Hey folks, wondering if some of the seasoned pros could help out with a super weird issue I am having.

So its regarding my 90 420sel.
A month ago I finally finished up the last stage in my a/c rebuild on the car by putting in a new condenser. And I filled up the system with 134a.. yea I know.. .. its all I had on hand to fill the system.

So far the system is holding a charge, and the cooling performance is as I expected with 134a.. semi ok.. I have r12 on the way. All my other vehicles use 134a so I have a few 20lb tanks of it.

My problem is definitely one of the weird ones I've seen. So I finally got around to trying the a/c and do some test drives around town, and im encountering a odd behavior. Ill kick in the ac, and the eco gauge will nearly peg to the right, and I literally cant get the car to accelerate above 60-70km/h.. It just wont go!!. Weirdest thing ive seen. Normally one would think the ac pressure is going way too high, but that's not the case. Ive tested the high pressure cutoff and its working as normal, and at idle, pressures stay within reasonable limits. Heres the kicker, I turn off the ac, and it takes about 15 seconds for the car to regain power, even though the compressor kicks off right away. So instantly im thinking its a fueling issue. But no where I can see if the CIS gets any type of signal from the hvac controller to tell it that the ac is enabled. I would think it would, to give the car a slightly higher idle when the ac is enabled. but cant seem to find any type of reference in the diagrams.

So heres part 1.5.. last weeked I decided it was time to do some R&R on the CIS sensors. I decided it was time to check the k-jet flow plate sensor readings as well as make sure I was getting a full throttle signal to ensure I was getting full throttle enrichment. Ive seem to be having odd performance issues with this car, even though the fuel system is in decent shape, and ive done a fair bit of work on it. So long story short, slightly adjusted throttle linkages to ensure full throttle, and installed a new plate sensor & calibrated. Overall car seems to run a bit better without the ac on.

Ok part 2 now..
Today I went to make sure my aux fans were working as expected, agian.. Just to rule that out.. And they were. But as I had the car idling in the driveway with the ac on. I would get random burst of rpm, and it would then settle down. I looked at the eco gauge, and over a 15-20sec period, the gauge would slowly move from the left to the center, then the engine would quickly rev up, settle down, and the needle would be pegged back where it belongs at the left.. then after a few sec, start to move to the center again (during this whole time the ac compressor continued to run! it did not shut off). Instantly I thought I should check the duty cycle. I pull out my multimeter. and sure enough.. it would start off at around 50%, where Ive had it set at all this time, then slowly creep down to 0%, then kickup, rev and go back to 50%.. Ah ha!! something is up here.
I turn off the AC, 15seconds later everything stabilizes, and the duty stays where it usually is around 52-47%

Only thing I could think of causing this... would possibly be the wires going to the EHA being backwards?? Or the CIS controller is on its last leg. Its like its trying to adjust the mixture, but in the end its doing the opposite of what it wants to do.
Is the EHA polarity sensitive? this car never really had the performance I would expect, my 560 would smoke the 420 on the street. I know its more powerful, but the difference is noticeable. Don't get me wrong the car doesn't run bad at all. It drives and idles pretty normal. But something has always been up in the fueling department.. like that last 15% is missing.

This one has me stumped

__________________
Just my MB's:
1990 420sel - Winter DD - 410,000km
1990 560sel - absolutely showroom mint - 360,000km!

I did another quick test. I started the ac, and unplugged the EHA.. Car idles like normal, duty cycle drops to 0%. but eco gauge stays at left where it belongs, idle does not jump every 20sec like it did with it plugged in and A/C on... didn't get a chance to test drive with unplugged eha but will soon.

__________________
Just my MB's:
1990 420sel - Winter DD - 410,000km
1990 560sel - absolutely showroom mint - 360,000km!

Just shooting from the hip, here is it possible you are introducing a major vacuum leak when you turn the A/C on?

__________________
Cruise Control not working? Send me PM or email (jamesdean59@gmail.com). I might be able to help out.
Check here for compatibility, diagnostics, and availability!

(4/11/2020: Hi Everyone! I am still taking orders and replying to emails/PMs/etc, I appreciate your patience in these crazy times. Stay safe and healthy!)


82 300SD 145k
89 420SEL 210k
89 560SEL 118k
90 300SE 262k RIP 5/25/2010
90 560SEL 154k
91 300D 2.5 Turbo. 241k
93 190E 3.0 235k
93 300E 195k

The EHA, via the ECU, is responding to the exhaust gas oxygen content as detected by the O2 sensor. Some O2 sensors, when in their dotage, have been known to deliver signals that are reversed in polarity from the normal. Unplugging the EHA has removed the O2 feedback of fuel/air mixture from the CIS. It is, in effect, operating open-loop.

Also, there is a connection between the compressor control unit, and the idle speed control unit, as you would expect.

Re: thread title
"delima", or, deli ma: the matriarch of the delicatessen

Oh, I understand that fully, but for some reason, only when the A/C is on is the CIS is doing the opposite of what it should be doing. Ive also replaced the O2 sensor a year ago, and flew thru emissions with no problems this year. I really want to reverse the polarity on the EHA control, but I suspect that wont do too much just for the plain fact that it is virtually just a coil inside. And relays for example activate with either polarity.

Today I did a little test. On my way home, I turned on the ac; and power totally disappeared. I pulled over, unplugged the EHA, and took off again.. Power was almost normal, car didn't have issues with accelerating. SO something is going on.. Ive read thru a few old threads about others 420sel having power issues, and there really wasn't much of an outcome. One person replaced the CIS computer, and the power came back. But none of these were related to A/C use only.

__________________
Just my MB's:
1990 420sel - Winter DD - 410,000km
1990 560sel - absolutely showroom mint - 360,000km!

DB:

In unplugging the EHA you have disabled the mixture feedback circuit, as noted above. Absent feedback, the mixture is controlled entirely mechanically by the airflow meter/fuel distributor package. That engine power returns to a near normal state suggests that there is an errant wiring connection between wires associated with the A/C compressor control unit (N6) and wires between either the ECU and the EHA, or between the O2 sensor and the ECU. Pulling the plug from the compressor, and repeating your driving test may tell you if the symptom is current related.

I will give that a try!
I could see that there was more then one wire going to the coil on the compressor, so im guessing that is the spot the cis gets its compressor activation feedback signal for the idle compensation.

__________________
Just my MB's:
1990 420sel - Winter DD - 410,000km
1990 560sel - absolutely showroom mint - 360,000km!