[cgrey8]

A buddy of mine has a 95 Jeep. I don't know exactly what model but I'm assuming it's a CJ-something. The previous owner ripped the L4 out and planted a 350 Chevy and TH350 into it. From pictures he snowed me, it's got the works...heads, intake, headers, who knows what else inside. The owner isn't an engine guy so about the only thing he could tell me is it's a 350 Chevy and it can bark the tires. Anyway it was an impulse buy for him and he didn't even stop to consider whether it would pass Ga emissions. And given the original owners live outside emission-testing Atlanta, it's got a carb on it...so no chance it'll pass. He put CATs on it and that helped, but still no pass. I don't know how far out it was to know if just dumping some higher ethanol concentration in the tank and dialing down the dizzy would be enough to get it to pass. But assuming that wouldn't be enough, I got to thinking more technical.

I know Chevy TBI systems are fairly straight forward...basically just a carb-shaped throttle body with injectors mounted on it. So how difficult would it be to rip an old TBI off a junker or off eBay and install it with make-shift wiring and a HEGO in one pipe each time he needs to get it tested. What else do those TBI computers need? MAP? TPS? Anything else? What kind of fuel pressure do they run? That is probably going to be the PITA part if an inline returnless pump can't be rigged into the engine bay. Also as you can imagine, there's no EGR. So whatever computer would either need to be tuned to disable EGR or somehow have it spoofed (if that's possible).

If all works, all would get ripped out and the carb right back on for another year. Emissions guys here are willing to turn a blind eye so far, but they just aren't willing to be so blind as to pretend what's on the rollers is something other than a 95 Jeep. It's only got to pass for another 3 years so a little pain is doable...but the fuel pump issue might be the show-stopper if an inline returnless isn't an easy thing to rig.

Note, the TBI doesn't need injectors large enough to run the engine into high RPM WOT, it only needs to run it well enough to get it to the emissions place, pass emissions while in Closed Loop, and get him back to his house. How easy could this be done? I told him I didn't mind helping to research this some and possibly helping show him how to get it done. But I don't want to own this project. So it's gotta be something that can be done in an afternoon.

Thoughts?

[decipha]

find out what numbers he put out and what the limit is

if the carb is dialed in then simply locking the timing to 10 degrees should do the trick

if its not dialed in then thats where id start, get the carb dialed in and see where its at

What ignition system is on it now?

if hes going to spend the time going fuel injection then he might as well stay with multi port injection

you can mount the regulator at the fuel tank so you only need one line no prob

[cgrey8]

find out what numbers he put out and what the limit is

When I asked yesterday, he thinks it was failing on HC (too rich). So putting some E85 in the tank would lean it out. I'll find out what the actual numbers were and that might give us some idea as to how much E85 in the tank would be needed to "fake" tune the carb. Or do you not think that's likely to work? Regardless, I'll get actual numbers so I know for sure what he's up against.

Oh yeah, I do remember him saying he thought the carb was a double-pumper. That would be Holly right?

...What ignition system is on it now?...

What do I look for? I'm Chevy-dumb. I know when I see an HEI dizzy, but other than that, I don't know what I'm looking for.

If there's a chance we can get it to pass without going to all that trouble, then that would obviously be better.

As I said, I haven't actually seen the Jeep myself. I only saw some pictures he was showing me of it on his phone. Nothing detailed...

[cgrey8]

I got his numbers. And it was HC and CO that were failing him.

Code: Select all

         25/25           50/15
     Actual  Limit   Actual  Limit
HC   192     126     208     129
CO   5.94    1.38    8.45    1.08
NOx  286     778     169     850

So yeah, what's the likeliness we can set the dizzy down and dump a 2 gallons of E85 in the tank?

If that's all it'll take, how do you fix timing on a Chevy dizzy like you do on a Ford when you pull the SPOUT? I'm assuming Chevy has their equivalent of a SPOUT jumper but if it's not computer controlled, why would it have a jumper? So it's gotta be something else that can do to the distributor to fix it at 10 just long enough for us to get a test to pass.

