Technical Air bubbles flowing in return fuel line - 650 Fiat 126 engine in 500. IMB28

[Bianca500]

Hi folks - firstly thank you to all the members here. The contributed forum content, from technical walkthroughs to guidance, is excellent - it's been a fantastic help.

I've got a '66 F, and have recently installed a 650cc 126p engine & box. It's got the 'modern' ecu and crank sensor timing system.

I am looking for some advice regarding it's current running situation. Air bubbles flowing quite fast through the fuel return loop, which whilst I've sort of got it idling, applying any throttle it will try and cut out. Presumably because the air:fuel mixture is being affected by the unwanted bubbles?

I see a fair number of folks on here that have done the swap followed the same approach for the IMB28 of having a T-piece in the engine bay, which I've also done. I recall other comments suggesting no bubbles on the push side is the goal.

From what it looks like, the bubbles are being introduced at the carb. Since the outlet of the Carb return is now on the suction side of the pump.

I've replaced all the jets, float device (set per the manual), and given it a light clean, so I'm not ruling out something I've inadvertently done in the process. I've double checked all the fuel hose clamps for tightness, which are all good.

Any thoughts what would be causing these bubbles? Caution: loud video.

Cheers!

 

[bugsymike]

It does sound like fuel starvation/blockage holding it back to me.
Unless you have a serious air leak.
Whilst it is not unusual to see a little air in a petrol fuel filter that does seem excessive.
I did suggest as a test on another thread to try a gravity fed supply to prove/disprove fuel / pump supply issues, though I do say only if it can be done SAFELY and whilst stationary. Maybe with a fire extinguisher handy.
I can hear a "spitting" noise, is that your stainless steel exhaust or do you have a head gasket sealing problem also?
Just a further thought, you say it has a "'modern' ecu and crank sensor timing system" have you heard the engine running well with this setup as in has this current issue just occurred or is it part of a new build, so could there be a problem there?
Not always easy to locate a sound like that, I sometimes find a damp cloth held around a suspect area may quieten a sound helping you to pin point it.
Don't leave cloth there for long as you may need the fire extinguisher for that too.

 

[Thomas Vaughan]

I have almost the same problem. Air bubbles from the pump. I've just installed a new pump and it's exactly the same. My idle/starvation/lumpiness is different from yours but I'll be following this thread for any further ideas.
If you do directly feed the carburettor, please show us how. I had difficulty assembling a good way of doing this.
My carb is out at the moment after a good ultrasound bath, so can't test anything at the moment.

 

[bugsymike]

I have almost the same problem. Air bubbles from the pump. I've just installed a new pump and it's exactly the same. My idle/starvation/lumpiness is different from yours but I'll be following this thread for any further ideas.
If you do directly feed the carburettor, please show us how. I had difficulty assembling a good way of doing this.
My carb is out at the moment after a good ultrasound bath, so can't test anything at the moment.

Something along the lines of this. Straight to carb. Just be careful, petrol I believe is inflammable.

 

[Bianca500]

Thank you for your responses and suggestions!

I didn't really have a safe way to rig up a direct gravity feed to the carb today, so instead I tried to quickly simulate a proper return feed by using a length of hose down to a jerry can from the return feed of the carb. I temporarily blocked the T-piece section off.

This immediately fired up and there were no bubbles in the flow - interestingly neither the supply OR the return. (this leads me to suspect they were the remaining air stuck in the closed loop?)

However - it's still suffering from some kind of fuel starvation under throttle, so maybe the bubbles were a red herring?

I still need to rule out the pump - but the flow to the jerry can did seem more than what should be 'consumed'. I'll try gravity feed next just to be sure, though.



Re the ecu etc. I've not heard the setup running. Being a later Polish-built engine, it was setup for the nanoplex from the factory. The engine I bought didn't come with it (or an exhaust) so I sourced one separately (APE red), and built the loom up. I've not heard what it "should" sound like so I appreciate its difficult to compare.

I believe the spitting sound is coming from the exhaust, though. Given this is a completely new setup, there are a lot of variables and uncertainties here!

 

[bugsymike]

Oh well, or you can do is work through one issue at a time.

 

[lazylobster]

Looks like you're missing the two rubber grommets that fit around the spark plugs. These are a must as help the cooling of the engine.

 

[Bianca500]

I was able to hook up a 'Motorcycle portable fuel tank' (£25 from Amazon) to test direct feed into the carb, and eliminate the fuel supply/pump from the equation.

The return feed just dumps into a jerry can as it was a short test.

The fuel starvation is still the same, so that does suggest something is up with the carb.

In terms of settings, the float is at the correct 8mm / 16mm adjustment, and the idle screw is 2 complete turns out from fully in - altbough I have tried other adjustments here to no avail.


However, this test I could hear the 'spitting' you described, a bit more frequently. Interestingly it felt like it was coming through the air intake at times.
I will do a compression test on the cylinders. Looking at other posts, 120-140 psi seems to be the target, and low variance between cylinders? Could be that the valves aren't sealing properly?


