Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Propane, Butane, LPG, GPL, C3H8, C4H10
Post Reply
Dunksterp
Posts: 14
Joined: Fri Jul 10, 2020 2:22 am

Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Post by Dunksterp »

Hey all!

I'm new here (thanks Frank). I'm based in the UK and made the brilliant decision of buying a Chevy G20 with a TBI SBC installed. Since owning it I've spent far too much on bodywork, and various other bits on bobs. I've now added an LPG system, it runs and drives, but no where near as well as it should.

For tuning purposes, I've purchased an Innovate Wideband AFR. I've also purchased an emulator for the stock ECU so I can completely adjust the spark map on the fly.

The LPG system uses an Ipmco 425 mixer. My main issue at the moment is that it is almost impossible to get a decent AFR at anything other than idle. It always goes lean, then slowly gets almost there at around 15.9:1 After thinking about it for a while, I wondered if it could potentially be down to "not good enough" ignition cables? They're currently just cheapys bought from rockauto.

Just another thing to add.... I'm completely new to this sort of thing and have no experience in engine tuning, so please bear with me.

Another thing I've considered is that the TBI throttle body is pretty restrictive, with it's injector pods right in the middle. Would switching to a carb and potentially a single plan intake be a better option?

Many thanks for any advice.

Here's some pics of the beast in question!

https://i.imgur.com/IZQbIj9.jpg

https://i.imgur.com/BRDQHu8.jpg

storm
Posts: 796
Joined: Wed Jan 10, 2007 10:10 pm
Location: NSW, Australia

Re: Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Post by storm »

Some questions for you.
1. Do you intend to run dual fuel? In other words do you occasionally run petrol?
2. How have you got the 425 setup with the TBI system? some pictures would be helpful here.
3. How have you got the innovate setup? My suggestion would be to set it up to show Lambda values not AFR. If you have it setup to show AFR what AFR are you using as base stoichiometric value (petrol is 14.7-1, LPG is 15.5-1)? Lambda, regardless of fuel type, is always a value of 1 at stoichiometric.
Fuel flow requirements calculations
Engine air flow requirement calculation: CFM = Cubic Inches x RPM x Volumetric Efficiency (VE) ÷ 3456

Dunksterp
Posts: 14
Joined: Fri Jul 10, 2020 2:22 am

Re: Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Post by Dunksterp »

Hi,

1. I sort of plan on running petrol, but only for those emergency times getting to another station, I plan on fully running LPG 99% of the time. I also have the ability to have multiple maps, so if I need to switch back to Petrol, I can just load the petrol map onto the ECU.

2. Here's the set up :

Image

Image

As you can see... Space is fun.. So is heat. I really need to find a way of getting cold air in some how.

3. The innovate is set up to just show AFR right now. I bought it 2nd hand so I don't have the cable right now to change the set up, so I'm assuming it's set up for Petrol.

Thanks again.

storm
Posts: 796
Joined: Wed Jan 10, 2007 10:10 pm
Location: NSW, Australia

Re: Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Post by storm »

I don't know what the petrol quality is like in the UK but if it is at all lower than top quality I'd be running it a bit more than 1% of the time. TBI is bad enough on fresh petrol but when it is stale TBI plays up. We found that out here in Australia with the Global J car (we called it the Camira here) and the late 1980s Ford Falcons.

If you think the ignition cables are a problem check their resistance, follow these steps https://www.championautoparts.com/Techn ... Wires.html

I'm assuming the emulator you mention is a Moates Ostrich2 or the APU1. If you have the Ostrich you will probably need Socket Booster 1 as well. With regards to the Innovate unit I'd be getting the laptop cable before I did anything else with it. Innovate equipment is good, although if you believe what is on the internet many people are complaining about long term quality, but it is always better to start with everything you need and make sure the readings are correct to begin with. You could always get a suitably equipped workshop to do an exhaist gas analysis in the meantime.

Thanks for the pictures, the 1 thing I would comment on is the fact that the air going into the engine is going to be warm at best and very hot at worst. That air filter is in a bad spot. Not knowing the design of those vehicles, (@C3H8) may have some hints here, I'd be trying to plumb it so it gets air at ambient temperature rather than engine bay temperature. You'll find you get a performance and an economy increase (if driven sedately).
Fuel flow requirements calculations
Engine air flow requirement calculation: CFM = Cubic Inches x RPM x Volumetric Efficiency (VE) ÷ 3456

Dunksterp
Posts: 14
Joined: Fri Jul 10, 2020 2:22 am

Re: Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Post by Dunksterp »

With regard to petrol, I've had another think about that and I think going forward, I'll use petrol to start her up and get her warmed up. If I start on LPG the AFRs read crazy low (like 19/20:1). I guess this is because the water temp in the vaporiser is too cold and it's not vaporising the fuel correctly yet?

