ECM test bench.

canadian

sy in progress
I know many moons ago Bruce Plecan had set up and ECM test bench (there's probably photos on diy-efi) so he could fiddle around with the in's and out's of ECMs.

I don't suppose anybody has any plans on how to set one up for our ECM's? I would like to play around with the input/output/logging of an ECM, but don't want to sit in the garage with the truck running for hours on end... mainly because that means I'd have to get the truck running.

Anybody have any ideas?
 

smeagol

Active member
You could just put it on a battery charger and put key on if you are just experimenting with logging. The computer is still on and sending data, just a lot of zeroes :wink:
 

leroy

Donating Member
Brian Dourty and I got an old computer with a power supply.....most power supplies have a 12v and 5v source. I went to the salvage yard and got some ecm connectors with wires attached, repinned as necessary and labeled everything. We bought pots, 555 timers and leds, but didn't get a chance to finish it. Brian can probably answer specific questions.

Jim
 

canadian

sy in progress
i *could* sit in the truck and try to program on a crappy old laptop, but would prefer to do it inside where it's nice and cozy. i may have to drop brian a line and see if he's got some wiring diagrams and whatnot lying around.
 

smeagol

Active member
I did have the info handy for getting the ECM connectors (either side, harness or ECM plug end).

I was going to hunt that info down again, as I have another project in mind :) (or should I be :evil: )

lemme know if anyone trips across that info.
 

turbodig

Active member
Re: ECM test bench.

canadian said:
I know many moons ago Bruce Plecan had set up and ECM test bench (there's probably photos on diy-efi) so he could fiddle around with the in's and out's of ECMs.

I don't suppose anybody has any plans on how to set one up for our ECM's? I would like to play around with the input/output/logging of an ECM, but don't want to sit in the garage with the truck running for hours on end... mainly because that means I'd have to get the truck running.

Anybody have any ideas?

I've got one set up on my desk... only about 4 wires on it are connected.

All you need to to to make it run (and connect to a scan tool) is hook
up power, ground, ign power, and the data lines from the scan tool.

The pinout is all in the Sy/Ty suppliment... basically, just connect the
sensors you want to the appropriate ECM pins. The diagrams in the
individual trouble code pages help a lot for this.

Much of Bruce's setup for the C3 vintage ECM still applies to our ECM...
the pinout just changes.

Really, all you need to have for good basic functionality is:

555 timer set up to run RPM
555 timer set up to run MPH ( just use a dual timer for both)
Hook up a real tps to the right pins
Hook up a MAP to the right pins (and have a mityvac to pump it up)
Rheostat hooked up to the MAT input (and ground)
Rheostat hooked up to the Coolant input (and ground)

You can add injectors, IAC, o2 to this mix, but it doesn't serve much
purpose unless you're investigating these particular subsytems.

As far as a harness, you need not be picky. Find any '730 equipped car,
and you'll find a harness. Late 80s Pontiac 6000s are good targets.

In fact, the whole '730 ECM will work, unless you're wanting to drive a
WG solenoid. It runs $58 Sunbird/Syty code just fine.

I run my setup off a 3 Amp Radio Shack Power Supply. If you want to
power injectors or an IAC or an 02 sensor with that, forget it. You'll
need 20 amps or so to set up full functionality. Bruce has his set up in
his kitchen, running off a car battery, with a charger running on the bat.

Needless to say, it's quite a kitchen. :)

Later,

Dig
 

myclone

Donating Member
Re: ECM test bench.

turbodig said:
canadian said:
I know many moons ago Bruce Plecan had set up and ECM test bench (there's probably photos on diy-efi) so he could fiddle around with the in's and out's of ECMs.

I don't suppose anybody has any plans on how to set one up for our ECM's? I would like to play around with the input/output/logging of an ECM, but don't want to sit in the garage with the truck running for hours on end... mainly because that means I'd have to get the truck running.

Anybody have any ideas?

I've got one set up on my desk... only about 4 wires on it are connected.

All you need to to to make it run (and connect to a scan tool) is hook
up power, ground, ign power, and the data lines from the scan tool.

