Page 1 of 4

Distributor question

PostPosted: Sat Oct 15, 2005 5:17 am
by stuartgb100
Having begun to tune my newly-purchased '71 Sprint, it soon became clear that the distributor was significantly worn. While talking to Lucas's Technical Department on another matter, I asked who they would recommend to carry out the rebuild. They said they always answer with H & H Ignition Solutions in the West Midlands.

I duly packed it off to them, with a note that set out the advance curve requirements (all as per Miles Wilkins' figures ... page 168 in the book "Lotus Twin-Cam Engine").

A week went by, and then a phonecall from the proprietor: He confirmed it had been worn, and that it was now ready. He said that getting that advance curve had been quite a challenge ... particularly stopping any advance below 2000 rpm.

To stop that advance, he had to source the strongest spring they had ever used ! Not only that, but he'd had to use two of them !!!

Bearing in mind they had never been asked to do this before (and he reckons they've rebuilt tens if not hundreds of twin-cam dizzys) did I still want it sent back to me with these 'funny' settings, or he'd reset it to
'standard' settings?

So that's where I'm at. What to do?

BTW, the distributor is a 41189A.

He'll send it with Miles' settings, and if I'm not happy he'll reset it for no extra charge.
I think I'll ring him on Monday, and get the advance settings that he would normally use.

Has anyone any thoughts/ideas?



Re: Distributor question

PostPosted: Sat Oct 15, 2005 8:36 am
by rgh0
I dont know what he means by "funny" settings or "standard" . The settings in the book are the lotus spec and any decent distributor place should have no problem replicating any of them . Getting no advance below 2000 rpm is not hard and does not require 2 extra heavy springs. It just requires the right preload on the springs so no additional stretch occurs until the force from the weights overcomes the prefload at 2000 rpm. I would be very suspicious of the person doing this work of your distributor as he does not sound like he knows what he is doing.

In my opinion most Weber Twink road engines on normal road fuels need around 25 to 28 degrees max advance and that the curve can be faster than the various Lotus curves with maximum advance coming in around 3000 to 4000 rpm.


dizzy advance

PostPosted: Sat Oct 15, 2005 2:11 pm
by john-c-elan+2
I presume you had deducted the 'static' settings from the figures sent to him ? ie. less 10 or 12 degrees depending on whether big valve or not ?

I had my 41189 rebuilt by Bestek (who I now see offering electronic diz set-ups on ebay) and it follows close to MW figures +- few degrees. This maybe only my rev counter error ? The more I delve into dizzys, the more often I find that its only the 2 springs that alter the advance, and the rate of advance - nothing to do with the weights.

Re: Distributor question

PostPosted: Sat Oct 15, 2005 2:11 pm
by M100
I thought that some of the ignition timing/advance curves in the Wilkins book were one of the things revised in the more recent second edition.

From what I recall they didn't match up to the Lotus workshop manual figures either!

Re: Distributor question

PostPosted: Mon Oct 17, 2005 3:47 pm
by stuartgb100
It would seem that the plot is thickening......

I cannot speak to the right person at H&H until Tuesday/Wednesday, so in the meantime rang Aldon Automotive. They are one of the most respected companies for ignition know-how (here in the UK). They produce their own distributors - one of which is a replacement for the 41189 series. I was interested to know what advance curve they used.

I began by explaining what I'd done with my distributor. I was passed over to the guy who does all Aldon's distributors, and explained again to him. It must be a coincidence, but Paul Matty's asked exactly the same question of them last week!!!

Aldon's advance settings are as follows:
Up to 1000 rpm (crankshaft) no mechanical advance
At 1500 rpm 2-3 degrees
At 2000 rpm 4-6 degrees
At 2500 rpm 7 degrees (maximum advance)

From memory, H&H's figures were similar.

So on the one hand H&H have been servicing/rebuilding 41189 distributors for customers for years, whilst on the other Aldon have been supplying their own distributors to customers for years.

In both cases, the similar advance curves are significantly different to Miles Wilkins'.

BTW, Aldons confirmed that they sublet work to H&H !