I tested a couple of my ignition leads and they're way lower than the resistance on Champion's website, so I guess they're fine.

Correct, I'm using the Ostrich 2.0 and the socketbooster on the 7747 ecm. I've disabled EGR, and for LPG, I've created a custom 14-14-14 spark map.

Image

Image

I've ordered a cable for the innovate unit, so I should be able to get that configured correctly next week.

The airflow in these vans is god awful. I've managed to put some 6" ducting coming from underneath the van to just below the air filter. We'll see if it makes any difference, but it can't hurt.

When I loaded the 14-14-14 spark map, I back the power adjustment screw all the way, then opened it up a couple of turns. At WOT I get just into the 15:1 ratio, but then at cruise it seems to just want to sit at 16/17. Which doesn't seem right. I've tried adding more and more fuel, but it never really seems to go much lower. I obviously don't want to toast the engine from running it horribly lean, so any ideas what I'm missing?

C3H8
Posts: 1134
Joined: Fri Sep 09, 2005 11:23 pm
Location: Winnipeg, Canada

Re: Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Post by C3H8 »

The Chevy G20 vans were always difficult to do because of space. Because of this we selected the 300/50 series mixers over the 425. I made a much cleaner installation with only a small sacrifice in airflow. 432 for the 300/50 and 476 for the 425. The 300 allowed the mixer to be closer to the throttle body and did not require a 90 degree turn of the flow. It was still using an open air cleaner but many owners went the extra mile and adapted there factory air cleaner to fit the 300 allowing the cold air intake to be used.

As for your comments on fuel air ratio. Something is amiss. The cold fuel air ratios should be much richer as the vapourizer would not be vapourizing the fuel completely when cold. The mixtures should be much richer.

And why would you disable the EGR. A properly operating EGR provides excellent fuel emission advantages by cooling the combustion temperatures and increasing gas mileage under light cruise conditions. Disabling the EGR is old school.

Lastly the 425 only is adjustable in two areas. The idle and full load. The mixtures you are seeing at cruise are fixed by the contour of the gas valve in the mid range and can't really be changed without changing the gas valve completely or using a lighted orange spring in the regulator which is only done with a fuel control system added. 16-17:1 is actually a pretty good mixture considering stoich is 15.5:1 on LPG and should get you pretty good mileage. This is also one of the reasons we selected the 300 mixer. On that mixer adjusting the idle mixture also will have some affect on the midrange mixtures.

I also don't understand you statement on the power valve. If you back the power valve out it is all the way open already and turning it in is closing the power valve. To correctly read the power valve mixture you will need to be above 3500 RPM and at WOT in 2nd or 3rd gear. The gas valve is not in the power range until this point. Also the preferred ratio would probably be 14 to 14.5:1 at WOT.

Dunksterp
Posts: 14
Joined: Fri Jul 10, 2020 2:22 am

Re: Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Post by Dunksterp »

Hi C3H8,

Thanks for your reply. Unfortunately I didn't know about the option of the 300 mixers, seeing pics, it looks like it would've been easier, but I'll have to make this work.

Have you any ideas what the AFR at startup oddity could be? What you've said makes sense, if it's not vaporised, it'll be vastly richer until it's warmer right?

The EGR is disabled because it had been deleted by a previous owner and it always flagged a check engine light.

Regarding adjustment on the mixer. I was advised that the "top" allen screw hole is for power, and fully closed was fully anticlockwise and fully open was fully clockwise (which seemed odd to me).

One other thing, what is the pipe connector circled in green below. It's currently plugged on mine, but I've seen pictures online with it connected to a pipe, but don't know where.

Image

C3H8
Posts: 1134
Joined: Fri Sep 09, 2005 11:23 pm
Location: Winnipeg, Canada

Re: Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Post by C3H8 »

The innovate monitor was not used in my days. I had a look at it and the one conclusion I can come to is that it uses an oxygen sensor for the readings. Typically oxygen sensors require a minimum temperature to work properly. Most require a temperature of 650 degrees to be accurate. Most of todays sensors have an electric heater to support the heat however they still take 30 seconds to a minute to heat up. Possibly the sensor is not warm enough to be accurate. One of the others that use this device may be able to add some information here.

