Parts Catalog Accessories Catalog How To Articles Tech Forums
Call Pelican Parts at 888-280-7799
Shopping Cart Cart | Project List | Order Status | Help




Go Back   PeachParts Mercedes ShopForum > Technical Information and Support > Tech Help

Reply
 
LinkBack Thread Tools Display Modes
  #1  
Old 09-16-2002, 09:42 PM
rcmktg
Guest
 
Posts: n/a
Of potentiometers and men

1985 190E 2.3 with CIS (Cursed Injection System). a few questions after an entertaining weekend:
1. Potentiometer reading between 14 & 18 pins on the potentiometer 4890 ohms non-adjustable, and about 50ohms over spec: enough to replace?
2. Readings between 14 & 17 pins: 750 at rest, 5670 at full deflection, as high as 5690 dear full deflection. Enough to replace?
3. Adjustment of potentiometer, reputedly very sensitive, yields only about 150ohm at rest, and less than 20 at full deflection is this right?
4. Checking the EHA, the final adjustment of control pressure, and therefore the last modifieir of fuel output by the injectors, I can modify its values correctly when testing for full lean and full rich conditions at the oxygen sensor, and the system "floats" when everything is connected, indicating the eha is correctly adjusting for downstream readings and engine operating conditions. But at wide open throttle, instead of going to 7-9mA, it fades off to about 2, effectively leaning the engine out. I dislike jumping to the conclusion that the computer is at fault, but the throttle switch works, the idle cutoff is working, thermal sensor and barometric sensor all seem ok. Fuel flow and pressure are at least 150% of minimums. I did find a leaky coil wire, and two if the injectors had lousy patterns so they were replaced, along with the injector seals, all of which gave me a better idle and part throttle, up to about 3200rpm under light loads, but the performance end of the throttle just isn't there. Oh, compression is 145 - 160psi and plugs indicate sooty misfire.

Man, I hate being ignorant, but this is just at the periphery of my bosch capabilities. ANY suggestions? ( other than offers to buy it for parts, it's still an ok vehicle at 151k)
Reply With Quote
  #2  
Old 09-17-2002, 08:43 AM
it leaks, its german
 
Join Date: Mar 2002
Location: raleigh nc
Posts: 1,111
is the fuel pump getting full system voltage? this almost sounds like a weak fuel pump or a weak coil. If thw pump is howling I'd be running volume tests on the pump.


Joe
__________________
Project Smoker, '87 603 powered wagon
Hauler, 96 CTD can you say torque?
Toy 73 Cougar xr7 convertible
Acme Automotive Inc.
Raleigh NC
919-881-0364
Reply With Quote
  #3  
Old 09-17-2002, 09:02 AM
Moderator
 
Join Date: Mar 1999
Location: Tucker, Ga USA
Posts: 12,153
At WOT almost all of the sensor imputs are ignored & the contact for WOT takes over. The computer will richen the mixture for max output at that time.

Many times the EHA doesn't change the differential pressure correctly to richen the mixture.
__________________
MERCEDES Benz Master Guild Technician (6 TIMES)
ASE Master Technician
Mercedes Benz Star Technician (2 times)
44 years foreign automotive repair
27 Years M.B. Shop foreman (dealer)
MB technical information Specialist (15 years)
190E 2.3 16V ITS SCCA race car (sold)
1986 190E 2.3 16V 2.5 (sold)
Retired Moderator
Reply With Quote
  #4  
Old 09-17-2002, 09:32 AM
Moderator
 
Join Date: Mar 2000
Location: Gainesville FL
Posts: 6,844
The action of the airflow potentiometer is monitored by the KE controller. The controller is looking for rapidity of change (or acceleration of position so to speak). When one makes a sudden airflow change (flooring the accelerator), the controller gives an instantaneous mixture increase of 5-10 ma (probably a 0.1bar change in differential pressure). This is like an accelerator pump in a carb and does not stay. The increase in current is momentary and drops to nominal in a second or two.
__________________
Steve Brotherton
Continental Imports
Gainesville FL
Bosch Master, ASE Master, L1
33 years MB technician
Reply With Quote
  #5  
Old 09-17-2002, 12:02 PM
rcmktg
Guest
 