Aside from this, my next logical step with the carb would be to swap it out with an IMB26 that I have from the original 500 engine. Whilst this would certainly point the blame at the IMB28 if I could get it running, I wonder if that is worthwhile use of time at this stage?
With the IMB28, my understanding is there are two jets; idle and main.

In this situation, if the idle is running (sort of) okay, could this suggest an issue with the main jet? I've cleaned everything out, and also used a new replacement. Could it be a jet sizing issue? I need to get a (larger??) bore main jet to allow more fuel in?

Also - thanks for the reminder about the spark plug grommets - I've added them in. There are a few bits I need to do once it's running like tidy up the cabling wrap / hose routing etc.

The air intake elbow had a hole drilled in the top for some reason... I've plugged it for now but need to get a replacement.

 

[Goldnrust]

I think a compression test is probably a good idea if you have a tester to hand. It doesn't take long (especially with only 2 cylinders to test!) and will help rule out some issues.

It is indeed possible that the valves aren't sealing properly, theres two potential ways for that to happen. One is that the valve clearances have closed up and the rockers are holding the valves open, this is an easy and free fix to just re-set the clearances to factory standard. The second option is a damaged valve seat or damaged valve, and sadly that means removing the head to inspect further and make repairs.

If either problem were the case, you'd expect to see low compression numbers on your test, but as the valve clearance are a 5 min job to check, I'd definitely do that anyway.

Interestingly you are running a separate air filter, we were only discussing recently in another thread how restrictions in the standard intake may effect the air fuel mixture of the carb. Do you have a standard air box lid and filter you could throw on just to do a quick test?

Finally, you said you cleaned out the carb, did you remove the emulsion jet? Its a bit deeper inside the carb assembly than 'a light clean' might get to. Mine was plugged solid when I rebuilt my carb.

 

[bugsymike]

I was able to hook up a 'Motorcycle portable fuel tank' (£25 from Amazon) to test direct feed into the carb, and eliminate the fuel supply/pump from the equation.
View attachment 428376
The return feed just dumps into a jerry can as it was a short test.

The fuel starvation is still the same, so that does suggest something is up with the carb.

In terms of settings, the float is at the correct 8mm / 16mm adjustment, and the idle screw is 2 complete turns out from fully in - altbough I have tried other adjustments here to no avail.


However, this test I could hear the 'spitting' you described, a bit more frequently. Interestingly it felt like it was coming through the air intake at times.
I will do a compression test on the cylinders. Looking at other posts, 120-140 psi seems to be the target, and low variance between cylinders? Could be that the valves aren't sealing properly?


Aside from this, my next logical step with the carb would be to swap it out with an IMB26 that I have from the original 500 engine. Whilst this would certainly point the blame at the IMB28 if I could get it running, I wonder if that is worthwhile use of time at this stage?
With the IMB28, my understanding is there are two jets; idle and main.

In this situation, if the idle is running (sort of) okay, could this suggest an issue with the main jet? I've cleaned everything out, and also used a new replacement. Could it be a jet sizing issue? I need to get a (larger??) bore main jet to allow more fuel in?

View attachment 428377

Also - thanks for the reminder about the spark plug grommets - I've added them in. There are a few bits I need to do once it's running like tidy up the cabling wrap / hose routing etc.

The air intake elbow had a hole drilled in the top for some reason... I've plugged it for now but need to get a replacement.

Apart from keeping the fuel cooler was there any other benefit having a fuel return as it wasn't there on the original set up?
If the spitting is coming from the carb, I would first check valve clearances are correct and if so personally I would do a "leak test" which would show up any poor sealing valves etc. and has the benefit over a compression test of pin pointing exactly which area to look.Although both have benefits.
I am a great fan of this on petrol and diesel vehicles although it can be time consuming.
As a test the 500 carb if OK before on other engine should run to see if any difference. Many years ago I ran a 1500 on a 997cc carb quite happily.

 

[smart51]

I will do a compression test on the cylinders. Looking at other posts, 120-140 psi seems to be the target, and low variance between cylinders?

The theory says a compression test should show atmospheric pressure x times by compression ratio, less a bit for the cylinders not quite filling fully.

14.7 PSI * 7.1 CR = 104 PSI, or 14.7 * 7.5 = 110 PSI depending on your engine.

I recently tested my 594cc engine (mistakenly with the throttle closed) and got 100 and 102 PSI. That was good enough for me not to retest with the throttle open as anything upwards from there is OK with me.

 

[Bianca500]

Thanks all


Firstly, the next set of tests took place with a simplified fuel line in the hope of removing variables. Hose straight from copper to pump. No filter. No T-piece. The IMB28's return line I capped off with a short section of hose and plugged it.
Why did I have it in the first place? I appreciate the nozzle on the return feed is tiny, but after reading the forum around the 650 conversion, there are some folks who have blocked it off, and some who have done the return either to the T-piece, or to the tank itself. The reasoning was to do with issues when hot starting. Without the return, there are reports of fuel vapourising. Given I had a few instances last summer with the 499, waiting for the car to cool down to start it, it seemed like a sensible idea to try the return feed loop. But for now, keeping things simple and blocked it off.


