Engine/Transmission (1998.5 - 2002) Anatomy of an APPS

[mhenon]

Wonder if Tim had a rough spot on the lower end of the OEM APPS?

 

[mhenon]

..... I'm starting to think about a somewhat plug and play version on the bus APPS. You take the APPS bracket off, you swap the APPS (2 screws, no adjustment, just screw it on), plug the OEM APPS connector into a box, plug the bus APPS connector into the box. Done, crank her up.

This is pretty much where I'm headed Bob. I'll first test the bus APPS the way you did in the OEM bellcrank. . but have the relay box in between. Only after this proves reliable will I go ahead and do the whole pedal assembly. After looking at Gary's post above. . those connectors would make it so you wouldn't have to cut any OEM wiring (like I did!). The relays box could have the OEM connector on one side and weather packs (for bus APPS) on the other. Couldn't be easier!

Mike

 

[mhenon]

Just had a thought... The bus APPS has upside-down logic that pulls the lines to +5 volts instead of ground. Rather than using relays to reverse the logic, it might work to just put a resistor from each of the bus IVS lines to ground, and switch the idle and not-idle lines between the bus APPS and ECM. When the bus APPS pulls one of the lines to 5 volts the ECM would see that, and when the APPS releases the line the resistor would pull it to ground and the ECM would still be happy. I'd try 1K resistors.



I don't have a bus APPS to try it out, but if anyone else wants to...



Karl

Karl, Let me see if I got this right. After adding the 1K ohm resistors to the bus IVS you would then "X" the idle/accelerate lines coming from the ECM. So "idle" from ECM would go to "accelerate" to bus IVS and vice versa. This way when bus IVS idle was hot the ECM accelerate line would see an "open' circuit just like it's supposed to.

Can you post a small schematic. I've got to think about this for a few seconds but think it's genius. would the lines from the ECM go after the resistor (between resistor and ground) or doesn't it matter? How do we figure out what resistance would be best? Wouldn't want to send some stray voltage back to the ECM. That could get exspensive!!!

Mike

ps: i'll have time this weekend to play with this.

 

[rweis]

Well, I am going to test this 1 step at a time.



Component - Using bus APPS, Using bus IVS running relays to ground for ECM



Try recalibrating the APPS (batteries disconnected 30 minutes, reconnect batteries, key to on, pedal to floor slowly and release slowly. Key off. Key to start, start engine, check CEL lights)



If that still gives 0222 Idle validation signals both low. Right now I use the same ground to ground the coil and the NO switch. Could this be giving feedback to the ECM? Put seperate grounds on the coil and on the NO switch. Check CEL lights.



Swap dc APPS back in recalibrate APPS check for CEL



Check IVS voltages with intent to make bus IVS voltages the same



Swap bus APPS back in



Recalibrate bus APPS expect CEL on engine start



Start engine expecting CEL 0222, get 0222



Change ground on relays to insert resistor between ground and ECM to provide xxx resistance and therefore xxx voltage to ECM same as dc APPS



If CEL is cleared, drive this way for a week.



If CEL is not cleared, use bus APPS for APPS only and procede with MS installation for IVS, remove bus IVS support relays etc, clip bus IVS wiring back to physical APPS and render IVS wiring inert.



Test MS IVS for CEL expecting no CEL rendering final solution bus APPS for APPS only, MS for IVS.



Bob Weis

 

[KarlC]

Karl, Let me see if I got this right. After adding the 1K ohm resistors to the bus IVS you would then "X" the idle/accelerate lines coming from the ECM. So "idle" from ECM would go to "accelerate" to bus IVS and vice versa. This way when bus IVS idle was hot the ECM accelerate line would see an "open' circuit just like it's supposed to.

Yes. That's the idea.

Can you post a small schematic. I've got to think about this for a few seconds but think it's genius. would the lines from the ECM go after the resistor (between resistor and ground) or doesn't it matter? How do we figure out what resistance would be best? Wouldn't want to send some stray voltage back to the ECM. That could get exspensive!!!

Mike

ps: i'll have time this weekend to play with this.

There shouldn't be a problem sending anything bad to the ECM. And it shouldn't matter what order things are connected up. In this case you could think of it as water -- doesn't matter what order a sink, washing machine, and dishwasher are connected to the pipes, the water still gets to all three.



The resistance value depends on what circuitry the ECM has internally connected to these lines. 1K is a guess but likely to work. I'll try to make a measurement on my truck today to figure that out a little better.



Here's a schematic.

 

[mhenon]

I think you're right on this. I was worried about sending voltage back to the ECM in a way it didn't want. A simple check of the voltage on DC pins 6 & 2 when they're active should verify that 5 volts is there but really... what else would it be!

Let me know if you come up with a different resistance value. I should have a few 1k's around.