Your info on the power valve adjustment is incorrect. Clockwise to close. CC to open. The factory setting is approximately 4 turns in from fully open.

The brass fitting you circled can be used to provide vacuum to the lockoff or to attach an SV. (start assist valve or primer). There are several silver screws around the base to allow convenient positioning of this fitting. That black pipe plug in the picture of your L vaporizer goes into the secondary chamber of your regulator. Placing an SV valve in there and routing it the brass fitting will provide quicker cold starts on propane.

By the way, there is nothing wrong with starting on gasoline. Gasoline goes stale eventually and using a little at a time for starting will require adding some from time to time to refresh it.

C3H8
Posts: 1134
Joined: Fri Sep 09, 2005 11:23 pm
Location: Winnipeg, Canada

Re: Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Post by C3H8 »

Oh! One last item. If you don't have anything attached to that brass fitting it has to be plugged. That is a massive air leak if it is left open and would drastically lean out the mixtures.

Dunksterp
Posts: 14
Joined: Fri Jul 10, 2020 2:22 am

Re: Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Post by Dunksterp »

C3H8 wrote:
Tue Jul 21, 2020 8:36 am
The innovate monitor was not used in my days. I had a look at it and the one conclusion I can come to is that it uses an oxygen sensor for the readings. Typically oxygen sensors require a minimum temperature to work properly. Most require a temperature of 650 degrees to be accurate. Most of todays sensors have an electric heater to support the heat however they still take 30 seconds to a minute to heat up. Possibly the sensor is not warm enough to be accurate. One of the others that use this device may be able to add some information here.

Your info on the power valve adjustment is incorrect. Clockwise to close. CC to open. The factory setting is approximately 4 turns in from fully open.

The brass fitting you circled can be used to provide vacuum to the lockoff or to attach an SV. (start assist valve or primer). There are several silver screws around the base to allow convenient positioning of this fitting. That black pipe plug in the picture of your L vaporizer goes into the secondary chamber of your regulator. Placing an SV valve in there and routing it the brass fitting will provide quicker cold starts on propane.

By the way, there is nothing wrong with starting on gasoline. Gasoline goes stale eventually and using a little at a time for starting will require adding some from time to time to refresh it.
Hi,

Thanks for the info. I've had another look and tried adjusting the impco to default for now. I'm still VERY confused about the AFR readings. I've now set the Innovate AFR unit up for LPG, but this hasn't changed anything as my readout simply displays the current air fuel ratio and not Lambda. On petrol the unit will display around 14:1 from cold start during idle and then go down to about 12:1 during acceleration (I've currently got it set to forced open loop mode to make things simpler tuning). So when I start using LPG at cold, it'll start at 20:1 and then very slowly go down to about 15:1 once fully warmed up. If I give it a bit of throttle (just parked, no load) the AFR richens up to around 18:1. During around town cruising, up to about 40 MPH currently it seems to sit around 17/18:1, but then if I floor it, it'll go to about 15:1. Are these dangerously lean AFRs? I REALLY do not want to damage my engine and mess everything up.

As a note, when driving it around, it feels pretty good, it doesn't hesitate or stutter, and generally feels ok to drive. I get a tiny amount of knock detected at full throttle, but I'll be adjusting my timing curve to account for that and remove it.

With regard to the brass fitting, I've had that plugged since I installed it as I thought it'd be a massive leak.

Another think I'm doing to tackle heat somewhat is wrapping my headers in heat wrap to try and transfer some away from the engine bay.

Thanks for any advice, I know it must feel like you're trying to explain to a simpleton, but I'm learning slowly.

EDIT - Another thing I've noticed while I've been browsing the forums today and looking for GM TBI specific stuff is that you've mentioned that the standard computer can handle the timing.... If that's the case, is my timing table a tad on the aggressive side, should I wind it back a few degrees?

Cheers

Dunksterp

Dunksterp
Posts: 14
Joined: Fri Jul 10, 2020 2:22 am

Re: Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Post by Dunksterp »

Well, after another afternoon of tinkering, I decided to take the mixer off and discovered a completely destroyed gasket on the TBI unit, between it and the collar. Found it with good old brake clean!

