Lotus Elan

40 Dcoe Tuning Part 2

PostPost by: type26owner » Fri Apr 09, 2004 3:16 pm

Identified two more issues to resolve now to get it all sorted correctly.

One is the cold start choke unit when engaged halfway causes the mixture to fatten to less then a 10:1 mixture but it does not raise the idle speed. If pulled on fully it will not assist the starting and it will not sustain the running after it does start. Since this unit supplies both additional air and fuel this should not be the correct result. It should rev the engine up to about 2k rpm and help not hinder the cold running performance. Suspect the stock setup is jetted way to fat.

Secondly, I've got a mixture leaning situation at high revs now. None of the tuning books properly address how to correct this condition BTW. By jetting the mains (100 w/Shell 91 fuel) for a mixture of 13:1 at ~3K rpms w/no load by the time it gets to 6k at WOT and at maximum load it's at 17:1. This is not okay and I've already got tiny 140 air corrector jets installed with little affect on this situation. Appears the best adjustment to try first is to install larger main venturis. The stock 30mm ones are undersized according to the graph (pg22, fig32) in the Weber Tuning Manual. This is assuming that the one labeled '4-stroke sport engines' is the correct one to apply for the twincam. By this graph a 36mm size is required which just so happens to be the largest size which fits and is available for the 40DCOE. So what I think is happening is the undersized venturi is generating to much low pressure at to low a rpm. This causes a very fat mixture at the intial rpm point of mixture's introduction. When it reaches the high revs it then is restricting the airflow so much it impedes the auxillary venturi from reacting to a high enough vacuum to keep the mixture constant. By enlarging the main venturi it should reduce the vacuum level and delay it a low revs and at the high end it should allow more air to pass through so according to Bernoulli's principle it could actually increase the vacuum level in theory. Have to test this theory. Just hope it's this simple to solve. Can't believe how lame the advised criteria on which the tuning books use to size the the main venturis. My engineering intuition tells me it's generally true but for a different reason then what's described. :(
Keith
p.s. Hoping at this point the stock exhaust system does not impede the induction system at higher flow rates. Going to attempt to measure the pressure drop at the downstream point the auxilary venturi protrudes into the main venturi at least for the no load condition. It may be possible to establish a correlation between the pressure drop and the mixture strength. These two effects should be slaved together ideally.
type26owner
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 1616
Joined: 18 Sep 2003

PostPost by: gjz30075 » Sat Apr 10, 2004 2:07 am

Keith, 30mm to 36mm choke is a heck of a jump. Wouldn't you be trading off a proper vacuum to getting too much air, hence leaning it out at high revs? Why not go to a, say, 32mm choke to see if your theory is correct?

Greg Z.
Greg Z
45/0243K Sprint
36/5727 pre airflow coupe
User avatar
gjz30075
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 2828
Joined: 12 Sep 2003
Location: Roswell, Georgia, USA

PostPost by: type26owner » Sat Apr 10, 2004 4:07 am

Hi Greg,
I'm aware of that situation and please realize I don't care to throw endless amounts of money at this situation either. Contacted my friend Mike Ostrov earlier this week to borrow some venturis for testing purposes. He's graciously offered me all the sizes he's got socked away as spares. One set is of 33mm which I'm hoping is enough of difference so I tell if it's the right direction of component to change in order to correct this problem.

Geez, a set of these venturis is nearly $100 for each set. Asked Doug at Pierce Manifolds how much a complete complete set of spares of every part offered for the 40DCOE and he guessed it would cost well over $20K to complete one. The real challenge here is to succeed correcting these problems without spending a fortune doing it.
Regards,
Keith
type26owner
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 1616
Joined: 18 Sep 2003

PostPost by: type26owner » Sun Apr 11, 2004 3:43 pm

Been pondering this situation and have realized maybe this condition is real and the typical limitation of a carbie. Trouble is the experts are so vague on this subject that their comments are really quite useless for me to make that determination. All you get is a statement like 'for getting the best drivability one should install this size main venturi'. Hey, wait minute here why is this so? What are the underlying physical laws that make that true? To make a broad sweeping statement like that without presenting any data or a complete examination of the system dynamics is setting off my bs alarm bigtime. All these tuning books suck. I wish one were written in a true engineering article format.