I tried the original air box filter and lid as suggested - didn't have any effect to the fuel starvation. I wonder if it might have more effect on an otherwise successfully running engine? For context, I did see how it behaved with just the elbow, no filter, and variably restricting airflow with my hand, but it didn't help starvation under revs.

I did a compression test and got the following results (in PSI)
Throttle open: 110, 111
Throttle closed: 105, 106

So that suggests it's bob on with what you mention smart51 - that's positive.

I hadn't taken the emulsion jet out when I cleaned it before, as It was quite firmly in there and I was being over cautious about breaking anything. However this weekend I persisted and got it out. It didn't appear blocked, but it was a bit grubby on the outside. Cleaned it up and reinserted it. I did notice that the main line of holes were not aligned with the exit path to the throat. So, I aligned them when I put it back in. Not sure if that makes a difference, but it seemed logical to at least try.
Whilst the rocker cover was off, I also checked the valve clearences cold the next morning. A couple were 0.05 out, so not huge but I corrected them.

In summary, the most notable change to this point is the extraction and cleaning of the emulsion jet. The fuel starvation under load is now resolved!! I was able to drive it around some flat roads locally. What a difference the 650 makes!

I could not see any bubbles at this point either - problem solved?

Taking it out again a few hours later, it was misfiring. Still drivable, and would cope with full throttle, but it was underpowered. The bubbles in the fuel line have returned, notably visible BEFORE the pump.

My next action is to investigate the fuel line between the tank and the fuel pump. Air is getting in somewhere.

I'm confident that if I hooked up the gravity fed fuel tank again, it would run very well. The emulsion jet was likely a separate problem that is now cured along the way!


So a bit around-the-houses, but getting there thanks to your suggestions. Appreciate the responses all.

 

[smart51]

How did you get the emulsion jet out? Its a part of my carb I haven't cleaned.

 

[Thomas Vaughan]

Bra beskrivning på hur man rengör förgasaren.
http://rezerwa126p.pl/?page_id=21759

 

[Bianca500]

How did you get the emulsion jet out? Its a part of my carb I haven't cleaned.

In the end I used a mini pick set, the straight (tapered) one, lightly tapped it with a rubber mallet to get it to grip the jet, and pulled it out. It came out attached to the pick, which I then twisted off to remove it.

It was pretty firmly stuck in there.

Like the bottom one in this set

 

[Goldnrust]

Good to hear of some progress there.

Hopefully if you can find, and fix, where the air is entering the fuel line, then you'll be all sorted

 

[bugsymike]

How did you get the emulsion jet out? Its a part of my carb I haven't cleaned.

On the twin choke carbs on Ladas (Fiat 124 copy) in the 1970s I used an match stick jammed in to pull them out and they were often filled with dirt and water droplets so well worth checking with any fuel related poor running.

 

[Toshi 975]

I give them a squirt of WD or carb cleaner and normally my little phillips screwdriver is a tight fit on the emulsion tube which I give a good wriggle to loosen it up. Had to use a bit from my tiny set recently which locked in on a tight fit so I could pull it out.

How did you get the emulsion jet out? It’s a part of my carb I haven't cleaned.

 

[Bianca500]

Been a little while and various incremental improvements.. thanks all for your suggestions.

There was indeed a small crack in the fuel hose, under the clamp where it fits the tank/fuel sender. There wasn't really enough slack in the original fuel line so ended up replacing the entire fuel line from tank to carb, including bending and bubble-flaring a length of new copper for the rear section for good measure.

No more bubbles!

It also seems I had the coil terminals inversely wired which I'm sure wasn't helping the situation!! This rectified, plus a resistor ballast wired in line on the ingition side should now be correct.
(I'm using the APE 20FF electronic ignition as shown here: https://www.fiatforum.com/threads/installing-electronik-ignition-kit.486992/#post-4633981 )

One thing that isn't apparent, is the difference between the two coil types. Is it just water-resistant connection plug? Or there is some practical difference with spark output?

What I have: (per recommended Fiat 126 twin coil)

What's also shown used: (per linked thread)

Now I'm at the point where it is running and driving, albeit not that well. I've set the timing with a strobe gun to 18 degrees at about 3k rpm, using the adjustment on the crank sensor, and been able to drive it around, which is positive.

Can anyone think of a reason why the power loss behaviour might now be intermittent? With all the above done, I've been out a few times in last couple days where, first run (after letting it choke idle up to temp) it's still lower powered / slight misfire(?), yet I park up, come back 20 mins later and go up the road it's full of beans and sounds great. Yet park up again after a short journey, 5 mins later it's back to the same loss of power.

I can't think of anything that might cause the intermittent behaviour other than some kind of temperature change somewhere?

 

[Toshi 975]

That first coil you show was standard fitting on the 126 from the mid 80’s onwards and needs to be fitted along with a ballast resistor.
The second smaller coil was fitted in pairs to many Fiat models including the Cinquecento, Seicento and Punto. some dealers now sell them as suitable for the 126 and I have known some guys running them but you need to work out your own way of mounting plus you need that special connector plug. I have tested the later type coil myself and have some of them along with the correct plugs. Also have some of the original type coils.