Page 2 of 4

Re: Distributor question

PostPosted: Thu Oct 20, 2005 12:02 pm
by rgh0
24 crank degrees centrifugal advance in the 103TC distributor is interesting as I have never seen a Weber engine in Australia that needs that degree of centrifugal advance no matter what the specification. The comment that static setting may be reducd to 6 degrees is interesting because if you gave me a 103TC distributor to put into my race engines that is around where I would end up setting it to reduce some of the extra centrifugal advance and keep total advance to 30 degrees.

What it demonstrates is that many different curves can be made to work as long as you are prepared to make some compromise somewhere in the rev range where it does not matter to your application.

Rohan

Re: Distributor question

PostPosted: Mon Oct 24, 2005 6:10 pm
by stuartgb100
Update:

H&H received Paul Matty's distributor and confirmed it has the same springs etc as mine. In fact, it's components are the same as all the other 41189's they've received over the years.

What I hadn't told you was that sometime back H&H inherited a 'box' of unused 41189's, which they mapped and then sold. This mapped curve is the one they use to rebuild and match.

When mine is returned (hopefully at the end of the week) they'll include a written map. I'll post the details.

BTW (and no association with them whatsoever) they quoted ?47 plus postage, and turn around within a week. In my case that includes a brand new shaft (I knew it was worn, but not that worn!).

Regards

Re: Distributor question

PostPosted: Tue Nov 01, 2005 8:01 pm
by stuartgb100
A final update.

Distributor has now arrived.

Based on the 'fact' (and my belief) that H & H did indeed once have a batch of original and unused 41189's, I finally decided to ask them to use the curve they matched from those 'originals'.

This was in part due to the fact that their curve closely followed the Lucas data that I had unearthed.

Here is the curve they have set mine to, together with the corresponding Lucas data:

(NB: distributor rpm and degrees)

No advance below 400 rpm

at 600 rpm ..... Lucas ..... 0-2 deg ..... H&H ..... 1 deg
.. 1000 ........................... 4-6 .......................... 4.5
.. 3250 ......................... ..6-8 .......................... 8

Of course the Static setting must be added. I am trying 10 degrees, due to the reduced octane rating.

I am still trying to establish what modifications Lucas made to the individual model numbers. For example, mine is a 41189D.

Would anyone have an idea as to how this might vary from the A version?

Regards,

Stuart.

Re: Distributor question

PostPosted: Tue Nov 01, 2005 10:07 pm
by type26owner
Stuart,
The exact timing is not all that important as far as getting the MBT (maximum brake torque) out of the engine. Here's a graph which shows the typical MBT curve deviation from 100% torque due to timing errors. As you can see it's not a big deal and trying to achieve the perfect timing curve is an exercise in quickly diminishing returns. What is more important is preventing detonation or pinging.
http://www.daytona-sensors.com/tech_tuning.html

Re: Distributor question

PostPosted: Wed Nov 02, 2005 11:03 am
by steveww
Very interesting Keith. I was always under the impression that ignition timing was more sensitive than that. You learn something new every day :)

Scatter

PostPosted: Wed Nov 02, 2005 11:46 am
by fasterbyelan
I fully agree with Steve's comments - thank you Keith. It did however lead me to the question, in view of the recent discussions on the forum, how much scatter are people experiencing with their distributor's and have they noticed any genuine improvements once they have been replaced/repaired?

Although from an engineering point of view wear is not acceptable, would variations in timing as a result of scatter be noticeable (providing it does not cause pinking and/or the amount of scatter is not excessive)? From the information provided I would have to say no.

Karl

1970 S4 FHC SE

Re: Distributor question

PostPosted: Wed Nov 02, 2005 2:36 pm
by type26owner
There's a direct one to one degree angular relationship which makes it very insensitive. The peak pressure has the highest mechanical advantage to exert the force to the crankshaft which happens at about 15 degrees ATDC. If that happens a few degrees from ideal one way or the other it makes very little difference. Kinda puts a new perspective on what to concentrate on to get the performance out of the twinkcam. :wink:

Karl, the scatter does matter since it can get excessive quickly. The issue is one of play in the distributor mainshaft in the axial direction. What causes this is the high helix gear teeth arrangement of the driven gear against the jackshaft. If you replace the dizzy with another type you still have the same driven gear arrangement causing the same scatter if it gets loose. The fix is really easy and the Lucas unit is top quality if you just don't let it go to shit.

Owning one of these cars requires you to have the necessary tools to work on it. I acquired a Sun distributor machine so I could service my own dizzies because you can't rely on anyone 100% these days. Those folks which are doing the recurving MUST know that the slope is not all that important. Shame on them to lead you down the path just to pick your pockets clean.

Re: Distributor question

PostPosted: Wed Nov 02, 2005 3:30 pm
by iain.hamlton
I changed my distributor mainly because the clamp no longer held reliably: the collar it works on had partly collapsed.

The car ran pretty well, although there was plenty of wobble in the shaft. This caused some scatter.... But i am right in thinking it would also make the gap at the points variable? And wouldn't this affect the qulaity of the spark as well as the timing?

With the new distributor the car runs better now I have the timing set to agree with the maximum advance in the manual (see earlier in the thread), and I am happy.

Question for Keith:

The link you attached is interesting, but why is a different advance curve required for Strombergs as opposed to webers? It appears webers do not need as much advance at high revs. This point is made by Maurice Wilkins in his book.

best regards, Iain

Re: Distributor question

PostPosted: Wed Nov 02, 2005 3:39 pm
by type26owner
Hi Ian,
The link you attached is interesting, but why is a different advance curve required for Strombergs as opposed to webers?

