digital egr delete egr bypass

[yellow3800]

Anyone know how to bypass the digital egr valve?

It looks like cold-idle is not possible when the valve is removed inferring that the valve opens to allow air into the intake manifold to keep it running. no valve=not enough air and the engine quits.

If someone has done this before it would save me some time.

[DTMAce]

Not sure I understand that.

The EGR valve allows exhaust gases through to sample them, not external air. A few people have simply deleted the EGR by removing it and mounting a cover plate, or even removing the mounting from the intake completely. Then they would either live with the SES code being thrown for it, or have that removed in the computer electronically via programming.

But I have not heard of anyone having the issue of removing it and the engine not running. That's a first. I can't see why that would cause your issue in that way. But I'm not the smartest engine guy here either. LOL

[3X00-Modified]

Ace is correct, the EGR should not be functioning when it's cold idle. If your car wont start and run cold without the EGR in place you have another issue somewhere else.

[yellow3800]

i could live with a temporary engine code if it didn't knock out all the 'quad drivers' the thing has. consequences are : no TCC lockup [otherwise it does on gears 3&4], suspected cold idle problem for first minute after cold start, slight midspeed misfiring, and other less significant things. The idle problem is solved by keeping 1300rpm or so, or maybe unhooking a small vacuum line will keep it going.

Other than quad drivers, and the 94+ guys I think have the digital egr, I'd suspect the cars would act the same, or take the same to be bypassed.

I really want to put it back the way it was and I prematurely recycled the metal adapter plate so I can't just bolt it back on. ugh. It all started when I needed to put my arm where that egr pipe-to-exhaust was, and once removed, didn't care to repair the egr pipe, and removed the assembly. I'm not claiming a performance gain here, or intent to.

It seems that if the egr unit is removed from the base, block plated, remaining plugged in and grounded, the ses light ignites with its fiery burning sensation we hate to see.

Spark plugs have been in the engine for over 16 years now, maybe I should check those out. They are fine though after that first cold startup minute, and full throttle. I doubt it is plugs because when the egr is operational, they were fine. Car runs like new when that egr is okay.

[DTMAce]

What year and engine are you using?

None of this makes any sense to me. EGR shouldn't be causing these things.

[3X00-Modified]

I ran with my EGR unplugged and none of those issues happened. I can do it now and all I have to disable is the EGR code and it will run fine.

If you have the EGR passage blocked and the EGR itself still plugged in somewhere then yes you'll still get a code for EGR flow, not quite sure how it senses it but it can.

[yellow3800]

its the lemon - 1990 3800 and buick computer. my code for egr is 26, then the rest of the standard egr codes because the egr here is a 'quad driver' [fancy for high-current junk]. what I don't like is that I am also getting other misc codes, map sensor code is one of them. because I have been driving w/out the egr for two months now, it may have crept into the equation, and I can't let a variable like that stay because if thats a faulty reading, the idle problem is the result of that - not my 16yr old spark plugs!

Not having tcc lockup is driving me batty, so I'm going to a pull-it yard and buying another egr mounting plate, so I can put this back. The only way I know to tell if my egr throws all 4 quad drivers I have out of commission is to have a buick reatta egr connector unplugged. if their tcc does not lock up then I'm okay because all of their 'quad driver' circuits fail at once for one fault. if their egr systems act like the digital egr berettas, then I know I made my wire harness wrong, which I doubt.

Either way, I'm not hellbent on removing this thing... but at the same time would like to know how to remove it when it breaks in the future. I figured someone else has done this by now.

I'm willing to bet that the pins ABCD on our digital egr setups are the same:
A=Blue
B=Brown
C=Red
D=Grn which is +12v.

ABC control each individual solenoid, like a low/med/high flow of air from the egr pipe to the intake manifold. the ecu may see that each solenoid resistance, which probably defeats shorting ABC to ground, as the ecu switches each on as desired. If anyone is reading this thread because your egr has a fault, check AandB solenoids for 20-30ohms resistance and solenoid C for 10-17ohms resistance. ...and of course D for +12v... directly quoted from mitchell repair. if okay, then check your ports by removing the valve for a visual. I don't know if beretta ecu's have an onboard circuit breaker like the reattas/lesabres/olds 88/98 etc, so check that out too. I have not looked at my board to see what it looks like. it is there on these other cars though, and since the lemon uses 1989 electronics, later years should have the protection of a circuit breaker so a low resistance solenoid doesn't kill the ecu board.

Am I wrong to give the computer system too much credit to think that the oxy sensor [just one] feedback voltage does not reflect the egr solenoids opening and closing [the egt does not drastically change after opening the egr solenoid if the passages are blocked] and thats why it throws a code?

I think thats what the problem is here. if a stock car gets carbon block in the egr passages, a code throws because it recognizes a blockage -only through oxy sensor feedback not changing as it should to reflect hot air in the intake thus lowering the egt, not by IAC movement because the IAC needs to open to allow more air. I agree with you that the egr blocked should not stall the car within 1min of cold start. I think my computer is slightly different, where if a 94 z26 solenoid 2 fails, one code may result and only the irritating ses light powers on. in my case, and all these buicks, I think it kills all quad drivers, tcc, fuel purge, shifting, elec. spark. timing, etc.

no, removing a 1/8" vac line kills it faster, so its not air the thing needs to stay running at its cold idle. my theory of egr opening to provide intake air so the iac remains calibrated is out the window.

so again, lemme find an egr mounting plate, put this back on, and watch all the problems go away. if this thing needs oxy sensor feedback, I don't see how this can be bypassed by use of a blocking plate unless firmware is reprogrammed. that I have yet to learn. The system is too autonomous for me to be turning potentiometers inside the car, or push-button tcc lockups.

[3X00-Modified]

Well I'll say it again, there have been many others who have removed the EGR with no other issues, Woody being one on a OBDI setup and many others on OBDII. Non have responded the way you've described, so I guess the big issue here is you have a custom harness with a Buick computer so any input we give is useless to you and can't be applied. And if your getting a MAP error then that will cause lots of running issues since the whole fueling system is run by the MAP sensor on a OBDI setup.

[KFLO 93 GT]

I know the Buick computer and it's diagnostic system is most definitely different from our obd-1 Berettas.

My thoughts are checking out the map sensor, swapping it for a spare, and checking out the oxygen sensor.

On my car, the o2 sensor is too far downstream to heat up properly on initial start. So I have to sit there and high idle it for a bit sometimes. A heated o2 will eventually remedy that, wrapping the pipe around the o2 bung in header wrap has helped a bit for the time being.

You've definitely got a problem of some sort Andy, like you, I'd start penciling off the cheap options first. Junkyard some parts up and start swapping them out. Egr, Map, o2 if you find one worth taking. Could be $20 well spent.

[scd88ga]

Mine was a 1994 Supercharged 3800 engine that I installed in my 88 Grand Am. I ran it with no EGR and had no issues.