That's a fantastic graph to show the transition starting just after 3500 rpm! Is that the slow application of throttle in 3rd gear?

Yep this is the game we play, dialing things in. Shocking what a difference the main venturis make! But when you think about it, it makes sense. The whole event of fuel draw starts with the main venturi in that the main venturi establishes the amount of gross airflow (volume and velocity) to which the rest of the carb's jets react.

The carburetor just amazes me that it works as well as it does, simply on the premise of air movement drawing fuel out of the carb.

Great pictures by the way. They show people a couple of PMO "physical' improvements

1. The secondary venturi is sleeved into the throttle body for a very secure, reliable long-term fit. Webers on the other hand have the venturi slotted into the throttle body and are secured with a piece of spring steel. Over time that spring loses its tension and the venturi can wobble in the slot, which eventually hogs out the slot and that permanently loosens the fit. A loose venturi = poor contact with the fuel supply port within the slot and you lose the venturi effect of the air flowing thru the venturi tube, which sucks fuel out of the notch. You end up with a lean mixture condition and weakened low rpm power.

I experienced this first hand at the racetrack when my buddies Craig and Denny were talking shop with me about carbs and asked how well the tall venturis fit in mine, since a close mutual friend had a problem with them being loose. I said I never thought much about them (ignorance, not dismissing the idea) and they said let's take a look. One carb had all three of them really loose, so we removed both carbs and shimmed all of them with some feeler gauge cut to size. Holy cow what a difference. The difference in low rpm power was a massive improvement when applying heavy throttle exiting slow corners. I was shocked.

2. The PMO main venturis/chokes are slip fitted into the throttle body and held secure by the secondary venturi sleeve. Nice secure fit with no external fastener bearing on it. Webers on the other hand are held in the throttle body by a shallow, blind hole drilled in the side of the venturi and there's an external screw outside the carb that holds the venturi in place. The screw is special in that it has a machined snout on it to engage the blind hole in the venturi, and you safety wire the screw to the exterior of the carb. The downside to this fastening is that you can easily overtighten the screw and distort the shape of the venturi and allow air to escape past it.

Here's a picture of where the Weber screw is on the outside. It's the hex head screw with the twisted safety wire on it, bottom right of the right-most red circle.

http://forums.pelicanparts.com/uploa...1435595588.jpg

And by the way, that screw isn't safety wired very well- has the opportunity to loosen a bit before the wire goes taut. Granted, we're talking about a screw that has a very low torque and I would bet the stiffness of the "loose" wire is enough to keep it tight

Thumbs up to Gordo for always providing detailed descriptions & pictures of what he's doing! SmileWavy

Emulsion Tubes - AFR Curve
 

Thanks as always for the great feedback KTL! Yep - 3rd gear, slow throttle application/acceleration up a moderately inclined stretch of road.

This latest graph/run really got me thinking about my fuel curve and how the emulsion tubes play a role in this.

Based on the graph, my F16 emulsion tubes really start drawing/providing fuel from ~3,900 RPM on up. Nearing this same point (3.5K-3.8k RPM) my 55 idle jets (which are stinking rich for most of the progression circuit RPM range) - have reached their limit for fuel mixture flow - and the engine begins to lean out (mid 14 to 15 AFR) until the mains really start to contribute.

I'm thinking if I put some emulsion tubes in that begin to feed at a lower RPM range (i.e. start to draw/feed from the main jets around 3,300 RPM), I may be able to smooth out the transition a little better.

I'm considering trying the F8 emulsion tubes.

Thanks again, Gordo

No problem. I'm always willing to listen about carb tuning because it seems almost every application experiences something different.

What makes you think the F8 is the emulsion tube to test? I'm not disagreeing. Just wondering what your thoughts are on that tube specifically.