The only other reference I have found is from a David Vizard book. He gives no detailed figures, but says:
"The purely mechanical advance distributor has an advance curve which runs out at 6500 crankshaft rpm........ The non-vacuum distributor also advances at a greater rate up to 2000 rpm."

So he is saying (like Miles) that advance occurs up to 6500 rpm, but he also says (unlik Miles) that advance occurs below 2000 rpm.

So, I phoned Paul Matty. They are also confused. They also expected everyone to be using Miles's figures.

Fortunately, Miles Wilkins is someone they regularly talk to, and they were hoping to contact him this afternoon. Perhaps we'll get to the bottom of it.

Will update in due course.


Re: Distributor question

PostPosted: Tue Oct 18, 2005 10:30 am
by rgh0
I dont see a lot of difference between the Aldon curve and the later Weber curves ( 41225 and 41189) in the Wilkins book.

10 to 12 crank degrees static advance and 14 crank degrees maximum centrifugal advance at 5000 crank rpm gives 24 to 26 degrees total advance most of which comes in before 4000 crank rpm. This is essentially the same as the 41189 curve in the Wilkins book. Dont confuse distributor RPM with crank rpm. The Wilkins book is in crank rpm the distributor companies talk in distributor rpm.

I would be a little more aggressive on the curve than any of the following but otherwise both the Aldon curve and the 41225 and 41189 Lotus / Wilkins curves are fine. To do better than any of these 3 you really need to run your specific engine and fuel on a dyno to optimise the curve.

The early Lotus curves with advance above 30 degrees ( 40930 and 40953) I think are just plain wrong for a road engine today. They may have been right in their day with the fuels available then but they dont work now even if the rest of the engine is still to the early specifications which most are not. The 40953 curve for the Stromberg engines are different due to the different characteristics of that carb and still seems to work right for those engines. The 41225A curve was an emission curve and I would also ignore it for all practical purposes today.

In general it all makes sense once you work through the details. The only things that dont make sense was the original comments on how diffiuclt it was to match the Wilkins 41189 curve



Re: Distributor question

PostPosted: Tue Oct 18, 2005 5:54 pm
by stuartgb100

Thanks for the comments .... but I'm not confused (AFAIK) regarding crankshaft and distributor degrees!!

However, there are plenty reputable bodies (apart from me) out there, who are confused .... and becoming more so, not less!

Here's an update:

I today received a call from Paul Matty's. They confirmed they spoke to Miles on Monday afternoon. As they thought, he not surprisingly stood by his figures. They also, this morning, had a conversation with Aldons.

So I again phoned Aldons. I asked more detailed questions. I now learnt something that I had forgotten, namely that Aldon do a standard 41189 rebuild to their own advance curve, and also a 103 TC distributor of their own design (which has different advance characteristics). It turns out this latter dizzy is very popular with QED.

I phone QED who confirm they basically only use Aldon's 103 TC when building their dizzy-based engines. They do not seem to feel the "no mechanical advance before 2000 rpm" is correct. And they have extensive dyno facilities for testing, don't they!

So another call to Aldon's. To muddy the water further, they advocate the 103 TC for all twin-cams EXCEPT the Sprint. For that they recommend their 41189 rebuild !!! Aldon's man is now reconsidering that recommendation !

Back to Paul Matty's. They have found a 41189 in good condition. It will be sent to H & H for rebuilding, but before the rebuild, H & H will be asked to map the existing advance characteristics (I should have thought of that for mine). So for the meantime, I am leaving my 41189 at H & H just for comparison. With luck will have results on Friday/Monday.

My conclusions (tentative) so far:

a) Every set of advance figures I have seen, show mechanical advance occurring below 2000 crank rpm. Except Miles'.
b) Lucas' own figures show advance below 2000 rpm.
c) Both Aldon and H & H (both very respected) state that the only way of stopping advance below 2000 rpm is with a much stronger spring(s).

So, my question is this: Why in decades of rebuilding 41189 distributors, have neither company had a customer's distributor arrive with the stronger spring(s) already fitted?