[cgrey8]

No thoughts?

[decipha]

nope no thoughts

without knowing which ignition setup is on it i can't say for timing, if its real old school then you'll have to re-spring/weight it to get the curve you want, you can also remove the vac line to the vac advance if it has one, if all else fails you can twist the dizy back to knock down the timing as a last resort

for the fuel yeah a higher stoich will lean it out, your best bet would be to prob get a carb tuner to dial it in for you so not only will it pass emissions but it'll run better overall, a win-win, if you were closer to new orleans i'd refer you to my guy roger hall, he's by far the most badazz carb tuner in the united states, if you find out exactly what carbs on it how it jetted / what metering blocks in it etc... i can run it by him to see what he says

[cgrey8]

It won't be for at least another week. I go on vacation at the end of this week. So it might be next weekend by the time I get over there to actually see the beast for myself.

[cgrey8]

I got a look at the Jeep.

It's got an HEI distributor on it with vacuum advance. Again, I'm not as familiar with SBC but it appears to be an aftermarket distributor. The coil is black, but the cap is red. So I'm guessing it's either Mallory or MSD. Aren't they one in the same today? In that Jeep engine bay, it's fortunately easy to get to so whether it's stock or aftermarket probably makes no difference.

It has typical SBC shorty headers that point straight down in the center. There's plenty of room to weld in a WB bung in either pipe. The annoying finding is both exhaust manifold gaskets leak. When he fires it up, I could see the black smoke puffing into the engine bay on both sides. It's obvious to see right where the leaks via the soot on everything right under the leak points. Someone put really thick multi-layer aluminum exhaust manifold gaskets on it. Are those supposed to be better for preventing leaks or are they known for causing leaks? My guess is the leaks are due to loose bolts as I've suffered header-bolt loosening myself until I installed the bolts with high-temp silicone gasket on the bolt threads AND MrGasket header bolt lock washers. Regardless, the leaks have to get fixed before a WB is capable of delivering accurate readings.

The carb isn't one I'm familiar with. It is a Demon but beyond that, I don't know what else to say about it. I didn't take the filter off to get a more detailed look at it (didn't have time and wasn't looking to get dirty right then). Google-searching info of what others think about them, the reviews aren't great. In particular, there are complaints about quality checking and difficulty in tuning them. Hopefully we can get it "close enough" with a combination of E85-dilution and retarding timing.

While looking around under the Jeep, I did notice one of the transmission mount bolts was missing. It's unknown whether it was missed when they last had to drop the tranny or if it's come out on its own. Anyway, I let him know he might not want to run it hard until he gets a replacement bolt in there.

[85GT]

Base on those numbers, if the CO is with cats on, then it's just running rich. Are they 3way? if so bouncing around stoich lets them work. 2ways would need an air pump to supplement. You can jet is down a couple of sizes if you want to get old school dirty. Distributor can be just rotated. Is there vacuum at the canister at idle? Should be going to ported vacuum. With no vacuum, 6-10degs would be typical advance at idle.

If you want to go the tbi route, i can tell you all about the one in my 91 caprice. Diagrams, tuning, etc. Not the most popular pcm, but tunable if you get the hardware. That would be the case for all of them. Need to be socketed and chips burned. It's VE table tuning. A matched pcm to throttle body (injector size) will get you a head start. There are formulas how to adjust those parameters, IF you know what size injectors you have.

[cgrey8]

Base on those numbers, if the CO is with cats on, then it's just running rich. Are they 3way? if so bouncing around stoich lets them work. 2ways would need an air pump to supplement...

I don't know what they put on and I doubt the owner does either. Hopefully they put 3 ways on knowing there's no smog pump. But there's no telling.

...You can jet is down a couple of sizes if you want to get old school dirty...

