3.0 SC/ 40 IDA Dyno Tuning
 

I had my 3.0L engine again on the MAHA rolling road last week.

The throttle response and the overall performance is now quite satisfying. As you can see from the dyno chart (see only the red line!) the engine has a small 'dent' between 4,300 and 4,600 RPM, which we couldn't eliminate during the dyno session.

We came out at @248 HP at the flywheel at 6,700 RPM with power starting to drop of at 6,900 RPM. Torque figures were at @214 ft-lbs at 4,800 RPM.

The engine is an early 'big' port 3.0L SC with 95mm JE pistons with 10,5:1 (ROZ 100/102 gas is available here at German gas stations), Bosch HKZ single ignition timing set at @30 degrees total, DC60 cams (John Dougherty), 40IDAs (34mm venturis, F3 emulsion tubes, 150 main jets, 170 air correction jets, 60 idle jets, 50 pump jets). The exhaust system consists of Dansk 1 5/8" headers with Dansk dual pipe 'barrow type" exhaust.

We desperately want to eliminate the dent in the curve between 4,300 and 4,600 RPM - what alternative jetting do you think we should try? Any further ideas are appreciated...

Many thanks for your help.
Christian

http://forums.pelicanparts.com/uploa...1283452952.jpg

46 Webers

Here are some suggestions in no particular order:

1) Try some F7 emulsion tubes, a little fatter on the lower RPM's than the F3s.

2) Check the fit of the auxiliary venturis to the main throttle bodies. I resurface the fuel interface surface on the aux venturi making it flat again; an imperfect seal here (it is metal-to-metal) will cause lost signal to the main circuit.

3) Check if you have the tall aux venturis that they don't rock in their mounts. If they rock (side-to-side) they will not seal properly. You can peen the edges of the rectangular end plates to help re-establish a snug fit and then flatten the fuel interface per the above.

4) Be sure the spring in the aux venturi provides a tight fuel interface

5) Check your exhaust gas for rich/lean at the RPMs in question to provide insight to the issue.

By the way, I run 34mm venturis in my 2.3 using 42mm throttle bores and am VERY happy with it, you might want to try 36mm venturis...

Paul Abbott
Performance Oriented

Paul,

Those are some fantastic tips for anyone setting up a pair of webers.

Thank you for sharing.

Jamie,

Hopefully the suggestions help;

"Sometimes the magic works and sometimes it doesn't." (from Little Big Man)

Paul Abbott

Those are great number for a Maha dyno. I love those types of dynos. INcredible amounts of information

Many thanks so far, especially to Paul. :)

We have tried 36mm venturis with the 40IDAs but had some problems here with throttle response especially when running partial loads. In addition we only gained 3-4HP at maximum RPM.

We will of course consider your recommendations concerning the emulsion tubes and the aux venturis.

Best,
Christian

That 4500 RPM dip is mostly due to camshaft profile.

One can sometimes compensate a little with exhaust changes and advancing the cams a little bit, but I would recommend trying different exhaust components, first.

I had a similar issue with my carbs. I found that by adding a 1 inch spacer between the heads and intake manifold the power came in about 500 RPM earlier. You need a fuel/air measurement to see if you are lean or rich at the lower RPM as well. I had a local machine shop make the spacers. I've been told by a well respected engine builder that for our 3 liter engines, a longer intake tract will help with low end torque and not lose the high end. That being said your cams also are not optimum for 4000-4500 RPM.

-Andy

Some more thoughts:

1) Optimize advance for each 1000 RPM and check that your distributor provides something like that.

2) Check fuel float levels on your engine when running and be sure fuel supply pressure is 3.5 psi at the carbs. Each float must be shimmed individually using the external fuel level gauge to check . If fuel levels are not matched then the main circuits will not be activated simultaneously for all cylinders which could provide a power dip.

The DC60 in a 3.0 litre is typically a 4500 rpm and up cam. Looks like the 40 Webers are holding you back above 7000 rpm but giving you a little below the intended rpm range. Are you logging the Air-fuel ratio?

Take a look at this dyno graph
http://forums.pelicanparts.com/porsche-911-technical-forum/246563-another-rsr-clone-project-dyno-day.html

Thanks Steve, Paul and John.

I know the 40IDAs might be to small for my displacement - but they are what I have at the moment. I am considering to move to MFI with 42mm butterflies and twin ignition next year and not going the 46mm Weber/ PMO route.

During the dyno session we found it strange that going to 36mm venturis made the dip at around 4,5k RPM even larger. As it is right know I do feel quite a lack in acceleration in this RPM range in the car.

Best
Christian

Totally predictable, especially with those cams

JMHO, but its sounds like you should be using DC40's not the 60's to get the powerband you expect.