And I am pretty darned sure that my Ford Twin Cams in the early 70's all showed similar earlier advance characteristics.

Nope, there's something wrong here.

I await a torrent of responses!



Re: Distributor question

PostPosted: Tue Oct 18, 2005 6:09 pm
by type26owner
Hey Rohan,
Chew on this. If the jetting is set too fat then the burn rate is slowed. IIRC, my stock AFR was around 9:1 at the lower rpms. That would call for more total advance to compensate. Peak cylinder pressure must happen around 15 degrees ATDC to extract the most power.

Re: Distributor question

PostPosted: Tue Oct 18, 2005 9:11 pm
by steveww
Just a quick note to say that I had a 40953 rebuilt by H&H and they did an excellent job 8)

Re: Distributor question

PostPosted: Wed Oct 19, 2005 10:17 am
by rgh0

I agree that the stock Lotus advance curves were probably distorted and maybe more avanced than should have been by the stock over rich setting on the carbs. On the other hand the curves are probably conservative to ensure no detonation. Where it all ends up with current engines and current fuels is anybodies guess !!!! The only thing to do is get the carbs setup right then get the distributor set up right. Easy but expensive on a dyno, a little more time consuming but possible on the road. Just spent 4 hours on the dyno this morning with my Elan I will post the result when I get a chance to properly analyse them.


Re: Distributor question

PostPosted: Wed Oct 19, 2005 10:51 am
by rgh0

The orginal Lotus specs and the Wilkins replication of these specs minus the errors in the Lotus manual had no centrifugal advance below 2000 crank rpm for the 41189 curve. Now Wilkins may have just replicated a Lotus mistake and if orginal Lucas data shows a different curve then I would believe the Lucas data. Wilkins use of the 2000 crank rpm number was based on a Lotus mistake in their manual where the mixed up crankshaft and distributor rpm and showed 1000 crank rpm where if the overall table is to be consistent it should have been 2000 crank rpm. But maybe Lotus were even more mixed up and got the first number in the table correct as 1000 crankshaft rpm but got the rest wrong, who knows !

I dont really believe this is critical, as whats going on between 1000 and 2000 crank rpm in terms of power and spark advance is not of great concern in a road or track car. If others have a curve that has advance that beings at 1000 crank rpm versus 2000 rpm then thats fine as long as the general shape of the curve and total advance is similar to the orginal curve. In reality the couple of degrees we are talking about is not critical unless you are seeking the last couple of percent of horspower in which case you need to set it up on a dyno and find whats right for your specific engine and fuel anyhow.

As for setting up to get no advance from 2000 rpm versus 1000 rpm I continue to maintain that this does not require a radical change in number of springs or spring strength ( rate in lbs /inch). It just requires a change in the amount of spring pretension. You need to have a shorter spring stretched more but at the same spring rate and number of springs so that the slope of the advance curve remains the same. Advance starts when the force of the centrifugal weights exceeds the preload on the springs. The slope of the advance curve is governed by the number of springs and spring rate. Curves with 2 slopes over the rev range are achieved by having a loose second spring that comes into action only part way through the rev range. Whther 1000 or 2000 crank rpm is right I dont know ut either is easily achievable. hard to determine from an orginal old distributor unless you have the orginal srping details as the old springs will have stretched and lost their preload by now !

I still question the understanding of this by all the specialist you have talked to so far as you say - confusion only increases. But then many of these places do it by trial and eror with no understanding of the engineering involved.

What is the advance characteristic of the 103TC curve. I will bet it has something like 18 to 20 crank degree advance coming in linearily from 1000 to 3000 / 4000 crank rpm if QED use it on their developed engines. This is the sort of curve I use and works best with minor tweaks in both road and track twin cams on Australian fuels.


Re: Distributor question

PostPosted: Wed Oct 19, 2005 2:06 pm
by type26owner
The only thing to do is get the carbs setup right then get the distributor set up right.

Yes, I totally agree. The AFR must be achieved first and only then the timing tweaked.