This will simplify things quite a bit if we can drop the relays from the mix.

Mike

 

[KarlC]

I just did some checking on my truck, and the resistor thing may not work. Trouble is that it would require somewhere around a 150-ohm resistor from each leg of the IVS switch to ground, and that would draw 33 milliamps of current from the 5-volt supply. I don't know, but that may be too much and I wouldn't want to chance it if it's on the edge.



Does anyone know how much current the 5-volt supply can safely put out?

 

[Gary - K7GLD]

Does anyone know how much current the 5-volt supply can safely put out?

The limit is most likely pretty much whatever a typical 5 volt regulator is capable of - but also depends on other items using the same 5 volt bus - but I'd guess a 33 ma drain wouldn't be a problem...

 

[KarlC]

I think you're right on this. I was worried about sending voltage back to the ECM in a way it didn't want. A simple check of the voltage on DC pins 6 & 2 when they're active should verify that 5 volts is there but really... what else would it be!

Let me know if you come up with a different resistance value. I should have a few 1k's around.

This will simplify things quite a bit if we can drop the relays from the mix.

Mike, my concern was that it would be 5 volts when measured, but later it might drop due to different load, temperature changes, etc. It would probably set a few codes and you'd probably be back to idle only. I doubt it would permanently damage anything. You could try it out, using 150-ohm resistors instead of 1K.



Karl

 

[rweis]

WAY better than a set of relays!



What I know about the bus IVS.



The bus IVS measured 13 ohms closed.



I think I read that the dc APPS IVS was looking for 20 - 125 ohms.



Hence why I am getting "low voltage on IVS line(S)" (S FOR EMPHASIS)



Notice it does function correctly, just LOW VOLTAGE.



I was going to add a "tad" (50 ohms?) to each of the ground sides (idle & throttle) of the relays to get the bus IVS more in the middle of the dc IVS.



I like the idea of swapping the 2 / 6 , much much simplier.



ASIDE Thoughts on MS mounting if you go this way



We are getting different ways to mount the MS to interact with the bell crank. Some require welding, some require drilling and tapping. What about mounting the MS inside on the "go" pedal linkage? I have had a MS mounted on the pedal linkage for my exhaust brake. Same idea. Might be more room, easier to get the MS situated to open / close based on pedal linkage movement instead of the bell crank where the available room is limited and not real conducive to adding brackets.



Wire it up to an anti theft switch as well? No IVS you are NOT going to accelerate the VP44, we KNOW that.



Just an idea,



Bob Weis

 

[Gary - K7GLD]

Up further above, I outlined a possible approach for guys wanting to abandon the electronic IVS in favor of a MS to potentially extend the lifespan of their APPS, or possibly resurrect one that failed due to a bad IVS section.



Just for fun, I've ordered the MS and plugs to make up an extender harness for that suggested setup, and will present that project when the stuff arrives, using the APPS removed from my truck when I installed the heavily modded one I'm now running...



If we keep providing good alternative working solutions, we stand a better chance of producing something usable for a wider number of guys here...



Besides - it's sorta fun, now that we know more about what we're doing! :-laf



MEANWHILE, my version is working GREAT - no issues of any kind!

 

[mhenon]

... Does anyone know how much current the 5-volt supply can safely put out?

I could rig up a small pot to the IVS supply (wherever I tap it) and back off on resistance until the supply voltage start's to drop. Measure the remaining resistance and you could come up with a "never to exceed" amperage.

Do we need the resister at all? The OEM APPS is sending that same 5 volts right to ground when it throws to "idle" or "accelerate" so maybe the load is being handled in the ECM as is. Just a thought.

Mike

 

[mhenon]

... Besides - it's sorta fun, now that we know more about what we're doing!

Gary, it has been fun. It's also been a great learning experience. Anybody can fix the easy stuff... this has been a real head scratcher at times.

For myself I enjoyed ever second of it. All that's left is to clean up the rough edges and tie a ribbon on it. Great thread!

Mike

 

[rweis]

I think the schematic of the IVS wiring shows ground, but it is not really ground. I think the ECM knows it is in idle or throttle by sensing some voltage (. 5 volts?).



That is why my 0222 code is showing up because I really did assume the IVS back to the ECM IS ground. On the relays I put the ground side of the coil to that IVS "ground", and I tied the ground side of the relay switch to the same "ground". Which probably provides a slight voltage, but by the 0222 code not enough voltage.



I now think that the ground side of the coil needs to go to battery ground (true ground), and the ground side of the realy switch needs to go to the IVS ground and maybe a 50 ohm resistor needs to be in that line.



The dc IVS circuit has some resistance in it and the bus IVS circuit just has to match the same resistance.