Posts: n/a
To answer questions /comments as they appeared:

1. The WOT reading is supposed to be 7 - 9mA I see 2Ma WITH THE ENGINE RUNNING. Isn't 2mA the lean direction from the 8mA nominal? ( I assume that from seeing the current rise with counterclockwise adjustment of the mixture adjustment screw. ) Is this lack of correct response a sensor (input) problem, or controller ( output) issue? If I don't see any increase in current for fairly large movements of the throttle, is the computer ignoring the potentimeter, or is this likely a potentiometer problem? Is it a proportional response: that is, small movements yield small increases, and larger movements yield larger responses?

2. As noted in original inquiry, the pump has at least 150% of service minimum volumes. System pressures stay in range from idle through any throttle positions, and control SEEMS to float as adjusted by EHA.

3. I did the resistance tests of the coil, but do not have tools for max output. I'll check with a local shop, since my ancient SunScope is not doing any oscillating these days.

4. Is there anything like a logic diagram for the CIS controller? I have verified. both at the switch and at the controller connector that the WOT switch is working, so what else is in the loop at WOT: temp sensor, potentiometer, baro-sensor? Of these, only the potentiometer is showing "above limits" values. Should it be replaced?
Reply With Quote
  #6  
Old 09-17-2002, 12:30 PM
Registered User
 
Join Date: Jul 2001
Location: SoCal
Posts: 1,303
'Potentiometer reading between 14 & 18 pins on the potentiometer 4890 ohms non-adjustable, and about 50ohms over spec: enough to replace?'

This is about 1% of spec - I'd say damn good.

'Readings between 14 & 17 pins: 750 at rest, 5670 at full deflection, as high as 5690 dear full deflection. Enough to replace?'

I don't know what they should be, but typically a pot's resistance may decrease at the full stop compared to at the end of the resistance element. Not likely a problem. Pot failure mode is usually an intermittant internal connection, a worn wiper (erratic reading at spots), or an open terminal. It is rare to get a significant change in resistance due to failure.

'If I don't see any increase in current for fairly large movements of the throttle, is the computer
ignoring the potentimeter, or is this likely a potentiometer problem?'

Again, I don't know what is happening. But, to maintain precision, it is typical for a monitoring circuit to use a current source to deliver a constant, precise current to the pot - then monitor the voltage change as the resistance varies.

Hope this helps, since I am a neophyte with CIS, but involved with electronic control systems for decades.

Steve
Reply With Quote
  #7  
Old 09-17-2002, 02:57 PM
rcmktg
Guest
 
Posts: n/a
Perhaps I should have been clearer: Potentiometer 14-18 resistance 3200 ( 3240 some listings) to 4800 ( 4840 some listings) Is 1% over range limit unacceptable?

14-17 580 to 840 at rest, 3800 to 5600 full deflection (620 - 870 and 5620 some listings) Same question?

IF I remember correctly, increasing resistance reduces current, so if the circuit is a counterbalance to the steady state load being sent to the EHA, less current here means more to the EHA, providing an increase in fuel delivery because of the change in control pressure. THis is evidently a time based adjustment, after which the O2 sensor again resumes priority, unless we are at WOT, at which point the WOT switch is supposed to put the EHA in a "nuetral" amperage of 8mA. Since I am not seeing 8mA in the WOT configuration, something is causing the system to go lean ( I am still assuming that 2mA is a lean condition setting for the EHA.