Now that I ranted about the quality of the info which we've all got to deal with these issues. Iit occurred to me there might be a simple way to cobble together a cure if I can determine it's utimately needed. Since at WOT the mixture is going extremely lean then the obvious solution is to increase the available energy which operates the induction system but only at the WOT condition. This could be done by just increasing the pressure gradient between the air pocket in the float bowl and the main venturi. Can't really do anything about the main venturi since I'd leave it sized for the mixture needs of a partial open throttle condition. What possibly could be done though is to slightly elevate the air pressure in the float bowl but only at WOT. The pressure source could be a small CO2 bottle about a 10 lbs size would be good for months of driving. It would supply a pnuematic control valve which I'd cobble onto the end of the Weber throttle spindle. It's designed to provide the extra potential energy via a pressure increase at the WOT position and releases that additional pressure at any other spindle position. The control valve would also have a second stage which would operate at a few psi to close two more pnuematically operated valves which are fitted into the vent holes of the carb body which are there to breathe into the airbox. Possibly most of the pieces necessary could be purchased for a few hundred dollars or less. Best guess is the pressure needed to do the trick is a few inches of mercury at most. Just a thought. :)
Keith
p.s. Hey Steve, if I can't resolve these issues then I will be installing my TJ fuel injection system afterall.
type26owner
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 1616
Joined: 18 Sep 2003

PostPost by: type26owner » Mon Apr 12, 2004 1:15 am

Was able to enlarge the fuel passages on sets of 40F2, 40F13 & 40F11 idle jets to make them into the NLA 45 ones I needed. Drilled a tight fitting pocket into a block of nylon which accepted the engagement end of the idle jet. This centered the jet body closely to the quill spindle of the milling machine. The carbide drill would hold the hole size to a few microns of error in diameter no problem. Measured it with the pin gauges I bought from Pierce Manifolds. 0.45mm just happened to be smallest size on the gauge. Didn't break a drill or ruin any jets in the process. Whew!

Found a set of 120 and 100 air corrector jets I'd forgotten I had. Installed the 100 ones to see what would happen. Seems like I'm able to move the errant slope of mixtures around the scale but not reduce the slope itself. Counting on Weber having done the engineering and applied the capability to overcome this condition. Just have to figure out how it's done. :(
Keith
type26owner
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 1616
Joined: 18 Sep 2003

PostPost by: type26owner » Tue Apr 13, 2004 2:10 am

Changing to a size 100 air corrector and a 100 main jet has apparently significantly changed the viscosity of the air/fuel mixture. Now when in 4th gear and at any rpm below 3k when the throttle plates are opened pass halfway the mixture goes extremely lean too. This is to expected if the mixture is trimmed to be too lean on the progressive holes by moving the plates away from them enough and the main venturi does not generate enough vacuum from the low airflow to extract the denser air/fuel mixture from the auxillary venturi. The stock jet settings were so overly rich this never happen before. Maybe this is what's referred to as being 'drivablity' but not explained in any detail so one would actually know it. Go figure.

The baffling part is the Weber Tuning Manual describes a simple carburetor on pg.9 and the leaning at the high rpms I'm experiencing is contrary the basic principles too. No help there either. :angry:

Going to start over again and make sure the AFM is giving me believable info. It's brand new technology and it's possible I've stumbled upon a glitch here.
Keith
type26owner
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 1616
Joined: 18 Sep 2003

PostPost by: BillGavin » Tue Apr 13, 2004 12:12 pm

Does the leanness at half throttle show up as a stumble?

It always pays to be sure of the measuring equipment, but so far it seems to be giving good data.

Have you tried different Auxiliary Venturies? You may be able to get a stronger signal to the jets. You may also want to try different emulsion tubes.

You're well into uncharted territory, it seems. With methodical work and the A/F meter, you should eventually win.

- Bill
User avatar
BillGavin
Second Gear
Second Gear
 
Posts: 136
Joined: 19 Sep 2003

PostPost by: type26owner » Tue Apr 13, 2004 12:57 pm

Hi Bill,
Does the leanness at half throttle show up as a stumble?

Would better describe like I turned the ignition off. Not good for throttle response but I'm getting fantastic mileage now. By backing away with these mixture settings one could definitely live with a small stumble by just not lugging the engine at low rpms but really benefit from the extra fuel mileage.

It always pays to be sure of the measuring equipment, but so far it seems to be giving good data.

I'm going to send it back to the factory and pay to have it requalified again. Could be I'm just using it wrong too. I'll cover those ducks next after I try the larger main venturis. This process could take a couple of weeks cause I've got other obligations to fulfill at the moment.

Have you tried different Auxiliary Venturies? You may be able to get a stronger signal to the jets. You may also want to try different emulsion tubes.