PMO makes mention of four sizes. F11, F16, F2 and F7. He says F11 seem to work best for street cars and mild cams. http://www.pmocarb.com/Images/Instructions-3.pdf

Emulsion Tube Selection
 

Kevin - considering the F8 entirely based on the Weber emulsion tube table and descriptions of their effect (center, IDF column):

http://forums.pelicanparts.com/uploa...1435706658.jpg

I currently have the F16's - which by the Weber description, essentially delays when the mains start to contribute. This appears to match what I'm seeing - my mains come on strong rather late ~ 3.9K RPM.

The F7 description says it's designed to contribute earlier in the RPM range. Meanwhile if I understand correctly, the F7 and F8 are essentially same, but the F7 is sized for larger displacement engines.

I'm thinking if I can bring in the mains earlier, it will reduce the lean hesitation at transition, while allowing me to adjust and reduce my idle/progression circuit richness (which when set up on the rich side - tends to mask the transition leaning).

That's my current theory, not sure what its worth though.

As a side note - I'm going to throw things for a loop soon. Just bought some SSI's to replace my 1.75 B&B headers....

Gordo

OK, I'm in. Great discussion thus far.

The F16 has 8 x 1.00mm dia holes near the top and the F8 has 2 x 1.00mm dia holes a little higher than the F16. The F8 will have a little later initiation onto the main than the F16. F8 also has holes along its height that the F16 does not have which will lean the mixture as RPM increases.

F7 is same as F8 except it does not have the 2x1.00mm holes at the top which would make it initiate quicker than the F8. You could try the F8s and if you think an earlier initiation would be good then solder the top two holes shut and you will have a F7.

The comparative chart says F7 & F8 are for "slight accelerations", this means when cruising you open throttles by 25% and your mixture is weak then a F7 or F8 would help. The smaller outside diameter of the F7/F8 e-tube when compared to the F16 provides more fuel within the annulus of the E-tube well and the outside diameter of the E-tube. This is referred to as a "reserve" which means fuel has already passed through the main jet and is ready to be sucked into the venturi without delay. The extra holes along the length of the body of the F7/F8 are to keep fuel from gushing into the engine due to the relaxed fit of the E-tube in its well.

I recommend 3/4 throttle runs in 3rd gear and use a throttle stop under the throttle pedal to achieve consistent charting results. 4th gear will be a little too fast for most public roads.

Your main jet size should be somewhere close to 36x4 and add a little for crummy fuel and add a little for better air flow thru PMOs so 36x4=144 + 10 = 155 mains...maybe. Try 160s to be safe.

My two cents.

Emulsion Tube Role and Function - Factors for Selection
 

Mr. Abbott - thanks so much for joining the mix.

I've continued my research on the the F7 and F8 emulsion tubes and e-tube principles in general.

No question - Mr. Abbott has the e-tube theory mastered and memorized (or tattoo'd on his forearm), meanwhile I continued to search hoping to find some AFR recordings that helped visualize how various e-tubes influence the AFR curve. I wasn't able to find recordings - but did find some additional helpful info.

Pelican "Gumba" tried the F7's - but found they made his transition / mid RPM too rich. He was unable to recall the config and details. Also found other forums (non-Porsche) with folks describing similar results trying the F7 - too rich.

Lastly, I found an awesome RX-7 thread ( Weber Emulsion tubes - RX7Club.com ) with a great discussion and description of the role that emulsion tubes play - same principles as Mr. Abbott described, but put another way:

Emulsion Tube Role / Function

"Emulsion tube, so what the hell does it do? The easy answer is that it mixes air with fuel as it enters the aux venturi and is sucked into the engine. But, that's not all it does. Actually that's the smallest part of what the tube is actually doing. The e-tube is actually a fuel brake. It slows the progression of the fuel metered by the main jet. Without the e-tube the car would continue to run richer and richer as the revs rise. See as the air speed increases and pressure drops across the chokes, vacuum in the float bowl increases two fold, so there has to be a way to meter the fuel down to use able amounts. The emulsion tube does this two ways. The first is by the width of the tube. The wider the tube is the tighter it fits against the main circut galley, therefore the less fuel it allows by and the leaner the mixture. The second way it slows the fuel's progression is with the air corrector jet and the holes in the tube. The emulsion has different series of holes along the length of the tube. The higher the holes the LOWER the RPM range effected. When the etube assembly is in the carb it is half sitting in fuel, half not. The depth of the fuel in the main circut is controled by the float setting in the carb. You see the higher the fuel level is in the etube galley the sooner the main circut will come on. Also, the less amount of holes above the fuel the sooner the main circut will come on. Air and fuel will take the easiest path to get to the vaccume created in the engine. If it's easier for air to rush past the ac jet and through the e-tube and into the carb than it is for the fuel to be pulled out of the float bowl that's exactly what is going to happen and that's why the main circut is comming on late.