Bob Weis

 

[mhenon]

I think the schematic of the IVS wiring shows ground, but it is not really ground. I think the ECM knows it is in idle or throttle by sensing some voltage (. 5 volts?)...

I'll check that out today. There should be some resistance between IVS ground and chassis ground if there's a resistor in there.

Mike

 

[KarlC]

I could rig up a small pot to the IVS supply (wherever I tap it) and back off on resistance until the supply voltage start's to drop. Measure the remaining resistance and you could come up with a "never to exceed" amperage.

Do we need the resister at all? The OEM APPS is sending that same 5 volts right to ground when it throws to "idle" or "accelerate" so maybe the load is being handled in the ECM as is. Just a thought.

Mike, You could test the 5 volt supply like you said to see how much current it would be capable of. Be careful if you do, since like Gary said it might put out 1/2 amp or so. Might burn out the pot you put across it. A possibility is that the ECM cuts off the 5 volt supply if it gets overloaded, and has to be reset by the dealer's scanner. I remember reading about something that worked that way but I don't remember what. Maybe Gary knows about this. I think a better approach might be to check that it will put out 50 milliamps ok. That way we'll know that it will put out the 33 milliamps with no problem.



I'm sure you do need the resistor. Otherwise the switch would be shorting out the 5 volt supply. In the DC APPS the switch connects to ground but the line it's connecting isn't tied right to the 5 volt supply, it's going through some other circuitry in the ECM before it gets to the 5 volts.

 

[Gary - K7GLD]

I think the schematic of the IVS wiring shows ground, but it is not really ground. I think the ECM knows it is in idle or throttle by sensing some voltage (. 5 volts?).



That is why my 0222 code is showing up because I really did assume the IVS back to the ECM IS ground. On the relays I put the ground side of the coil to that IVS "ground", and I tied the ground side of the relay switch to the same "ground". Which probably provides a slight voltage, but by the 0222 code not enough voltage.



I now think that the ground side of the coil needs to go to battery ground (true ground), and the ground side of the realy switch needs to go to the IVS ground and maybe a 50 ohm resistor needs to be in that line.



The dc IVS circuit has some resistance in it and the bus IVS circuit just has to match the same resistance.



Bob Weis

I dunno if the "ground" for the APPS/IVS is actually at direct ground potential, like the engine block or vehicle frame/body, but if you look at the ECM schematic, virtually ALL the various sensors are connected to the SAME ECM ground bus. Which is also why when the ground connection is bad, you get all sorts of sensor error codes...

 

[mhenon]

OK... EVERYONE SITTING DOWN

The Williams APPS works as is... no capacitors, resistor or relays!!!!!!!!

I finally got a day to work on this thing and it's done. Bob, just get rid of all the extra goodies you added to get the resistance and voltage to match the DC APPS. The ECM will learn the new bus APPS just like it does the OEM APPS. No need to supply 5 volts to the IVS either. Our IVS wants ground where the bus IVS wants 5 volts. I thought maybe there was some interplay between the TPS and IVS circuit requiring this 5 volts to make it all work. Not the case. . the IVS circuit is a dumb relay triggered by the wiper voltage (a guess here) but once triggered it could care less if it relays 5 volts or ground.

The idle voltage at pin 23 was a little high (. 724v vs . 525v with the DC APPS) when the bus APPS was in place on the bellcrank. Just back off the the idle set screw and get it as low as you can. This turned out to be . 634 volts. I then cranked in on the set screw until the idle voltage came up a hair. . like . 635v. Didn't want the stop point to be the internal stop on the APPS. Same thing on the upper stop screw. WOT on the DC APPS was 3. 336 volts and 3. 836v on the bus APPS. I forgot to write down where I set that but it doesn't matter. . the ECM will learn it. Probably any figure under 4. 5 volts or so would work for the high side.

With everything buckled up I did the key on pedal to the floor reset and took off for a test run. No codes or hiccups of any kind. I tried everything from WOT to running down the road in 5th gear at idle. Also tried throwing the switch on my Quadzilla XZT+ while running. . just anything that I thought might confuse the ECM in case we were pushing the envelope somewhere. It runs perfect. I will test it for a few weeks before installing the Williams pedal assembly in the cab but think it's a done deal. This really looks good.

Mike

 

[mhenon]

Here's a few pic's of the install. The Willaims APPS along with these 6 pin connectors:Deutsch DT Series 6-Way Connectors makes this about as plug & play as you can get.

First photo show the Williams APPS as installed on the OEM bellcrank. Second shot (a little fuzzy) is the idle set screw that needs to be adjusted to bring the voltage down. Third shot is the finished install (minus the cover). Thats about as stock as you can get.

Mike

 

[Gary - K7GLD]

That's great Mike - get some miles on that baby, and keep us informed! Oo.