That's what needs to be done. I'll need to get more details on the carb to know what kind of tuning hardware is needed. Some people say Demon carbs are knock-offs of Holleys with just enough changed to be "different." I don't know how true that is and being I've never tuned a carb before, I don't know what I'd be obligating myself to by offering that as an option. The owner isn't so much for working on vehicles...just driving them.

...Is there vacuum at the canister at idle?...

Canister? As in emissions canister? Pretty sure that was ditched. If you are referring to the vacuum advance can thingie on the distributor, I don't know where its tube goes to. But when I get a closer look at the setup, I'll do a more thorough verification of what's connected where. My lookover of it was superficial and was literally only about 5 minutes.

...If you want to go the tbi route, i can tell you all about the one in my 91 caprice. Diagrams, tuning, etc. Not the most popular pcm, but tunable if you get the hardware. That would be the case for all of them. Need to be socketed and chips burned. It's VE table tuning. A matched pcm to throttle body (injector size) will get you a head start. There are formulas how to adjust those parameters, IF you know what size injectors you have.

I'm hoping we don't have to go that route at this point. I'm really crossing my fingers on the E85/spark retard hack to get it to pass. But if that effort fails, TBI is going to be easier than TPI. I'm fine with the spark remaining mechanical/vac instead of computer controlled. So all we really need is Speed-Density computer fuel control capable of Closed Loop operation. Would MegaSquirt be cheaper and/or easier to work with than tooling up to tune a stock GM EEC? Or are there even cheaper options for something as simple as SD TBI control?

[85GT]

From my understanding, 3 ways still need oxygen introduced to work properly. But the small amount available when a mixture is lean between switches is enough for that. I wonder how effective they are if it's always rich.

I've never had a demon but do believe most parts are 'holley' design. Especially the jets. There would be two screw-in jets in the front metering block and two in the back or a metering plate. In the pics I see, looks like the fuel bowls at each end cover these blocks. Changing jets requires removing the bowls to access the jets. There is no 'tuning hardware' per se. It's wrenches and screwdrivers and the jets. Only aid would be a WB these days. Other contributor to main fuel flow would be the power valve that allows more fuel when vacuum drops. It screws in between the jets. Think if it as the descriminator between 'closed loop and open loop' fueling. These go by vacuum level where they open. If he has a 'wild' cam, that could be causing it to open too soon and go rich too early. There are also more fine tweaks that can be done like accelerator pump cams and squirters, Idle bleeds, etc to optimize driveablity.

Yes, meant the timing advance canister on the distributor.

Don't know if megasquirt would be a cheaper route or not. Besides doaner TBI parts, you can get the tuning hardware from Moates. Between the burner, aldu cable for logging, socket and flash prom, I think you'd be around 200 bucks. And a little soldering to replace existing chip with adapter/socket for the newer memory devices out there. TunerPro RT would be software. Free to try and use. Option to buy to avoid the start up delay.

[cgrey8]

From my understanding, 3 ways still need oxygen introduced to work properly. But the small amount available when a mixture is lean between switches is enough for that. I wonder how effective they are if it's always rich...

Effective, just less effective. I did some research on emissions and CATs a while back and I was surprised by what I found.



Before the CATs, the HC and CO and NOx content is quite high. And while you can see the trend of HC and CO drop as the mix goes lean, they are still present at stoich (1.00 lambda). And ditto about NOx as the mix goes rich.

As you can see, if you could maintain an engine dead at stoich (0.997-1.00 lambda), the CAT can operate optimally without switching. However holding exhaust lambda in the 0.99-1.00 isn't easy. But what the graphs don't show is the CAT effectiveness graph is not static with time. The CATs have the ability to store a small amount of O2 when the mix is rich and CO when the mix is lean. So for short periods of time when the mix is just outside of this range, the CAT is still effective. But once the mix is sustained rich or lean, the CAT's effectiveness starts looking like the lower chart. So if the EEC can keep the mix constantly switching in the .97-1.02 range, you still get effective emission reduction from the CATs.