Now all of us rotards using webers have experianced the annoying 3000 rpm cruise stumble. Now we all understand that the cause for the stumble is a lean spot from the progression from the idle circut onto the main circut. So how do we fix this? First by extending the idle circuts amount of time functioning to supply enough fuel to run smoothly past 3000 rpm when the main circut starts to come on. So by richining up the idle circut we can over come the stumble. This however has it's draw backs. Rich idle causing terrible gas mileage, strong petrol smell, fouled plugs, as well as a rich sluggish bottom end.

The other option is to somehow cause the main circut to come on earlier. There are two ways to make the main circut come on earlier. The first way is by controling the fuel level inside the main circut galley you will control how soon the main circut begins. So how do we raist that level and how much can we raise it before we have problems? The answer is not much. With a weber IDA you have to set the float to between 5.5-6mm. .5mm of varience to work with ("0.0019 incase you don't understand metric.) anything more than that and you'll flood the car anything less and it'll run grossly lean. Now you see how important float level is don't you. The second way is by metering how much brake that etube is putting on the fuel.

First a thinner etube will allow more fuel to pass by. 2nd a smaller air corrector jet will allow less air through the etube braking the fuel less; and third by getting an e-tube with less holes in it or holes in different places will allow more fuel into the aux venturi."

It goes on to describe results of swapping in F7's, F8's and F3's. Although it's all in described in relation to a rotary engine - I think most of the observations and points still apply.

Great feedback and info - meanwhile I'm still thinking of trying a set of F8's, leaving all else constant, just to see if I can get the mains to contribute earlier / at lower RPMs.

I'm obsessed and can't help but play. ;)

Thanks again gents,

Gordo

This thread has great theory discussion with applied experimentation and results. Thanks for providing so much detail. Looking forward (not dreading:) to tuning the PMOs on the 2.8l RSR engine I just acquired. I will try to contribute once I get it going.

cheers,
dug

Experimenting - 36 Chokes, 160 Mains, 170 Air Correctors, 55 Idles, F16 e-tubes
 

I didn't order the emulsion tubes yet but was itching to experiment today. As such I played with the jets that I have on hand.

I swapped/reduced my main jets from 180 down to 160's. Also swapped out the 180 air corrector jets and installed 170's.

I normally make one change at a time and then do a run to record the results - but I needed to reduce the upper RPM richness significantly, and I didn't have a lot of time today.

My main interest today was seeing what kind of influence the smaller air correctors had in relation to the RPM's when the mains kicked in.

Since it's on the previous page - here's last week's run
Last week's recording - 36mm Chokes, 180 Mains, 180 Air Correctors, 55 Idles
http://forums.pelicanparts.com/uploa...1435987434.jpg


Here's today's run - same stretch of road (remote, oops ran up to 6k RPM...), same driving profile (3rd gear, slowly accelerating), relatively same weather conditions:
This week's recording - 36mm Chokes, 160 Mains, 170 Air Correctors, 55 Idles
http://forums.pelicanparts.com/uploa...1435983969.jpg


During last week's recording using the 180 air correctors, the mains appeared to kick in at ~ 3,900 RPM. With the 170 air correctors the mains appear to kick in ~ 3,400 RPM. I don't believe the smaller main jets would have any influence on this shift to the left (but they did lean out the upper RPM's some).