Now, The CD service manual says to test the pot at the controller harness, which is actually a partially shared junction with the barometric pressure sensor ( called an altitude sensor, but really senses changes in atmospheric pressure). I get a different reading at the connector than I do at the pot itself, I suspect due to this sharing of a lead with another variable resistor. At this point, confusion takes over, for the test of the baro sensor is an amperage one instead of a resistance check. I manually changed the pressure and saw changes in resistance through the baro sensor, with no dead spots

My guess is that this systemuses these variable resistance elements, such as the O2 sensor, coolant temp, air plate deflection potentiometer, and barometric pressure sensor as a series that starts with a known voltage and amperage, passes it through the variable resistors to achieve either a counterbalance to the nominal output current, or it becomes the output current. In my situation, the output current seems ok at idle and up to 1/3 or so throttle, but after that it goes away.

If the problem was electrical, as in the ignition system, the only two things I can identify as candidates are secondary voltage leaks and a weakness in the coil leading to an extended field recovery time or a weak maximum voltage.
Reply With Quote
  #8  
Old 09-18-2002, 12:30 PM
Registered User
 
Join Date: Jul 2001
Location: SoCal
Posts: 1,303
I reiterate that I am simply approaching your posts by starting at electronic basics. I find them confusing because you are not using termnology consistent with what I am familiar. So here goes:

'Perhaps I should have been clearer: Potentiometer 14-18 resistance 3200 ( 3240 some listings) to 4800 ( 4840 some listings) Is 1% over range limit unacceptable?'

Problem #1 - I don't know what you are referring to by 'listings'. Are you getting your information from various sources? 1% in a pot is excellent.

'I get a different reading at the connector than I do at the pot itself, I suspect due to this sharing of a lead with another variable resistor. At this point, confusion takes over, for the test of the baro sensor is an amperage one instead of a resistance check. I manually changed the pressure and saw changes in resistance through the baro sensor, with no dead spots'

If your primary concern is proper functionality of a potentiometer, it should be measured disconnected from the rest of the circuit, and certainly with no external voltage or current source applied. Else, any resistance readings will be meaningless.

Once you have determined that the resistance readings are correct at the extremities, and that there is no sign of intermittant operation at any point in the normal range of its travel, the pot has been tested adequately.

'My guess is that this system uses these variable resistance elements, such as the O2 sensor, coolant temp, air plate deflection potentiometer, and barometric pressure sensor as a series that starts with a known voltage and amperage, passes it through the variable resistors to achieve either a counterbalance to the nominal output current, or it becomes the output current.'

The computer can sample an input one of two ways. It can supply a known current OR a known voltage. It cannot simultaneously do both. Ohm's Law A=V/R shows this. If the ECU supplies a known reference voltage, it will sample the current flow through the circuit by monitoring the resulting voltage drop across an internal calibrated resistance. If is sources a calibrated current, it simply monitors the voltage across the sensor resistances. Knowing which mode of operation the ECU is in will make more sense of your measurements. If the manual says it is sourcing a current, you need to measure voltage drops. If it asks you to measure currents, it is sourcing a voltage reference.

Steve
Reply With Quote
  #9  
Old 09-18-2002, 03:48 PM
rcmktg
Guest
 
Posts: n/a
The "listings" I referred to were three different service manuals: an aftermarket book on Bosch CIS, a printout from Haynes that they sent when their original publication proved to be lacking, and the CD rom copy of a Mercedes service manual.

I have a somewhat functional knowledge of automotive electronics, and even german-style electronics, where pins labeled 30 are always on, 15 is on with ignition, etc., having been at one time certified by VW in automotive electronics. THe problem with a rational application of Ohm's law is the requirement for knowing two of the three variables, and Mercedes seems to only provide one, leaving the technician to either perform blind testing or guess at the other variable.

I agree that testing the devices individually is best, but having one method in one book and another somewhere else is aggravating. If I assume, based on provided information, that the 8 volt value uesd internally tot he computer is also used for power to sensors, well then we could establish a current amperage through any known resistance, right? And knowing the desired amperage at the EHA unit, we could establish what the resistance of the EHA should be, and possibly establish how much energy is required to move the EHA coil against a given fluid pressure, right? THen, if we knew the logic diagram for the system, we could see how the various sensors (inputs) and motors or coils ( outputs) interact, making the diagnostic process one step better than "push these buttons on the tester, get these values or replace the parts."

