preventative service...but it is recommended to do this not when cold and brittle, but when warm, and plastic cover allows bending.
[From www.Brickboard.com]... posted by John Mc on Fri Nov 28 20:30 UTC 2003 [used with permission, because I couldn't have explained it better myself! Authors comments added.]. Link to Thread throttle 1800 [E]: http://www.brickboard.com/RWD/index.htm?id=714766
PS - those contacts in the throttle switch. The
computer doesn't really need to know the actual position of the throttle
(absolutely correct! See:
Misnamed!) - it
gets all the info it needs to know about the air going into the engine through
the air temp sensor and the manifold air pressure sensor. That switch does two
1) Idle circuit: To squeak by early emissions laws the computers had a special idle circuit which was individually tuned on the line with an adjuster on the box. This fine tune adjustment is ignored when the throttle is anywhere but closed. So there are a couple of contacts in the switch to indicate this.
2) 'Accelerator pump': When you suddenly open the throttle on any car the engine needs just a bit of extra gas or it will stumble slightly. To do this they have the throttle switch send some pulses to the computer - which just adds an extra injector firings for each pulse (typically injectors are fired only when the trigger contacts in the base of the distributor pulse). There are two independent injector circuits (they fire in pairs) and if you look in the switch a single contact slides up between a sort of interlocking zipper sort pattern of contacts (herein referred to as Comb Contacts), so as the contact slides up it first hits one side, then the other. Of course you only want this to happen when the throttle is opening, so that portion of the switch is sort of friction loaded (herein referred to as Drag Switch), and when the throttle opens it rotates slightly on the shaft and closes a couple of contacts to complete the circuit, when closing those contacts are pulled open so you don't get the extra injector fires as the throttle closes. This 'electronic accelerator pump' action works even if the engine isn't running, turn the key on and if you listen very carefully you should hear the injectors clicking as you open the throttle, not when you close it. If it isn't working the engine will run just about as well as with it working, in fact you can totally unplug the throttle switch plug and barely tell the difference. I think the engine is a little slower to respond to quick throttle openings, but it's not something you'd particularly notice unless you were looking for it. Excellent explanation of how it functions normally...the question is what causes MaCs?
Good day everyone,
I hope someone can assist here. My 71 1800 has been having a problem. She has the original engine B20E. What is going on is, there is a HARD hesitation while driving at a steady speed/rpm. We have not been able to recreate in the shop. Seems it is only while under load (driving). It feels like a flaper or whatever in the air intake is closing off all airflow for a split second. The hole car then "bucks" hard. Has any one had this condition, and knows what to do? She runs good, and is at 248K miles. P.S. The condition goes away with a slight increase or decrease in rpms. Only happends at steady rpm under load. We have replaced the Throttle Control Switch with a new unit. I have preformed the adjust check. I count 15 clicks than a large gap (around half to 3/4 throttle) then almost full open, I get a few more clicks. This gap makes me think it would be in the area of 50-65 mph on the road, and that's the area where the bucking is hardest. Am I doing doing something wrong in the adjustment of the switch? Could it be ellswhere? This was slightly happening before I added items like the fireball elec ignition, and replaced the injectors. So I don't think that's it, but on these system, I just don't know. I was even thinking timing chain slop?
turn the car [Ignition] on, don't start. as you SLOWLY move the linkage, the injectors should click once every time a new spot on the TPS is found. there are supposed to be 20 separate spots, but 17 or 18 will do.[I wonder if he feels the same way about his paycheck? Although the upper region of swept Comb area is not used except under high load/WOT! If its not 20, there is something wrong!] if you don't get a click, move the engine slightly, the trigger points may not be open. good luck, chuck.
Link to FI troubleshooting manual: http://volvo1800pictures.com/document/fuel_injection_fault_tracer/fuel_injection_fault_tracing.pdf
From: http://www.swedishbricks.net/faq/fidjet.html [author's comments]
Date: Tue, 23 Mar 93 12:35:53 -0500
From: bw738@cleveland.Freenet.Edu (Ed Wetherell)
Subject: mysteries of d-jet
gonna love this one, about 90% of my problems were the throttle position switch! i swapped on one i had lying around and things changed drastically for the better. both units are of equal age and both are grooved on the acceleration path. both check out the same and both are spottlessly clean. i have no clue why some of my probs were effected because they occur over most of the throttle range and would even happen with the position switch disconnected!! [proving that it was not the TPS at fault!] as you probably know, [TPS is not necessary for base function, just for quicker acceleration and smoother return to idle]. anyway, i suggest you swap yours out if you have another handy.. looks like they might be bad and your average work w/ a DMM wont pick it up.. i am going to put the old one back on to be sure that the position switch was really the fault, but i am beginning to wonder how much witchcraft hans and gunter used when creating this monster!! or is uncle olaf pissed at me??
There's no witchcraft in play here!...its simply explained by the fact that a DMM only allows a static test, and so is not adequate for troubleshooting a transient condition as is most certainly happening at the TPS! Given the inputs to the Enrichment Circuit Flip-Flop that we have all (including the author, until recently!) been looking at, and are quick to point the finger of blame at, are the wrong area! After extensive further consideration (see above!), I now have corrected my position and believe the Comb Select Contact of the Drag Switch Selecting Contact with its slow closure and copper contact to be the source of the problem, although I have yet to explain precisely HOW this plays with the rest of the ECU to result in the MaC...but I'm working on it!
Also, when changing the TPS without a specific reason...simply "Troubleshooting by Substitution", replacing the TPS is NOT the only thing one is doing!! One is also moving and changing the adjustment of the TPS on the Throttle shaft as a side-effect, and this can have affects on the operation which one might interpret as improvement or having "fixed the problem"...for any meaningful test conclusions to be drawn, after replacing the TPS, it must again be adjusted correctly!
Excellent info on
setting Idle involving the TPS: Setting idle on a B20F D-Jetronic FI
(Crucial to Note: The Throttle Stop is not the Idle Adjustment!)
Sources of external materials are attributed. Otherwise, this page is Copyright © 2015-2017, Ronald Kwas. The terms Volvo, Bosch, Porsche (a family name, and a two syllable word!), VW, Mercedes, Renault, and Cadillac are used for reference only. I have no affiliation with any of these companies, or suppliers offering components and services, other than to try to keep Volvo's and Bosch's products working for me and reliably, and to help other owners do the same. The information and highly opinionated thoughts presented here are from my own experience, and very carefully considered opinion, and should be used in conjunction with normal, careful shop practice, or can be ridiculed and laughed at, or worshipped, at your discretion! Remember, you alone are in control of your knuckles, and future!
As always, if you can supply related additional objective info or experience, I’d appreciate hearing it, and will consider working it, along with the odd wise*** comment, into the next revision of this article. You are welcome to use the information presented here in good health, and for your own noncommercial purposes, but if you reprint or otherwise republish it, you must give credit to the author or link back to the SwEm site as the source. If you don’t, you’re just a lazy, scum sucking plagiarist...and the Boston Globe wants you!