The pinout is all in the Sy/Ty suppliment... basically, just connect the
sensors you want to the appropriate ECM pins. The diagrams in the
individual trouble code pages help a lot for this.

Much of Bruce's setup for the C3 vintage ECM still applies to our ECM...
the pinout just changes.

Really, all you need to have for good basic functionality is:

555 timer set up to run RPM
555 timer set up to run MPH ( just use a dual timer for both)
Hook up a real tps to the right pins
Hook up a MAP to the right pins (and have a mityvac to pump it up)
Rheostat hooked up to the MAT input (and ground)
Rheostat hooked up to the Coolant input (and ground)

You can add injectors, IAC, o2 to this mix, but it doesn't serve much
purpose unless you're investigating these particular subsytems.

As far as a harness, you need not be picky. Find any '730 equipped car,
and you'll find a harness. Late 80s Pontiac 6000s are good targets.

In fact, the whole '730 ECM will work, unless you're wanting to drive a
WG solenoid. It runs $58 Sunbird/Syty code just fine.

I run my setup off a 3 Amp Radio Shack Power Supply. If you want to
power injectors or an IAC or an 02 sensor with that, forget it. You'll
need 20 amps or so to set up full functionality. Bruce has his set up in
his kitchen, running off a car battery, with a charger running on the bat.

Needless to say, it's quite a kitchen. :)

Later,

Dig

What about an O2 input signal? Wouldnt the lack of x counts throw a code and put the ECM into limp home mode? How bout the rheostats for the MAT and coolant temp? Can pots be substituted here since those are low current apps? No laughing if these are tard questions.. OK?

TIA
 

Ian Turgeon

Cascading Inspiration
as far as the 02, bg left mine unattatched after working on my truck, The only way I knew was by hooking up a scan tool and noticing a problem switching between open/closed loop. no code, no noticable drivability differences.
 

kwick6

Donating Member
myclone- not sure if you seen this, or if it will help, ( Bruce's testbench ) but I would also like to know if you must have an 02 feedback.

It sounds to me like Dig was saying if you want to power an actual 02, then the 20A was needed....*or* seen below it looks as though Bruce just made a low voltage signal to replicate the 02 cross counts.



The hardest thing to do, is the feedback O2. I don’t have a final solution
for that, but by using the ignition coil signal as a trigger for a 555
(adjustable for “on” time), you can generate enough feeback to keep
a cross count running, and not have the ecm get to dumb, or set any codes.
To get a .9v regulated signal I ran the 555 output to a 1K resistor, then
a LED, a LED has about a 1.8v drop across it. So two 10K resistors
across that give you a voltage divider of 2, and an output of .9v..


Scott
 

kwick6

Donating Member
SyTyArchives said:
as far as the 02, bg left mine unattatched after working on my truck, The only way I knew was by hooking up a scan tool and noticing a problem switching between open/closed loop. no code, no noticable drivability differences.

Probably using an open loop chip. :wink:

Scott
 

Sy#26

New member
Got a question for Dig (or anyone) -

I've got a benchtop ECM running & notice one thing that bothers me. MAPs seem to differ by around 3psi from the bench version to the one in the truck. Set something in the PROM to happen at, say, 16psi on the bench and it happens at 19 or so psi in the actual ECM in the truck. Same ALDL software, cal'd the MAP (I use a pot on the benchtop) on both units.

Any suggestions?

Thanks -

Barry - Sy#26
Bangor, MIi
 

myclone

Donating Member
leroy said:
plumbing lag + reaction time of the sensor?

Jim

Agreed.. and poss tolerance stack on the components? Whats the tolerance of just the resistors in the ECM(Ive never paid attention when I had mine open)?

Prolly completely irrelevant but Ive noticed a ~3psi "offset" in my boost gauge reading vs datamaster. I checked the cal on the boost gauge with a high accuracey gauge at work and its linear and right on the money as far as measured vs actual. I figured it was tolerance stack between the MAP, ECM, laptop, and the software. Dunno.
 

turbodig

Active member
Re: ECM test bench.

myclone said:
turbodig said:
canadian said:
I know many moons ago Bruce Plecan had set up and ECM test bench (there's probably photos on diy-efi) so he could fiddle around with the in's and out's of ECMs.

