Idle spd controller unit digifant, repair?

[denver_westy]

Hello my idle was intermitantly going to 3000 rpm, especially as I was approaching traffic at a stop light , idles speed controller was going haywire. I opened up the control unit behind the passenger tail light and here's what I found. Is this repairable or is the circuit board to far gone? The board is pretty burned. The wire traces don't look separated from the board.

I plan to order an NTE185 and NTE125 from mouser.com to see if I can fix it, I read these are what I need on another post. I'm not really an electronics tech so I dont really know what I'm doing, I think I need to use a solder iron and wick to remove the solder then pull out the transistor. I'll try to google for instructions on the electronical stuff...

Thanks in advance.

[stevey88]

The PCB has carbonized due to heat . They will conduct electricity. The best thing to do is after noting where the traces for the 3 legs of transistor go ( take photo ), scrape off all the carbon. You may end up with a big hole but it is OK as long as you can solder the pins of the transistor to where they suppose to go. You may have to solder them on the component side as the circuit side is no longer there. Note the position of the transistor, on one side you can see some metal around the mounting hole, the other side you can not. Put it in exactly the same way.
_________________
Steve

87 Westfalia full camper 4 speed

[stevey88]

Another thought. The transistor went bad because it is no longer pulsing and is permanently turn on. So the idle valve is open fully and cause the high idle speed. Because it is on all the time, the transistor heat up and burned out. Note that in normal operation, the transistor turn on only a brief moment and then turn off. This is called " duty cycle " So if the on time is longer, the idle valve will open more. Changing the transistor will not help in this case. Your hope is that the transistor simply shorted itself due to a spike on the supply voltage. This can happen if you unplug the idle valve while the engine is running. The idle valve is an inductive device ( it has coil ) so there are "back emf" if you just break the circuit.

Need to get my hand on one to trace out the circuit. It should be easy as it is a single sided PCB with not that many component on.

Jason, where are they ?
_________________
Steve

87 Westfalia full camper 4 speed

[morymob]

If u have one to go by this can be repaired,a good rainy day project. Get thenumbers off ALL transistors and the flat pack,also check the big black/silver band diode for being shorted.Problem with component shorts/burned,they tend to take out other components. Unless hot signs show the resistors are ok but replace ALL trans/diodes,and flat pack,theyare pretty cheap and there are subs which will have a diff # but function the same. Transistor types are NPN/PnP and are NOT interchangeable so keep tabs on which is which,also which way any diodes are installed. Solder runs ,replace with wire jumpers as needed. These seem to be pretty pricy now for some reason so good luck,and verify all your repairs when thru.

[JBange]

I happen to have a disassembled unit on my desk right now. Here's a picture of a (mostly) intact circuit board. Blue is the power transistor leads, red lines are the traces they connect to, and green spots are the points the traces go to. The power transistor has a metal side and a plastic side, the metal side faces the camera (indicated by the black arrow).

The OEM power transistor is a 4A 80V PNP model, part number 2N5195 (datasheet), but looking up the NTE185, it looks to be functionally identical.

Replacing the power transistor might solve the problem, but it's hard to say. That crazy analog computer design is highly inscrutable, so it may be being fried by something else upstream. It's a cheap part to replace, though, so you got nothing to lose by trying.

I'm actually working on an Atmel microcontroller based replacement. I've been saying "it would be easy to..." for four or five years now, but now I'm actually getting down to soldering parts to a perf board. One advantage to a digital controller is that it makes all the various idle points adjustable.

Hopefully it works as easily as the bench tests indicate and I'll be trolling here for beta testers in a couple months...
_________________
John Bange
'90 Vanagon non-GL
'90 Vanagon GL

[denver_westy]

Well I changed the transistor. Decided to skip the diode. Bought a new one for a buck at a store called electronica in denver. The guy said to old one was fine when he tested it with a meter but I changed it anyway. Also sprayed down the boards and connectors with electrical parts cleaner. Fast idle seems to be fixed! yahoo. Idle is steady at 900 at least for the last 50 miles around town over 3 days.

[jimbelmont]

I've searched and can't find an answer - what is the difference between the "B" and "D" version of the idle control module p/n 251 907 393?