Nope. There are only two sizes available. The stock 4.5 ones are the larger ones. Have any of 3.5 ones laying around by chance? Again the written advice about what to expect from a particular emulsion tube is really not useful. I got the stock F11 ones and a set of F7 which are suppose to be fatter at the low rpms. Looks like the F16 ones are the direction to go possibly though but how does one to know without swapping and sucking. Even the Weber Manual is totally vague on this subject. :(

You're well into uncharted territory, it seems. With methodical work and the A/F meter, you should eventually win.

Possibly. Going to definitely find the limits in no uncertain terms that I can understand and not blindly believe what's been written. The experts could be wrong. I've succeeded many times in building devices that 'the experts' had deemed impossible. In my opinion the Weber designer really was a true master of his craft. I just need to kind of reverse engineer this stuff to explore the potential fully and find the true limits. Fun Stuff! :D

Oh, forgot to add I removed the wires from the third progressive hole already. Can't say for certain but after having swapped lots of idle jets the trend is the ones with a smallish well diameter universally do not improve the situation on my twincam. Haven't the faintest idea why this would be the case , it just is.:blink:
Regards,
Keith
p.s. If only Carroll Smith had written a book on Weber tuning. He's got my respect and nearly blind trust.
type26owner
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 1616
Joined: 18 Sep 2003

PostPost by: type26owner » Wed Apr 14, 2004 2:27 am

Okay, here's my new theory. It's so simple and elegant, it's completely stupid and as far as I can tell completely overlooked by the experts. Hope it turns out to be actually true. The air/fuel mixture properties for each of the two jetting systems which must be met first is to match the fluid VISCOSITY by deriving the necessary ratio of air to fuel ratio via the jet(s) selection itself. Secondarily, the total quantity must be then be adjusted to get proper mixture to support the required type of combustion ALL THE WHILE MAINTAINING THE VISCOSITY. In other words one has to dance with the main venturi and the progressive holes viscosity sweet spot and then properly feed the combustion process. For the idle jets this means finding the best ratio for the viscostiy and then staying with that ratio. Weber has made this a complete nightmare with their random designations. It might actually be easier to modify to suit the needed jets then to deal with the Weber ones. The main jet and air corrector RATIO OF AREAS must be determined for the best viscosity and then both jets must be changed to increase or decrease as a matched set in order to get the proper amount of mixture to optimize the combustion. The limit here is to determine the necessary viscosity to best match the dynamic range of the airflow of the main and auxillary venturis to maintain a somewhat constant slope of the changing vacuum level to achieve a steady state mixture over the rpm range.

How this is done as separate tuning steps I have not solved yet. Hoping to make some headway otherwise it's such a complex problem that tuning the standard way is almost impossible and so daunting I'm considering giving it up as a hopeless endeavor for a rookie like myself to even do it.:(
Keith
type26owner
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 1616
Joined: 18 Sep 2003

PostPost by: BillGavin » Wed Apr 14, 2004 11:58 am

That makes sense to me.

I don't know how you would home in on the right series of idle jets, other than the usual method of an educated guess on the series and size, then trying it with the idle screws backed out a certain amount - what is it, 1-3/4 turns for the early, fine thread needles?

There are lots of combinations that will work at this point, so the next step is to make more educated guesses about the emulsion tube, main jet, and air corrector. Then more "suck it and see" until the transitions are right and the mixture good through the midrange.

If your theory is correct, there should be a matching of an idle jet series and an emulsion tube that will match the emulsion viscosities, at least in the transition zone. (I can hear Rod Serlings voice now, "Welcome to..... The Transition Zone")

None of this stuff is linear, so it seems that there will always be some aspect of hit-or-miss. I'm not altogether sure of the influence of the emulsion tubes' characteristics - we have ID, OD, size of holes, number of holes, and distribution of holes, and maybe more. The idle jets are simpler, having just ID, air bleed, and delivery size, but still non-linear.

Complicating it all, the behavior of the jets is considered to vary considerably depending on whether they are factory-made or redrilled, as the radius at the ends of the holes will vary. Idle jets, as far as I can see, are just drilled at the factory, so they should be OK, as long as you can keep track of what size you have.

Something else to consider is that we're working with a number of discrete steps for each variable, and their effects are non-linear, so it probably is not even possible to get perfect mixture through the range. We can, however, measure A/F ratio fairly precisely: this brings up the question "how close is close enough?".