Seems I could continue to reduce the size of the air correctors (to initiate the mains earlier), but at some point as the air correctors get smaller I would expect the upper RPM Air Fuel Ratio (AFR) to get too rich. It might be possible to offset this with smaller main jets, but at some point, I think they would no longer be optimal to feed the entire RPM range.

Additionally, I would ultimately like to lean out my idle/progression & cruise - which will require opening up the idle air corrector jets (50 idle jets go too lean at the progression circuit's upper RPM range). When I do this, the engine is clearly going to lean out earlier in the RPM range.

With the 55 idle jets, the engine's stinking rich at lower RPM's but going lean around 3,200 RPM (before the mains contribute). As such, I think I want to shoot to get the main jets contributing around 2,800 RPM.

I might be able to get there just reducing the main air correctors, but I tend to think I'm going to go to a emulsion tube that is designed to bring in the mains earlier.

That's it for the fun stuff.


Now for the entertainment:
Good Initiative, Poor Judgment - Disconnected Acceleration Pump Linkage

Today, I decided to disconnect my acceleration pump linkage - so I'm not worried about pressing the pedal slow enough to keep them from activating and skewing my AFR readings.

Removed the linkage and... You got it - I couldn't start the engine without a couple shots from the acceleration pumps. Oops, back on they go :rolleyes:.


Have I mentioned - http://forums.pelicanparts.com/off-topic-discussions/631556-i-love-tools.html

I also decided I'm going to start reaming a set of jets to get things configured. I'm going to end up messing with the jets once I get the SSI's and figured it would be cheaper in the long run. Purchased this kit on eBay:

http://forums.pelicanparts.com/uploa...1435985841.jpg


And finally, a note for Mr. Abbott

I considered soldering some of the upper holes on the F16 emulsion tube, but with the 8x100 dinky holes and my soldering skills - I pictured making nothing but a molten mess ;)

Thanks again folks.

Gordo

Hi Gordo,

Great info via your graphs.

I'd put in smaller idle air correctors first followed by a 155 main jet followed by a 160 main air corrector. You will be amazed at how much affect the idle air correctors have on the progression fuel curve. Your 50s will probably be good all the way through transition with a smaller idle air corrector.

Don't worry about the main air corrector getting too small, just make an adjustment, evaluate the response and make a new adjustment. You are getting pretty close now so make one change at a time.

I agree about soldering holes on your F16 e-tube. I hadn't looked at your e-tube design, the e-tubes in the Webers are not as complicated and the eight holes all in a circle would be difficult to solder. However, the F8 e-tubes are another story, easier to solder than the F16.

Keep with your jet changing process until you have gone as far as you can and then consider e-tube swap. You now have the ability to solder holes shut on your jets and then resize them with your reamers, when finished you will want to buy jets with sizes per your final configuration. One little catch, when you solder an orifice closed you will need to pilot drill it open before you can insert the reamer. A pin vise and a couple of small drills are in order, I like McMaster-Carr for such items. I recommend pin vise #8455A31 for $11.28.

Plan of Attack
 

Mr. Abbott,

Thanks for tips and recommendation on how to proceed. I was thinking the same (play with the jets until I either find a combo that works, or determine that I need to change the emulsion tubes).

On the topic reaming - most of the info I found online shows people just using the reamer to start and create the hole as opposed to pre-drilling with micro-drill bits sized slightly smaller than the final desired size.

Is this an acceptable method, or is it better to pre-drill?

Thanks again, Gordo

Reamers do not have the ability to start a hole. They have a sharp point like a needle. They will not ream without a starter hole to begin with. Typical application for reamers is to enlarge a existing, smaller jet orifice.

This forum is making me want to build an EFI setup rather then run these carbs. If you did it all over again, would you do the same???

Carbs vs EFI
 

No doubt - I'm good with the carbs and would do it all over again.

I had the carbs dialed in pretty solid about a month or so after I first fired off my rebuild. I've also driven the car for the past ~9 months and they ran great (did need to clean an idle jet at one point) - plenty powerful, easy start up, rock solid idle etc.