The other thing worth note about these graphs is this is on a "properly" tuned engine. From what I understand, if you retard the timing, the NOx goes way down even if the mix is lean. And according to decipha, retarded timing also reduces HC too which confuses me, but I'll take it. So as it relates to getting this engine to pass emissions, my plan is to mix the fuel as close to stoich as I can get it and retard the timing which will kill power and hopefully force the throttle into a higher range where fuel demand is a bit easier to control than just off Closed Throttle where there are more dynamics in play in the carb.

...I've never had a demon but do believe most parts are 'holley' design. Especially the jets. There would be two screw-in jets in the front metering block and two in the back or a metering plate. In the pics I see, looks like the fuel bowls at each end cover these blocks. Changing jets requires removing the bowls to access the jets. There is no 'tuning hardware' per se. It's wrenches and screwdrivers and the jets...

Without knowledge of this carb, I was envisioning the same sophistication of Carter style carbs which have jets, metering rods, and springs to tune them. If it's only jets, do you buy a kit of jets or do you need to know the jet size you need and order that? Also where do you get Holley jets? AutoZone or do I have to go somewhere more special like Summit?

...If he has a 'wild' cam, that could be causing it to open too soon and go rich too early...

It didn't sound excessively aggressive or lopey. It had a slightly high idle, but it did seem to run pretty smooth.

If I do need to go the TBI route, I'll ask more questions about that later. But good to know it's not going to break the bank if that's the route we have to go to get it to pass.

[85GT]

Very good info. I did similar research aways back too, just didn't recall all the details. Might be a good addition to the FAQ section. High HC was always an indicator of incomplete combustion. From either poor spark or excessive fuel. And, CO tracks excessive fuel too, but not poor spark from what I remember. Have to think about how retarding it helps lower... hmm. I think leaning it out will get you there. Probably the e85 would do it.

Yeah, similar sophistication. Just different. Holleys don't use metering rods like carters or quadrajets. The metering rod function is replaced more or less with the power valve. That and air flow through the venturi/booster pull the required fuel in. There are springs to control the opening of the rear two barrels if it's a vacuum secondary carb. Other flavor is the double pumper. It has mechanical secondaries and an additional accelerator pump and squirter. These double pumpers are regarded as less emission friendly. Probably because of that extra shot of fuel. If he has one of those, slapping on a universal model 1850, (600 cfm, vac sec) may get you closer quicker. These use to be like 80 dollar carbs lonnnng time ago. One time I looked, the seemed insanely more expensive like everything else i guess.

Local speedshops, do they even still exist? lol. Would sell jets by the pairs on skinned cardboard cards. A 600 cfm ish holley would run someing in the 65 size jets. A 750/780 would have like 72-73's if I remember right. The cheapest route, way back, was Chrysler's Direct Connection. They offered a plastic parts box with an step assortment. It had something like sizes 60 up to 80ish. Think in single number steps. It was WAY cheaper the the two packs. Those use to be like 5-7 bucks a pair. While the whole assortment pack was like 30.
I picked one up for tuning my 70 and the 85 when it was still carburated. The 83-85 mustang holleys actually had 'close tolerance' jets for emissions. They added a third digit to the size to indicate where in the whole number range is was. 632 rings a bell from the GT. Autozone surprises me at times. I'd see if they had either, then go back once you know what sizes are already in the carb. Else see what summit may have as a similar assortment. Futzing with mail order and two at time might be too time consuming.

Also, throw a vacuum gauge on there if you can. That'll tell you if the power valve may be opening too soon. They're rated in inches of vacuum for their opening point. Easy enough to see with a gauge where it should open. Definitely below your idle and cruise values. 6.5 or 11.5 inches was a typical out of the box valve. Forget which one. Might have been the higher and the lower was the one most recommended. Is it an IM240 test on rollers or similar? Maybe throw in like 3.5 just to keep it out of enrichment for most of the test since looks like you have headroom on the NOx

[85GT]

Have to correct myself. Looks like two flavors of demon carbs. There is the Street version that looks carter/AFB based. You might be dealing with metering rods after all.