That's an easy question to answer. The AFR (air/fuel ratio) determines the burn rate. The fastest burning AFR is right around 12:1. Suspect the Stromberg supplies the mixture a bit on the leaner side and requires more total advance to achieve the peak pressure at the ideal angle of the crankshaft.

Re: Distributor question

PostPosted: Wed Nov 02, 2005 4:21 pm
by iain.hamlton
So, if I understand, the Webers produce a rich mixture at high revs, which protects against detonation; makes timing less critical; makes more power (but at the expense of emission and fuel consumption). Sounds like a deal!

best regards, iain

Re: Distributor question

PostPosted: Wed Nov 02, 2005 5:31 pm
by type26owner
Bingo! If I were you I'd stick with the Strombergs. The Webers provide a tiny bit more power but at the expense of the using more fuel. I tried to set the cruise mixture to lambda 1 but the engine was surging due to it being too lean. Suspect the difference is the Strombergs are much better at atomizing the fuel. The potential energy the Weber depends on just a few millimeters of water and the vacuum signal is not enough to disperse the fuel and vaporize it efficiently. I've had to make for this shortcoming by providing more heat by running the engine at the maximum allowed by the cooling system to offset the Latent Heat requirement. The Stromberg on the other hand always maintains a pressure differential of inches of mercury. That's several orders of magnitude of more energy. Duh, kinda obvious which carbie is going to be the easier one to tune. I would look into adding a Helmzholt resonance to the second firing cylinder to solve the unequal mixture problem because of siamese ports though. The crossover pipe is most likely quite beneficial too. Those points are just an educated guess though. Has anyone else seen the trick adjustable resonators on the F1 cars these days?

It's wonder the Webers work as well as they do when setup correctly. Too bad most Weber users never get there. I've tried to put the required info out there in the public sector. All the stupid, wrong technical articles are still up on the internet, even the ones on the LotusElan.net. If you believe those articles then you've drunk the poison Kool-Aid and you're dead already and don't know it yet. :lol:

Re: Distributor question

PostPosted: Wed Nov 02, 2005 8:27 pm
by steveww
By messing with the needles the Strombergs can be set up to run richer. My tweeked twinc runs at 12.8 to 13.8 AFR under various conditions. Running the stock curve for the 40953 with no problems with pre-ignition. Currently getting 130bhp around 6000 - 65000 with more to come after some extra tweeks over the winter.

Re: Distributor question

PostPosted: Wed Nov 02, 2005 10:05 pm
by type26owner
Hey Steve,
If you would add a TPS and record the data with LogWorks we could then do a side by side comparison of the attribute differences of the Stromberg versus the Weber. I think there is an overlay capability of LogWorks which I've have not played with yet too. Bet that would be quite revealing.

Re: Distributor question

PostPosted: Thu Nov 03, 2005 11:13 am
by rgh0
I set the ignition maximum advance on my latest engine to 32 degrees based on the previous engine dyno work as a starting point. On the recent dyno run changing from 32 to 30 degrees was worth 2 kW at the rear wheels at 5000 rpm or around 3%. Not a huge number but not insignificant.

However I agree the shape of the curve is not signficant as long as you get the best maximum adance and get the advance in as quick as the fuel and engine you are using will accept. The H&H and Lucas curves would be too slow in getting to maximum advance for a typical road Weber engine on Australian premium unleaded. They a more like the original early Lotus curve in the manual. But maybe they suit the current UK fuels. Surely someone has some real personal dyno data rather than basing it on what others say without details of the type of engines and fuel and how it was tested.

Rohan

Re: Distributor question

PostPosted: Thu Nov 03, 2005 11:15 am
by M100
type26owner wrote:Stuart,
The exact timing is not all that important as far as getting the MBT (maximum brake torque) out of the engine. Here's a graph which shows the typical MBT curve deviation from 100% torque due to timing errors. As you can see it's not a big deal and trying to achieve the perfect timing curve is an exercise in quickly diminishing returns. What is more important is preventing detonation or pinging.
http://www.daytona-sensors.com/tech_tuning.html


Despite the reference document being somewhat respected I'd take that information with a huge chunk of salt especially when applied to the real world non-steady state conditions.

Assume you intend to time to an optimised MBT but you then have a 10 degree timing error, which according to the graph, gives approx 95% of max torque. That might possibly be the case at full throttle with a fixed load but in no way does that replicate real world driving conditions. As many others will have observed 10 degrees timing error imposed on the timing curve, either advance or retarded can quite easily make the engine undriveable - sometimes totally.

type26owner wrote:Those folks which are doing the recurving MUST know that the slope is not all that important. Shame on them to lead you down the path just to pick your pockets clean.


Unless your engine only ever operates in a very limited rev range by definition the slope of the timing curve *has* to be important. OK so you might need the maximum advance to come in rapidly on the twincam but attaining that advance has to be controlled, and that control is essential to keep the engine from running into detonation. Part throttle conditions are also significant (particularly for road use) so a better solution might be to add vacuum advance as well.

A totally standard engine using an optimised mechanical distributor setup when converted to a mapped ignition (measuring throttle angle, revs and manifold pressure), would always, without exception, increase the area under the torque curve, be more driveable and ultimately quicker on both the road and the race track. The distributor is a mechanical bodge, albeit a relatively good one.

I had considered getting a copy of Taylors book but after seeing that snippet I think I might stick to Heywood instead!