Meanwhile, I should have started down this trail with a new thead titled "I'm a 911 Owner and I Can't Leave My Carbs Well Enough Alone...".

Seriously - if it weren't for the wideband indicating that i might be able to make them even better (and learn a few things along the way), I could be like most other folks with a good set of carbs: fast, happy and oblivious.

As for EFI - I considered the TBI option. Cost was comparable, but in the end I found the idea of mechanical tuning much more attractive than adjusting computer/chip settings. Both require considerable study and experimentation (or money if you prefer to let someone dyno-tune your engine for you).

Bottom line - if I went the TBI route I'm confident I would still be playing around.

Search Pelican for DIY EFI'rs (try Tec3, Haltec, Motec etc.) and you will find similar threads. Believe it or not - dialing in an EFI / engine management system can be considerably more complex.

To me, it's all about the enjoyment of learning something new - adjusting, driving, observing, correcting - optimizing. If I'm not enjoying it - I'm doing it all wrong.

On the other hand if the carbs/engine were running like crap at this point (which seems pretty uncommon on a new set of PMO's - Mr. Parr's pretty good at getting them set up close enough right out of the box) - I would have tossed in the towel, tucked my tail and handed off the car to the pro's a long time ago.

Perspective...

Gordo

Isolating Idle and Progression Circuit
 

Still itching to experiment as I wait for my parts and tools to arrive.

My latest theory (based on my AFR recordings):

In the idle/progression to main jet transition region (~2,800 RPM to ~ 3,500 RPM) the main jets appear to contribute in a relatively binary fashion - off or on (not contributing, or fully contributing).

From what I've read, and from various graphics depicting carb fuel delivery - it would seem that the main circuit contribution tends to taper in - preferably providing sufficient fuel delivery to maintain a relatively steady AFR as the idle circuit reaches its fuel delivery limits and the main circuit begins to overlap and contribute.

Bottom line - I'm curious if I'm getting any contribution from the main circuit below ~3,400 RPM.

If I get the weather and opportunity today - I think I will do a recorded run following another one of "professor" Abbott's tests:

"It is easy to isolate the idle circuit during slow speed testing: remove the main air correction jet and emulsion tubes. The RPM where progression circuit begins to weaken is where the main circuit needs to begin to add fuel to avoid a lean transition. By knowing where the idle circuit begins leaning out you can then reinstall the emulsion tubes and main air correction jets and then monitor mixture strength during initial transition operation. If the main circuit is disabled in this fashion and the vehicle is driven on open roads then the driver must be VERY aware that the vehicle has been stripped of main circuit operation and is thereby limited in its performance." Performance Oriented

Short drive staying in the low RPM range, light throttle applications while keeping a constant eye on the LM 2 AFR... Don't think I would try this without a wideband as I will be confirming the limits of my idle/progression circuit - the point where engine demand exceeds fuel delivery (i.e. leaning out with AFR headed in the 14 range).

Gordo

Resized Idle Air Corrector Jet
 

Weekend fun playing with my PMO carbs...

--------------------------------------------------------
Removed Emulsion Tubes/Mains and Air Correctors

Began by removing my emulsion tubes (complete w/mains etc) for a recorded drive. I wanted to see what the idle/progression circuit AFR profile looked like in isolation.

Here's the recording: driving down a windy, rolling backroad ~ 3,000 - 4,000 RPM, running entirely off of my idle/progression circuit (which has been set up quite rich)
http://forums.pelicanparts.com/uploa...1436714568.jpg


Observations

Although I've read that during light throttle / lower RPM street driving and steady RPM cruise the carbs are primarily running off of the idle/progression circuit, I was still surprised to find that the main circuit appeared to only be drawn upon / contributing under increased load conditions (relatively moderate and greater throttle applications, uphill etc.).

I also found the idle/progression circuit could deliver good AFR #'s (13 to low 14's) all the way up to ~ 5,000 RPM if you go light on the throttle, and are on relatively flat roads. Again - this was with an idle set-up that has been too rich at lower RPM's.