Or 46 Webers.....

+1 with Steve. Cams with this kind of duration are going to yield numbers way up in the rev range and really nowhere else.

Your results are very good so far. If you convert your numbers and match the dyno graphs with Tom's 3.0 you will see the difference the 40's make over the 46 PMO's.
Drive it like a 911 S. Keep it above 5000 rpm until you get the MFI done.

Steve is right, for a 3500 to 7000 rpm band the DC40 is better.

Hi Christian,

You might be interested in my 42mm conversion for your 40s, more breathing without too large a throttle bore.

Paul Abbott
Performance Oriented

Hi Paul,

I have had my 40IDAs extensively rebuilt very recently but certainly will consider your services for future needs.

Thanks,
Christian

Christian,

I had a very similar dip/"dent" in my curve, the shop who built my motor tried a number of exhaust combinations (different headers/muffler mods) and was able to straight it out a good bit.

Jimmy

I'd like to add a data point for reference, and maybe get some feedback as well.

My 3.0L single plug Electromotive crankfire 9.8:1 (Mahle) with 46 Webers and ..... uh ..... can't recall the cams at this time, Holcombe equal length headers w/no crossover, was jetted with 55 idles, 160 air correctors, 200 mains, F24 e-tubes, 42mm main venturis, 4.5 aux venturis and typical 50 pump jets.

On the dyno, mine was quite lean at ~ 15.0 and power flattened out at 6000 rpm, peaking at only 201 HP. Power production was disappointing and the bigger concern was the lean condition. Since I was quite high on the main jet scale, I changed out the venturis to 36mm since it appeared the 42's were much too large for my 3.0L. This decision was based on a friend's 3.4L currently using the larger 42mm chokes and smaller mains (180), reference manual (Bob Tomlinson's Weber Tech Manual) venturi size vs. cyl. capacity chart and same friend's former 3.0L used the 36mm venturis with 46 Webers.

My dyno guy (Dynojet) is a drag race & carb specialist and he also indicated that "smaller" carbs are advised in order to pull more fuel since I was already so high in main jet size, yet not getting the rich A/F I need. So with the smaller 36mm chokes, I also changed the e-tubes to F7. Reason being is because I understand the F24 to be for very large mains, which with the smaller chokes I shouldn't need to use?

My Halmeter AF30 doesn't seem to be too reliable because even with a new O2, after a few laps on the track the signal fell back off the chart in the lean zone. I'll get it back on the dyno to confirm where i'm at, but i'm just wondering how I could potentially still be lean at my current configuration (36 chokes, 200 mains, 160 A/C, F7 tubes, 55 idles) compared to what Christian is currently running. My next step would be to take a big swing at the air correctors by going down a few sizes.

Lastly, I found my engine to have a power dip/stumble at 4000-ish whereas it wasn't there before. I'll take the stumble if it means i'm running richer. I just don't want to melt my pistons by running too lean!

What are people's thoughts on the downsizing of the air correctors? Since these are a threefold effect compared to the main jet, smaller air correctors would certainly richen the mixture. But it seems my result would be sort of uncommon based on other people's typical jetting which is smaller mains and larger air correctors.

Thanks,
Kevin

I suspect you will be very rich with the current set-up, but there is no way to tell for sure except by getting an O2 guage or dyno time. You can get a clue by looking at the soot in the exhaust and at the plugs. Get a wide band O2 guage in there. I put one in my race car last year and it has proved very useful in setting up the carbs. You can get them for around 200 dollars now. My cockpit video captures the display so I can see what happened mixture wise during the session.

-Andy

Andy,

I was hoping for rich so we'll see when I take it back to the dyno next week. I did take notice of the exhaust and after the venturi swap the car sounded sort of "fat." We adjusted the mixture screws and it smoothed out a bit but still sounded deep. The tailpipes do show blackening, so I think that's a good sign.

I do have a wideband O2- i've got the old bar graph Halmeter AF30. Funny you mention the video too. Mine's mounted where the radio would be and you can see it on the video. It provides a rather stable reading while at pit speeds and during the warm up lap. Once the race started, the display began to fall backward to lean after just a few turns on the track. Then it showed nothing but a power-on signal.... :confused:

Are you overheating your O2 sensor?

Seems odd that it was reading OK and went away after time. That might mean heat build up and it is affecting the reading. Does it go back to "normal" after a cool down?

Did you do a free air calibration of the O2 sensor? Makes a significant difference in accuracy as it affect both the ability to measure O2 and unburned fuel.

I was thinking overheat of the sensor too. But the old sensor gave a reading all the time- albeit completely inaccurate.

