1993 300E 2.8 Saga
 

Here's what's been going on with this car in case anyone is following.
Bought the car running badly at idle, the usual idle surging but otherwise running great at loads. I knew nothing about the HFM-SFI system at the time. I mistakenly thought this would be an easy repair..Boy, was I wrong.

Step 1:) I check the plugs. It had some incorrect autolites. I replace with FR8DC+ copper core. Would have liked to have non resistor but they were out of stock.
Step 2:)I also checked the wiring harness and it was in bad shape. I replaced with an updated one. NO change..car runs the same.
Step 3:) I check the regen/purge valve..it's pintel is moving but not shutting off vacuum. i bench test and it passes. Turns out it was installed backwards!!I installed it correctly but still no change.
Step 4:) Pull codes again, release codes and 49 at pin 8 keeps coming back. I had already checked the OVP to be the latest version, so I pull it out and bench test..fails to close the contacts on 4-6 relay. I take it apart and there is a tiny particle holding the small relay open. I fixed it and passes all test. reinstall, checks codes all clear now. WOW! Big change in the running of the car, idles now like it should, I think great.. all fixed..till i take it for a drive. falls on it's face on acceleration. pull back in the shop. What I think I've learned on the hfm for all those that have the surging idle problem, I think that just means it's in limp mode?
Step 5:)Now, I'm getting really frustrated..pull codes again, get code 4 on pin 8. I clear it, and it comes back.I pull the HFM/maf and bench test. Seems to perform. Starts with around .9Volts with no air flow. As blow air across it, the voltage rises smoothly up past 2.5Volts. I repeat this a few times. I never cleaned it. I reinstall it and now the code hasn't come back, but car still runs the same. starting to think, i'm chasing my tail now..lol
Step 6:)Decide to check the O2 sensor..seems to respond ok. I check the heater voltage...none! but, bad news it it gets it directly from the hfm ecu...no fuse or relay to check. i give up for the day..now today I get the car up to operating temp, recheck the heater voltage and still zero. I plug th harness back in, and with a sharp probe recheck and haha, it's there..over 12 volts.That's odd, so I unplug and it's still there.
That's where I'm at so-far with this step child of mercedes(300e 2.8HFM)..I'm still wondering about the maf, and also the throttle servo, since this is drive by wire, and now that it's out of limp mode it's in play.

If the throttle is responding, it's not in limp home. Limp home means the mechanical rod is all that's opening the throttle plate. You'd know if you were limphome because you'd be pressing on the pedal like king kong and getting 15mph out of it.

OVP?

Try replacing it....instead of fixing it

Any codes?

Have you checked codes for the throttle assembly? Pin 14, I think (search is your friend here).

I have the same car, and it had the same symptoms you originally posted. Turned out to be a dead MAF and aged O2 sensor. I had pin 8 code 4, and another - forget the exact code - that implicated the o2 sensor. A fresh MAF fixed code 4. Driveability was perfect, idle was improved but still a bit lumpy. A fresh O2 sensor delivered a perfect, seamless idle.

What does "falls on it's face" mean - does it not respond to the throttle at all? Or is it slow, or what?

Update: Thanks for all your suggestions. I Apologize for the delay, It's been hot here so I haven't been in the shop much and this is a spare car, but I rewired the throttle servo and that took care of the hesitation in the car. It passed emissions now too on first try. Here are the results: HC-0.38 out of 1.00,CO-2.29 out of 12.00,NOX-0.57 out of 2.50. The car runs pretty good now too, just one problem left..the idle speed sometimes drops low enough at a stop sign to kill the engine. I haven't had time to look into that.