I was expecting the idle circuit to lean out and drop off ~ 4,000 RPM's however it appears to continue to contribute well beyond that and well into the main circuit's territory.

I tried my normal 3rd gear hill climb drive/profile with this configuration - but the engine went too lean right off the bat. The idle/progression circuit could sustain acceptable AFR's during the climb - but only with very slight/gradual acceleration.
--------------------------------------------------------


Resizing Idle Air Correctors

My engine has been running rich from ~1,000 to 3,500 RPM. I normally run 55 idle jets, but I've tried 50's and they still ran too rich. Based on feedback from Mr. Paul Abbott and others - I decided I would try to resize my idle air corrector jets to help lean out my lower RPM AFR's.

Removed the PMO idle air correctors and measured them. Mine were 1.25; Mr. Parr (PMO) told me he used 2 different sizes, but I can't recall what the other size was:
http://forums.pelicanparts.com/uploa...1436715216.jpg


I initially reamed the idle air correctors out to 1.35. I was surprised how much this minor change influenced AFR's at idle (without any mixture screw adjustments). Opening the idle air correctors from 1.25 to 1.35 resulted in my AFR's going from mid 12's to mid 16's at idle (rough idle...) - mixture screw adjustments quickly fixed this (starting back at 2 turns out...).

Short drive showed the 1.35 idle air corrector was still too rich at lower RPM's - as such I didn't bother to record, but opted to head home to open them to ~ 1.45.

It was slow going reaming out the PMO stainless steel idle air correction jets - I assume brass jets are a little easier to ream:
http://forums.pelicanparts.com/uploa...1436715794.jpg

http://forums.pelicanparts.com/uploa...1436715813.jpg


Here's my normal 3rd gear hill climb with the idle air controllers reamed out to ~ 1.45 (graph mislabeled - should be 170 main air correctors)
http://forums.pelicanparts.com/uploa...1436716401.jpg


Nice - this is getting very close to where I want to be.

1.40 Idle Air Corrector Observations

My lower RPM driving (1,000 to 3,000 RPM) and cruise is now in the 12-13 range, and the combo seems to be providing sufficient delivery to support my 3rd gear hill profile (when combined with the main circuit).

My main circuit has been rich since I swapped my 38mm venturi's for the 36's - still have some playing around to do up there...
--------------------------------------------------------


Will play more once I get my new exhaust installed (going from 1.75" B&B system to 1.5" SSI setup).

My bro' (Pelican Cabmando on the left) was in town to help (or at least help drink a few beers ;) ).

http://forums.pelicanparts.com/uploa...1436717237.jpg

It was a great weekend.

- Gordo

Gordo,

When you disconnected the accelerator pump linkage, did you remove it completely? Or just remove the nut from the shaft and let it move freely? I'm about to start playing around with the accelerator pump injection volume for my racecar to address rough running when blipping on downshifts. I want to reduce the volume more that I seem to be able to achieve by just adjusting the nut (I only get it down to a little under .5 cc). Ideally I'm looking for a way to quickly make the change at the track, and then be able to reconnect as well.

Thanks.

Scott

Scott,
You can move the whole area of adjustment up or down by replacing the float bowl control valve. But not easy at the track.

Thanks, I have heard about that option. I'm looking to get a sense of which direction I need to go with (hopefully) quick adjustments before making the float bowl control valve change. I've also heard of reshaping the "hatchet" actuation lever, though that seems tricky and possibly not so easy to get right. And of course more permanent.

Not to hijack, but Richard will give you a redesigned hatchet if you call; not at all permanent but you won't change back after you've done it. So far as I know, fbcv won't do anything about rough running, especially when blipping the throttle. It will address bogging (too much gas) or stumbling (too little) when you get on the gas after shifting.

I do have the updated hatchets. I am not sure whether I have too much or too little fuel when blipping, but since it runs well otherwise, I am looking to experiment by adjusting injection volumes up and down.