TheSamba.com :: View topic - Want to extend the life of my AFM: Bend wiper?

Tue Oct 07, 2008 10:06 pm

Weird question, I know.

I want to avoid paying $160.00 (plus core) for a new AFM right now. Already dropped thousands into our rig over the last year and, well, would better get away with spending that money 6 months from now - you know, the spouse does the check book....

My rig stumbles/hesitates when under load and giving gas above 3,000 RPM. Bentley states that CA emission models have a speed limiter that engages when the TPS is tripped (can't find that limiter for the life of me), and stumbling could mean an ECU/wiring issue. Wiring was verified by me (through rebuilding). That said.....

I've done the resistance checks on the AFM and there is a real dead spot around cruising speed - just over 3,000 RPM. I want to exclude this as an issue before going out to buy an ECU and/or fuel regulator (another possible culprit).

Someone here (I forget who! I'm sorry.) suggested that one can bend the wiper on the AFM where it comes in contact with the carbon to a "good" area. Looking at it, I can see where the two contacts have created their own "trenches" and where, if I bend the wiper in the right way, contact can be made on "good" carbon. Problem is, the wiper has 4 places to bend, as the wiper itself appears to be cut out of a flat piece of metal and, although a single piece, has rectangular holes between each of the 4 metal strips.

Long story short.....

Anybody try this? Any suggestions on how to bend all 4 at the same time? Advice on how to measure where to bend?

Thanks in advance for providing advice on this weird question.

Wed Oct 08, 2008 6:01 am

is it a definate groove and have you tried cleaning it.....why not just source a used one that doesnt have the dead spot and then tinker with yours if your still interested in experimenting with bending it (which i might attempt by taking two bodkins or upholstery needles....one on top of the wiper and one below to force a crimp that draws the contacts ever so slightly inward)....

Wed Oct 08, 2008 6:21 am

Don't kink the wiper or it will snap. That's really the only fear here, it's an easy fix.

Wed Oct 08, 2008 9:29 am

Thanks for the replies.

I started cleaning it and then stopped - the carbon matrix comes off! I believe this is normal and, therefore, should not be cleaned or touched up in any way.

Rather than get a used one (with the same issues, likely), I'd rather get a new one. Just want to try to avoid it for now if I can.

Upholstery pins, eh? I'll give that a crack. Only need to bend it a wee bit, and want to make it right the first time. Thanks for the suggestion!

Wed Oct 08, 2008 3:59 pm

I had a similar dead spot I was able to get out of, about 5 years back with a 1.9
Don't know if this will be the same, but..
If I'm remembering right there were 4 screws that held that plate in.
I unscrewed those and used a dremel to *very* slightly lengthen the hole so that the plate could be positioned on new turf.
WARNING will robinson WARNING
I have no idea if this is a sick thing to do, or smart. At the time I was about to go subaru and so I didn't really care too much if I upset the apple cart.

btw, before doing this I had already tried loosening the screws just a little, applying pressure with drivers in the gap to force the plate back a bit, then retightening, and felt that made a difference. That's when I decided to expand on the idea a little.

Wed Oct 08, 2008 4:16 pm

Quote: I unscrewed those and used a dremel to *very* slightly lengthen the hole so that the plate could be positioned on new turf

30 years in electronics tells me it will work. I have not try this but I will If I need to. But I will test the resistance curve after the mod.

Wed Oct 08, 2008 7:00 pm

That would work fine. The voltage curve wouldn't be changed in any measurable way, and these things just aren't that sensitive anyway. Bending the wiper has long been done, but I think repositioning the board is a better idea.

Wed Oct 08, 2008 8:02 pm

I just loosened the four screws on mine and moved the board closer to the shaft. The holes were large enough to allow a bit of travel; it doesn't take much.
The rare Vanagon zero-dollar repair.

Wed Oct 08, 2008 8:49 pm

Super duper! Didn't attempt the pin bending trick yet, and I'm going to try repositioning the board tomorrow, instead.

Awesome, folks. I hope that the thread will help others in our shoes in the future via the search function :-)

Thu Oct 09, 2008 10:48 am

sheesh...just be wary of moving the board right or left.....it doesnt take much movement at all to throw the afr off......also not to sound like someones mother, remember to mark its current position so you can return to it should you screw up somehow....g'luck

Thu Oct 09, 2008 9:42 pm

Well, I was able to move the board on which the carbon resides and on which the wiper swings. The board on my AFM only has three screws (two left / one right). I loosened them and pushed the board forward instead of backward, as there seemed to be more carbon to play with. While I haven't yet driven it, the resistance values definitely changed for the better - where there was a dead spot there is now resistance within tolerance.

Thanks to everyone, especially for the great advice on moving the board ever so slightly. Seems to have worked like a charm.

Thu Oct 09, 2008 10:04 pm

Hey, guys, to test the pot in these things, there's a much easier test that actually shows the net function of all them resistor-patchy-thingys on that there board there (too much sarah palin, sorry). It is after all just a regular pot.

Put your VOM in a range to measure 5V, and probe between pins 2 and 4, either with the top off (easiest) or thru the back of the plug (pull boot back). Do this with ignition on, or with it running. You'll see scaled 0-5V according to the flap position, toward 5 being fully open (it'll probably max out in mid-4's). At idle it should be around 1V, mas o menos. Either work the flap by hand or rev the motor and you can watch the voltage scale up. It should scale smoothly. You'll be able to catch any drop-outs easily.

This is the actual signal the ECU sees for load. The IAT sensor is a separate 5V circuit (betw. 1 and 4).