It's been my experience that the twinkcam is prone to detonate with 10 degrees of static timing, at low rpms, under high loads and at small throttle openings. The sound it makes has best been described as the noise large rocks make when banged together under water. Once you open up the throttles and fill the combustion chamber with an overly fat mixture the burn rate slows by alot and it is tolerant of much more advance.

Re: Distributor question

PostPosted: Wed Oct 19, 2005 2:36 pm
by iain.hamlton
I recently fitted a new 43D4 distributor after finding the collar where the clamp grips was damaged on my original 23D4.

I was unhappy with the way the car ran with the new distributor, so measured the advanced curve and plotted it against the Lotus manual. The 43D4 starts to advance earlier and is a bit erratic, and has greater total centrifugal advance. I retimed it at 2500-3500 rpm, when it matches the manual closely. This equals a static advance of 6 degrees.

Here are the measurements against the workshop manual:

41189, big valve twincam. Measurements taken using a timing light probably +/- 1 degree or so
revs; from manual; 43d4 retarded 6 degrees;
600; 12; 6;
750; 12; 6;
800; 12; 7;
850; 12; 8;
1000; 12; 8;
1000; 12; 9;
1200; 14; 14;
1200; 14; 14;
1500; 17; 16;
1500; 17; 16;
1800; 19; 18;
2000; 21; 21;
2500; 26; 26;
over 2500; 26; 26;

The car goes well, but tickover is a bit odd as it is more retarded than originally. You will notice the timing agrees with the manual from about 1200 rpm, which I suggest, is all that matters. It will now tick over at 850 rpm; previously, it would not stay below 1100rpm when warm.

As indicated earlier, Paul Matty's are going to get an original 23D4 measured. I'll be very interested to hear how they get on. Thanks for the help, Roy.

Of course, what would be really interesting would be for someone to try it on a dynometer or rolling road and measure the timing advance to give maximum power at each engine speed. It may be that the advance curve proposed in Mr. Wilkin's book may be better.

best regards, Iain

Re: Distributor question

PostPosted: Wed Oct 19, 2005 6:27 pm
by stuartgb100

The problem with Paul Matty's 41189 is that they DO NOT know if it is original .... it could have been altered over time. So even if in good condition, we cannot be sure that it is an original curve. All we'll get is a comparison with mine (which was most definitely worn), but H & H will be able to compare springs, weights and the shaft itself (7 degrees IIRC).

I managed to speak to a Lucas chap. He confirmed that with the demise of high-octane fuel, it would be likely that the original distributor characteristics would need to change as follows:
a) retard the static advance ..... probably by 2 degrees
and ideally
b) change the advance curve so as to duplicate the original curve, but earlier in the rev range.


The 103TC figures are Static (10 degrees) at 1000 rpm, Static + 8 degrees at 2000 rpm rising to Static + 24 degrees at 5000 rpm.
(crank rpm and degrees quoted).

I would have thought QED are using these for engines up to approx 140 / 145 bhp? After that Electronic?

I should have asked Aldon if they only supply one variant. (g).

Aldon did say that dyno tests for the specific engine under test, might well result in the static advance being reduced ..... even down to 6 degrees BTDC. Ergo, QED alter static advance rather than the distributor's advance curve (seems reasonable).

Unless someone has a definitive Lucas specification (and they didn't just hand them out) or a mint 41189 (never even fitted), then I doubt there will be any "for sure" outcome.

In any event, I for one am beginning to become convinced that the original Lucas spec required some advance before 2000 crankshaft rpm.



Re: Distributor question

PostPosted: Wed Oct 19, 2005 9:36 pm
by mikefromengland
i own a +2s130.i was using my original dizzy 41189a but as it was getting on a bit decided to renew it.bought a 43d4 from paul matty sprint spec.set it at 10 degrees btdc static instead of 12 because of the awful 95 ron unleaded.i also use valve masterplus with octane booster.i have done a few thousand miles and all seems fine.tickover is a bit rough but then i dont let it tick over for long.i think we are all to used to modern cars and the smoothness of them .lets keep these lotuses on the road but most importently enjoy them and dont be so critical have fun i do thats what these cars are about kind regards mike