What about an O2 input signal? Wouldnt the lack of x counts throw a code and put the ECM into limp home mode? How bout the rheostats for the MAT and coolant temp? Can pots be substituted here since those are low current apps? No laughing if these are tard questions.. OK?

TIA

Generally, the o2 isn't too big of a concern, unless you're actually messing
with integrator and transport delay stuff. The .450 excitation voltage
internal to the ECM ends up being right around the switching point, so
it's kinda self-emulating. Plugging in an o2 would likely generate more
codes, since the sensor would go lean permanently and trip the code.

A rheostat is, by definition, a variable resistor. A Pot is a variable resistor
with an extra leg. Tie the wiper of the Pot to one of the legs, and you
have a rheostat. It's pretty rare to see a purpose-built rheostat in use
anymore, it's cheaper to use a Pot.

Short answer... yes. :)

Later,

Dig
 

turbodig

Active member
Sy#26 said:
Got a question for Dig (or anyone) -

I've got a benchtop ECM running & notice one thing that bothers me. MAPs seem to differ by around 3psi from the bench version to the one in the truck. Set something in the PROM to happen at, say, 16psi on the bench and it happens at 19 or so psi in the actual ECM in the truck. Same ALDL software, cal'd the MAP (I use a pot on the benchtop) on both units.

Any suggestions?

Thanks -

Barry - Sy#26
Bangor, MIi

Might be grounds. You've got a lot less wiring on that bench, less voltage drop.

You also might be seeing a pressure difference in where your pressure
gauge is sitting, vs the signal to the MAP.

I've noticed slight differences in what calculated MAP should be (ie, 15psi
should be 200 kpa ) and what it really is. I haven't found a really good
explanation for it, other than possible problems in how baro is calculated.

On mine, though, everything is consistent between what I see on the
bench vs. the truck itself.

Hooking up a pot to the MAP might give you some insight to this... see
what the voltage is at the pot when the scan tool reads 100 kpa. Both on
the truck, and on the bench.


Later,

Dig
 

kwick6

Donating Member
Just run open loop, no 02 feedback.

Not necessary to play with all that int/BLM crap, unless your trying to pass emissions.


Scott
 

UNIsackV6

Member
Bill just an FYI I don't even have a knock sensor anymore!!! Roller valvetrain and all no way it would be accurate, Dig is right sounds like piston slap remember forged parts need a few minutes to warm up to operating temps. My truck runs like crap at first and then whaammo get the fuck out of the way!!!! Start logging things like EGT versus boost and RPM etc...if you're EGT is good and its says its knocking its probably BS, but I don't want you to toast your motor, stick with what BG and Dig tell you and take baby steps you've come to far to start all over if you know what I mean!!!! Somehow I missed the whole warm up part of your post because as soon as I read it from Dig's post I was thinking DUH!!! Yeah sounds like piston slap...keep at it...
 

myclone

Donating Member
Re: ECM test bench.

turbodig said:
myclone said:
turbodig said:
canadian said:
I know many moons ago Bruce Plecan had set up and ECM test bench (there's probably photos on diy-efi) so he could fiddle around with the in's and out's of ECMs.

What about an O2 input signal? Wouldnt the lack of x counts throw a code and put the ECM into limp home mode? How bout the rheostats for the MAT and coolant temp? Can pots be substituted here since those are low current apps? No laughing if these are tard questions.. OK?

TIA

Generally, the o2 isn't too big of a concern, unless you're actually messing
with integrator and transport delay stuff. The .450 excitation voltage
internal to the ECM ends up being right around the switching point, so
it's kinda self-emulating. Plugging in an o2 would likely generate more
codes, since the sensor would go lean permanently and trip the code.

A rheostat is, by definition, a variable resistor. A Pot is a variable resistor
with an extra leg. Tie the wiper of the Pot to one of the legs, and you
have a rheostat. It's pretty rare to see a purpose-built rheostat in use
anymore, it's cheaper to use a Pot.

Short answer... yes. :)

Later,

Dig

Thanks Dig.. I was thinking there was a reason that rheostats were used instead of pots (inductance, noise, better flux capacitor resolution, etc ;) ).
 
Top