If I sound frustrated, it's because vehicle manufacturers seem reluctant to give diagnostic authority or education to their front line, the service technicians. This seems especially true of the information provided to non-dealer techs, though I remember a lot of "do this" and "change that" when I was a dealer tech on the late 70's and early 80's.

I will look again for drop-outs or dead zones in the potentiometer. I rashly thought that if it worked at the ends there needed to be continuity. I still wonder about how being above the specification limits affects performance. How many milliamps to the ohm at 8 volts?

If anyone is still reading at this point, I used the 8mA target for a standard temperature engine as my method of adjusting the idle mixture after changing injectors. At that setting, Cold start is less than desirable, maybe 5 seconds cranking at 60F air temp, and stumble for about 30 seconds after starting. I still have about the same 1/3 throttle limitation for real power, but the engine will not die under wide open throttle. I even saw 70mph on a flat stretch of highway, where before I was limited to about 50. Having fun now
Reply With Quote
  #10  
Old 09-18-2002, 04:46 PM
Registered User
 
Join Date: Jul 2001
Location: SoCal
Posts: 1,303
'The "listings" I referred to were three different service manuals: an aftermarket book on Bosch CIS'

If this is the SAE publication by Bosch, it is by far the best system description for determining qualitative function of the system. I would rely on the Mercedes CD to pull up the numbers and locate the parts, but it is otherwise probably less instructive. The key, though, is regardless of method, all procedures should converge to the same result. E.g., if I found the Mercedes procedure to produce baffling results, I would use the Bosch method - that might show up the problem.

'How many milliamps to the ohm at 8 volts?'

Current is inversely proportional to resistance. 8V across 8 Ohms will produce 1000 mA, and across 80 Ohms you'll get 100 mA; across 1000 Ohms, you'll get 8 mA.

Note that 8V - measured from ground - on one side of the circuit must be compared to the voltage on the other side. That might not be at ground potential. If the ECU uses a current source in its return line, you might measure a fixed current even though the sensor resistance is varying and the high side voltage remains at 8V. This is because the current source allows the voltage on the low side to vary as necessary and this is then the voltage monitored by the ECU to determine sensor condition.

Measuring current is also not trivial. Most meter circuits use a low resistance internally to create a small voltage drop it can measure, due to the current flowing through it when hooked in series with a circuit. This voltage drop must be considered in relation to the circuit you are measuring. If you are measuring a circuit which is sourced by a very small voltage, the meter's voltage drop can affect the operation of the circuit. For this reason, and to avoid having to break a connection in the circuit you are measuring, it is often handier to measure the static resistance of some device already in circuit, and measure the voltage drop across that element. The current can then be calculated from Ohm's Law.

Good luck! I know these systems can try one's patience.

Steve
Reply With Quote
  #11  
Old 09-18-2002, 05:14 PM
Moderator
 
Join Date: Mar 2000
Location: Gainesville FL
Posts: 6,844
I hate to break this up, but component testing is a poor way to do diagnostics. Diagnostics should be through performance testing, not the testing of design criteria. Somewhere a bjillion words back was a condition existing on a car.

If you wish to evaluate the potentiometer, measuring airflow rate, then the proper way is through results. Two forms: raw data which is not resistance but voltage. The control unit is looking for voltage. But it is more interested in rate of change of voltage. This spec isn't available but the result is. The result is an appropriate snap increase in EHA current. My best guess is 5-10ma from nominal.

If you wish to fix a car you need a hypothetic system answer to explain basic testing: What is the base complaint? Does an EHA current problem explain the condition? Is it a general lack of fuel? Is it fuel? All fuel control starts with the hydraulics. It is only modified by the control electronics. The final answer is to see the EHA current as expressed by differential pressure. Lots of test values.
__________________
Steve Brotherton
Continental Imports
Gainesville FL
Bosch Master, ASE Master, L1
33 years MB technician
Reply With Quote
  #12  
Old 09-18-2002, 08:06 PM
rcmktg
Guest
 
Posts: n/a
OK Steve, with my Radio Schlock $49.95 digital multimeter hooked up in as an ammeter in the EHA posiitve voltage line, idling in normal mode, I have 7.5- 8.2mA general values Bashing over to WOT, the reading falls off to 2mA, when everything I read says it should default to 7-9mA. That's why I'm trying to work back through the system one stinking component at a time. I'm going out right now to see if I can get the "accelerator pump" effect, and watch the readings as I cycle through available rpm and throttle settings.