The cost of this exercise is also an issue: I have a lot of Weber stuff left over from my racing Mini and time trial Midget, but everything is in pairs, so I'd still have to spend lots of $$ to play around much with the Lotus.
User avatar
BillGavin
Second Gear
Second Gear
 
Posts: 136
Joined: 19 Sep 2003

PostPost by: type26owner » Wed Apr 14, 2004 1:32 pm

Bill,
Everything clicks into place if you have the correct perspective and it then becomes a simple matrix to solve. To do it you just pick a size of main jet that is suppose to be the mean and then step through the range of air correctors to find the combination which does not lean out at either or both ends of the rpm dynamic range of the jet system. If one cannot meet this best viscosity condition then probably the size selection of the main venturi is not optimized for the cfm of the induction needs of the engine. This superimposes the two flow components and the slopes from the venturis to the mixture viscosity on top of one another as a best fit senario for finding a steady state mixture. It does not have to be linear. Just the slope shapes and magnitudes have to match up fairly well. Once the equilibrium is established then it's just tuning for the mixture strength to achieve the desired 'drivabilty'(I can be vague too :D ). This is a best compromise fitting procedure. I'd be happy if I could halve the difference in the slope I've got right now. That is if the difference was only 2 lbs of air/hour instead of 4.

Suggest we figure out the conceptional solution before plunging into the practical aspects like the quality of modified jets. By doing this correctly one should be able to optimize both jet systems in like 20 swap & suck steps total. Tricky part is to optimize the size the venturis with no fuss or bother. I'm lazy and desire this to be so dead simple I can do it even after having consumed some beverages. :D
Regards,
Keith
type26owner
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 1616
Joined: 18 Sep 2003

PostPost by: steveww » Wed Apr 14, 2004 3:27 pm

Talking of viscosity.... The viscosity of fuel changes with its temp. How does the Weber correct for this?

I am still considering sticking with the Stromberg head and putting on a pair of SU HIF44. I have just received the dimensions from Burlen Fuel Systems so it is out to the garage with the tape measure to see if they would fit. Going to SU would give me a better choice of needles / springs. Now for the controversial bit :o A well set up SU is as good as any Weber side draft. As far as I can see the Stromberg head is well designed and should have no problems getting the required flow. However I am not looking for absolute power as the car is mainly used on the road and the extra economy will keep the range good from a small fuel tank. Thus on a long trip my car will be much faster than Weber equipped cars as I will have to stop for fuel less often ;)
User avatar
steveww
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 1838
Joined: 18 Sep 2003
Location: Northamptonshire, England

PostPost by: type26owner » Wed Apr 14, 2004 7:18 pm

Hi Steve,
Talking of viscosity.... The viscosity of fuel changes with its temp. How does the Weber correct for this?

The density difference between gasoline and air is about 570:1 so any viscosity variation of either due to temp is in the noise. What is key to this and is a first order effect is the mixing the right ratio of air to gasoline to modify it's fluid properties to best suit the small potential energy from the vacuum sources. This dominates the functioning of the carbie.

IIRC it's called latent heat. It's got built-in cooling. Suspect it's best to establish a thermal equilibium by having an enclosed engine bay and NOT overcooling the engine.

Not going to comment on your other topics cause it's well outside my expertise and my interests as of right now.
Regards,
Keith
type26owner
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 1616
Joined: 18 Sep 2003

PostPost by: type26owner » Wed Apr 14, 2004 7:42 pm

With the 30mm main venturi installed I can state that the main jet system starts contributing at about a 2k rpm threshhold when the viscosity is close.

Having run around exploring the lean mixture realm I've stumbled upon another problem I didn't know existed. On a frequency of about one in five times when into the accelerator pump I get a bigtime cutout. Suspect it's vibration related and I'm getting foaming or the check valves are not sealing as a result. The Weber Manual covers this condition in detail and I'm going to change the natural frequency as per their recommendations. Hope that is the cure. :(
Keith
type26owner
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 1616
Joined: 18 Sep 2003

PostPost by: type26owner » Thu Apr 15, 2004 10:10 am

Been hitting the books and the found a description to a condition which perfectly matches my accelerator pump cutout problem. Compressing it a bit it says under some condtions when the throttle is opened quickly the atomized fuel will wet out onto the walls instantaneously and cause a weak mixture and a resulting 'misfire momentarily' is possible. Of course it does not describe what actually are those conditions! I think maybe I've met them though but the good news is it can and will be tuned away. :)
Keith
type26owner
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 1616
Joined: 18 Sep 2003
Next

Total Online:

Users browsing this forum: No registered users and 3 guests