My Halmeter has no calibration feature. Instructions say just plug it in. However knowing that other AF meters do a calibration, I hooked up the sensor and keyed the ignition on to pre-warm it before installing it in the exhaust pipe. I also let it cool off before running it in place.

FWIW, I have been very happy with my AEM wideband sensor not too expensive but is very close to the Dyno sniffers. Very simple, no calibration by the user just plug and play.

5-4-3-2-1..

the Daytona Sensor schill hasn't popped up yet?

anyway...

The Halmeter is a narrow band sensor, not good for much except "yes/no" indications of mixture..

Per the OPs post.. I have to check my notes for what combo I have on my 3.2L 914/6 w/Weber 40's .... through many iterations I have a E-tube/jet /venturi combo that is fat enough under load, and leans out to good power at high rpm..

Also do what hundreds of other people have done regarding that 300 RPM block... Just stick your foot on the loud pedal and drive through it..

Thanks for the tip on the Halmeter. I'll get a wideband to keep a better eye on things. I saw the AEM sensor and what's up with it being noticeably cheaper than other brands? One thing I like about the AEM is it's analog to match my other inaccurate gauges..... :D

Can you elaborate on point 2 there? Never heard that before and don't understand why. As others have said, thanks for the great tips!

There are four fuel floats for the pair of carbs for 911 engines, two per throttle body. The fuel well between cylinders 1 & 2 supplies fuel for the main and progression circuit for cylinder #1 and for the accelerator pump for all three cylinders. The fuel well between cylinders 2 & 3 supplies fuel for the main and progression circuit for cylinders 2 & 3. A similar arrangement is used for cylinders 4 through 6.

The main circuit is activated by enough air being drawn througn the main and auxiliary venturis so that the pressure drop associated with that velocity sucks fuel from the fuel bowl up through the emulsion tube well and out the passage-way that ends into the center of the auxiliary venturi.

If the fuel level is low in one fuel well due to a mal-adjusted float then it will take MORE air velocity than a correctly adjusted fuel level to draw the fuel into the auxiliary venturi. The rusult will be some cylinders transitioning to the main circuit while others are still on the progression circuit which is loss of power or flat spot.

Paul Abbott
Performance Oriented

Paul,

thanks. I run Weber 46's that I've customized a bit myself so I'm quite familiar with their general operation. However for some reason I think I read more into your statement than you made. What I took out of it was that even a very slight variation would cause loss of power, but went back and read your comment and you didn't say anything like that. :)

Mine are adjusted to within 3-4mm of each other but just wondering if a little extra effort would yield some results.

Chris: I have for years connected the float bowls of the individual carbs together with a rubber hose connected to fittings where the drain plugs go in the float bowls. This at least equalizes the float bowls of each individual carb.
aws

aws,

I would think that linking the float bowls together would be the best way to ensure they were almost never at the correct level except when the car is parked. Under acceleration most of the fuel would go to the back bowl and under braking most of the fuel would go to the front bowl. If you've ever seen a video of a brake fluid reservoir when a car is on the race track you'd see why this would happen.

-Andy

My main motivation when connecting the float bowls was to equalize fuel flow to the bowls because one bowl feeds 2 main jets and the other only 1. Truth be told the fuel condition in a race car carb float bowl at full bore is probably more of a froth than a liquid. (A good reason Richard Parr of PMO does not like solid engine/trans mounts) I had never considered the effects of g-forces on the float level--- although I have never had a mixture or drivability issue in years of doing this.

aws

Regarding the 2 jets on one bowl, i've seen a larger needle valve used on that bowl. My carbs are using 200 valves on the 2 jet bowl and regular 175 valves on the single jet bowl.

Hi Chris,

The bowl supplying one main jet also supplies fuel for the accelerator pump so there is a parity of fuel demand from both bowls.

The loss of power due to unequal float heights is hard to quantify, let's just say that unless they are equal then the system won't be optimized. The Weber tolerance on fuel level setting is 20.5mm to 21.0mm (0.020 inches) from the top edge of the carb throttle body at 2.5 meters of water which is 3.56 psi. I routinely see equal shims used for setting float heights but the the floats NEVER have uniform geometry, the little tabs aren't set correctly with the fulcrum pin and the top of the float body. Also, an off-road excursion may bend the brass tabs from the fuel sloshing so it is prudent to check your vehicle after such an event, not only for structural items but float levels too.

Lateral forces due to acceleration loading (lateral and longitudinal) will slosh the fuel in the bowls, theoretically the floats have their center of buoyancy so that regardless of fuel movement the float height isn't adversely affected, sloshing is another matter. Fuel bowl dams help keep the delivery to the mains from being exposed during lateral accelerations.

Paul Abbott
Performance Oriented