If mA readings fall off as throttle opening increases, is the engine going too rich or is it incorrectly interpreting some sensor?

Last edited by rcmktg; 09-18-2002 at 08:42 PM.
Reply With Quote
  #13  
Old 09-18-2002, 08:39 PM
rcmktg
Guest
 
Posts: n/a
Just started up and watchedammeter (yes, that can be taken multiple ways when left as one word) Outside air about 70F, car sat in sun all day, so probably a little warmer than that. Startup current: 20mA, falling to 13 within 1 minute, and then on towards 9. Throttle blipping DURING WARMUP yielded 10 or more mA gain for a couple seconds, max. Slow rate opening of throttle yielded ever decreasing current, down to less than 6mA at approx 3500rpm, and 2mA at WOT. Something is giving a wrong signal to computer? Computer is ignoring suggestions and doing things as it chooses? Engine ( remember: 152k mi., 145-160psi compr, some blowby, new secondary wires, plugs, cap) blowing raw gas out exhaust?

After warmup, ABSOLUTELY NO "accelerator pump" effect. Did I read that potentiometer is really only active during warmup phase ( Bentley publications: "Bosch Fuel Injection Management" (published 1989, 1991), or is it supposed to be in the loop all the time?

Given these tests and results, it apears the WOT switch is not making its over ride of the rest of the system. Is there something else that could prevent that from happening?

Coil tested to 30k volts min. output, timing about 10deg BTDC ( Snap-on induction clip variable timing light, un- calibrated )

Idle system pressure 80-81 psi ( KD aftermarket gauge, uncalibrated)
Control pressure 75psi, variable with EHA by 1 - 3 psi
Other items mentioned earlier posts
Reply With Quote
  #14  
Old 09-18-2002, 08:50 PM
rcmktg
Guest
 
Posts: n/a
One other thing keeps nagging at me, since I don't have a bench setup or special lines to allow final output testing of injectors:

Could I have crud in the fuel distributor, say in those little screens? An old 80 volvo we had got some really bad fuel and contaminated the entire system. Rusty micro silt in every screen, nook and cranny of the system, including the injectors and distributor.

If I rigged up a set of special lines so I could measure injector output, does there happen to be a range published for that type of test? Seems to me it was 5cc/min at idle and 25-30? at WOT for a Rabbit engine on a K-series CIS
Reply With Quote
  #15  
Old 09-19-2002, 09:32 AM
Moderator
 
Join Date: Mar 2000
Location: Gainesville FL
Posts: 6,844
If I get a chance I'll look for the spec on accel enrichment. The system you are working with is MBs first KE and they quickly went to a plus/minus current control.

It sounds to me that you are just responding to lambda control with the 2ma situation. You might try disconnecting the oxygen sensor and see if you get any accel correction. With O2 sensor disconnected you should stay at 8ma except for any correction that is functioning (starting, warm-up, accel).

What is the customer complaint? How is performance?
__________________
Steve Brotherton
Continental Imports
Gainesville FL
Bosch Master, ASE Master, L1
33 years MB technician
Reply With Quote
Reply

Bookmarks

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On
Trackbacks are On
Pingbacks are On
Refbacks are On




All times are GMT -4. The time now is 01:10 PM.


Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2017, vBulletin Solutions, Inc.
Search Engine Optimization by vBSEO 3.6.0
Copyright 2011 Pelican Parts - Posts may be archived for display on the Peach Parts or Pelican Parts Website -    DMCA Registered Agent Contact Page