[stevey88]

I have one of each opened up and will attempt to fix them when I have more time ( Working on a newly acquired Westfalia Fullcamper ). I think they perform the same function but the construction ( and the circuit ) is different. The B version ( early version ) has single sided circuit made of cheap PCB material, the later use double sided fiber glass circuit board. I would assume the D version is better as they changed the circuit ,assuming to improve it. Having said that, I think I will post what I think about the ECU with the D version.

One more thing, the B version has a resistor that is " Select on test ". That means after assemble of the PCB, it need to be put on a test fixture and a resistor of the correct value to make the unit operate in the right way need to be selected and soldered on a stand-off on the board. The D version get rid of this step so it could be a cost reduction of the B version.
_________________
Steve

87 Westfalia full camper 4 speed

[jimbelmont]

Forgive me for some of my questions, I do not have much electrical knowledge, I'm a mechanical kind of guy.

I've got an idle control valve that does not pulse, it snaps open when the ignition is turned on and stays that way, no pulsing. The Bentley tests I've performed on the ICV seem to suggest the issue is with the module so I pryed it open and took a look around. The wirewound resistor showed some heat discoloration and thats all. The power transistor that drives the valve has no visible thermal damage. Everything looks like it should be working, but obviously theres a problem.

I would think that the power transistor would provide the power to drive the valve, and the pulse duration would be provided by one of the ICs located somewhere on the board. Your comment here suggests this is not the case, that the transistor is pulsing its output:

[stevey88]

jimbelmont,

Let me explain a little bit of the output circuit of the unit and hope you can understand. The ctl unit supplies a pulsed signal to the idle valve. When it want to open the valve, it apply plus 12V. When it want to close the valve, it apply nothing. As the valve takes some time to fully open after it sees 12V, if you take away the 12V before it is fully opened, the valve will open a little and then try to close because of the internal spring. If you vary the time you apply the 12V and the time you do not and then apply 12V again, the valve will open a small amount depends on the length of time it sees 12V and the length of time it see nothing. This is called " duty cycle ". Also this is why it is vibrating. Keep in mind one side of the valve terminal is grounded through a resistor. So in order to increase idle rpm, you apply a longer 12V to the valve so the valve open more.

In your case, the valve is fully open so 12V is applied to the valve all the time. There are two situation this will happen:
1. The transistor is good but the drive to the transistor turn it on all the time. Changing the transistor will not help in this case.
2. The transistor is shorted out - collector to emitter.

To test if the transistor is shorted out, cut the circuit trace to the base of the transistor with a sharp X-acto knife. do not cut a too wide a gap as you will need to solder them back again. Use a DMM to measure to make sure the trace is indeed cut. Put it back to the socket and if you valve is still open, the transistor is shorted. If the valve is no longer open, you have another problem.
Find the base of the transistor here:

http://www.datasheetcatalog.org/datasheet/stmicroelectronics/4127.pdf

Hope this helps.
_________________
Steve

87 Westfalia full camper 4 speed

[mecheng87]

hello stevey88, morymob and Jbange

Wondering if one of you has scienced out these modules such that you know or have measured the critical signal levels using a scope or DVM at various test points on the PCB when the module is plugged into a running engine at a specific RPM? Or is it simpler to just replace the various transistors and diodes.

Cabot

[stephenvw]

My Vanagon 87 ISC might be dead? I have more than one question puzzling me.

(1) I was to adjust the screw for idle speed (it is 900, OK, but I just try to get it right). So I warm up the car, then disconnect the coolant temp sensor. rpm goes up 1500 and stay there. Obviously the ECU thinks it is cold and put more fuel, isn't the ISC will restrict the air so that MAF will tell the ECU to put less fuel??? It does not.

(2) While doing that, I unplugged the ISC valve. The engine died right away. Then I reconnect the ISC valve in. But it can not re-start the car any more. I recall reading a thread about VW Golf, that if we unplug the ISCValve while the engine running, it will damage the ISC. Is that true?

(3) But when I put the Coolant Temp sensor back, it starts and work as normal. That means the car has something changed since I messed with it. It used to be able to start the car with the coolant Temp sensor unplugged. Now it does not. Why?

Stephen