Image

Image

Since replacing this, my AFRs make a hell of a lot more sense. I've also adjusted my timing table quite heavily after reading this thread

viewtopic.php?f=5&t=1300&start=75

Image

Things feel more peppy. It's definitely better. The AFR on startup and idle is still strange, it'll start at about 17, then very slowly goes all the way down 13:1 at idle when it's fully warmed up. I thought it was meant to behave the other way round?

storm
Posts: 796
Joined: Wed Jan 10, 2007 10:10 pm
Location: NSW, Australia

Re: Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Post by storm »

Dunksterp wrote:
Sat Aug 01, 2020 2:00 pm
Image

Things feel more peppy. It's definitely better. The AFR on startup and idle is still strange, it'll start at about 17, then very slowly goes all the way down 13:1 at idle when it's fully warmed up. I thought it was meant to behave the other way round?
That table doesn't look right to me, it is setup poorly with the axis in the wrong location and the rpm to low.
What is your memcal ID? It is usually a 4 letter code sometimes followed by 4 numbers. It will be written on the label on the outside of the ECU and also on a tag on the memcal itself.
Where did you get the ADX and XDF from?
Others may have seen it but I've never seen a chev V8 with a timing of 0 degrees at any rpm / map on a naturally aspirated engine.
Fuel flow requirements calculations
Engine air flow requirement calculation: CFM = Cubic Inches x RPM x Volumetric Efficiency (VE) ÷ 3456

Dunksterp
Posts: 14
Joined: Fri Jul 10, 2020 2:22 am

Re: Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Post by Dunksterp »

storm wrote:
Sat Aug 01, 2020 3:19 pm
Dunksterp wrote:
Sat Aug 01, 2020 2:00 pm
Image

Things feel more peppy. It's definitely better. The AFR on startup and idle is still strange, it'll start at about 17, then very slowly goes all the way down 13:1 at idle when it's fully warmed up. I thought it was meant to behave the other way round?
That table doesn't look right to me, it is setup poorly with the axis in the wrong location and the rpm to low.
What is your memcal ID? It is usually a 4 letter code sometimes followed by 4 numbers. It will be written on the label on the outside of the ECU and also on a tag on the memcal itself.
Where did you get the ADX and XDF from?
Others may have seen it but I've never seen a chev V8 with a timing of 0 degrees at any rpm / map on a naturally aspirated engine.
So the van originally came with a 16146299 ECU, with a AWLD chip in it. I've since replace the ECU with a 1227747 because it offers far more tuning options and is a straight swap, it had an ANLW chip in it. This has been removed and I'm now using an Ostrich 2.0 with the Socketbooster. This set up is all working as far as I can tell.

I've got all my bin, adx and xdf files from Gearheadefi.com :

http://www.gearhead-efi.com/Fuel-Inject ... rmation-42

I'm using the AMUR bin as it seemed to make most sense for the van. This is the default spark table:

Image

I've basically butchered the table somewhat to try and bring in 32° by 2800 rpm using excel to smooth it out. I'm all ears on this and very willing to learn.

storm
Posts: 796
Joined: Wed Jan 10, 2007 10:10 pm
Location: NSW, Australia

Re: Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Post by storm »

Well that possibly complicates things a bit.
I have read your threads on GearheadEFI and don't see anywhere if anyone asked you what transmission type you have. I'm assuming, considering the memcals you have been using, it's an automatic but it is an electronic or manual (meaning TV cable) automatic.
The next thing, and this is one reason I ask what type the automatic is, you have swapped a 92 ECU for an 87 ECU. I know you think it has more features but if the "new" ECU and trans don't match you are going to be chasing your tail. Not being, at all, knowledgeable about Chevy Vans this may take a little while to work through trying to find what is going on.
Fuel flow requirements calculations
Engine air flow requirement calculation: CFM = Cubic Inches x RPM x Volumetric Efficiency (VE) ÷ 3456

Dunksterp
Posts: 14
Joined: Fri Jul 10, 2020 2:22 am

Re: Tuning a '92 Chevy G20 350 SBC TBI with LPG installed...

Post by Dunksterp »

storm wrote:
Sun Aug 02, 2020 3:40 pm
Well that possibly complicates things a bit.
I have read your threads on GearheadEFI and don't see anywhere if anyone asked you what transmission type you have. I'm assuming, considering the memcals you have been using, it's an automatic but it is an electronic or manual (meaning TV cable) automatic.
The next thing, and this is one reason I ask what type the automatic is, you have swapped a 92 ECU for an 87 ECU. I know you think it has more features but if the "new" ECU and trans don't match you are going to be chasing your tail. Not being, at all, knowledgeable about Chevy Vans this may take a little while to work through trying to find what is going on.
Thanks for having a check, but I'm running a 700R4 or a very early 4L60 (not the electronic version). There are no electronics on my tranny. It selects all gears with no issues and locks up the TCC.

Post Reply