pre-programmed curves on 123 ignition - help please?

[christopher storey]

Look, a timing light is the only way to go. You can check both types of advance with this with the car stationary and you under the bonnet . First, disconnect the vac advance and check the timing at idle , which should be within spitting distance of 10 degs BTDC. The advance to the various rpm in the manual and check the total mech advance at that speed , which will be 10 degs PLUS the crank advance for the relevant speed.
At 3000 rpm you should get about 10+24 = 34 . Check that it does not continue to increase beyond 3000 rpm

Now reconnect the vac pipe . With the engine running on a light throttle at about 3000 rpm you should see total advance of about 50 - 52 degress BTDC. Sharply open the the throttle for a split second and the total advance should drop to 30 to 34 degs BTDC and then increase transiently back towards 50 ish as you release the throttle. If you get these figures everything is hunky dory . If you do not get them, and the differences are more than minimal , then the first thing to check is that your timing markers are correctly positioned, best done with a dial gauge in nos 1 or 6 cylinder ( search on this forum for the exact technique to use because with domed pistons it's a bit fiddly )

PS I hadn't realised this was not an E type,so the figures might differ slightly , but it makes no difference to the principles involved

[John ball]

Hi Steve and Tom,

Firstly, yes the TDC mark is correct. It has been check more than once with a dial gauge. The engine is correctly set statically at 10 BTDC. I used the timing gun and it produced the numbers I came up with. I took the car for a short drive on the flat and up and down a hill. It felt harsh, but did not Pink. But the car is quite loud.
I never took the car above 4,000 rpm both with the gun and on the road as I could see and hear it was not right.
I think this comes back to a comment I made before that the 123 vacuum is a switch not a proper vacuum??

I then went back to the generic standard preprogrammed vacuum setting that has a maximum of 10 degrees.
I also used the Distributor Doctor curve from MarkRado post in a previous set of posts, that has 19 at 3000 and 22.5 at 4000, plus my static of 10 BTDC.

The car drove much the same, but not as harsh and I think pulled slightly better up hill.

As I said before the test, I used Chris’s highest vacuum numbers, plus the highest advance numbers.
Both of these from the Jaguar manual.

Yes, I agree my car is not a standard 3.8 E Type, but being modified, it is closer to that than a standard XK120.

You guys are patient, but getting fed up with me. It gives me something to do in Covid times !!

[John ball]

Hello Chris, I think your post crossed with mine when I was typing it.

When I raised the revs to 3000 I never realised I was meant to see such a high advance number, as your post saying over 50 degrees. I just saw a much higher than anticipated constant advance degree and backed off.
Yes, I was doing it with the vacuum connected.

So, tomorrow I will go back re-enter the curves I set out in my previous post, try with and without vacuum to see if I get your numbers. As I said on the previous post, I was using your second set of higher vacuum figures and also the highest ignition advance from the Jaguar manual figures. I think I am right in saying the 3.8 can take higher advance than 4.2, thus assuming 3.4 can take even more. Although I am told higher compression should take less advance ??

Did you agree the two curves, looked ok ?

[Tom W]

I don’t know if the 123 vacuum sensor is a simple switch or a proper MAP sensor. One hopes it’s a proper MAP sensor, and the fact you can programme different values to different pressures suggests it is. Does the app show a varying vacuum value as you rev the engine?

There are several factors that determine the correct amount of mechanical advance. Combustion chamber design plays a large part. Compression ratio and fuel octane play a significant part too. Reducing the ignition advance can stop an engine pinking, but if it needs reducing excessively, it’s a sticking plaster solution to having an excessive compression ratio for the fuel being used. Intake manifold and fuel system, mixture strength and fuel atomisation/distribution also make a difference.

The optimum ignition timing can only really be found on an engine dyno, a rolling road or by spending a lot of time making repeated timed runs on the road. You need a way of accurately measuring the torque the engine produces under wide open throttle. Then you can try changing the ignition timing. If the engine makes more torque with more timing, then it needed more timing. If it pinks, you’ve gone too far. If it doesn’t pink, but makes no more torque, then you’ve gone too far.

A dyno session is very useful at arriving at close to the best settings. To set the timing by driving on the road, you need to make timed runs over the same RPM range, and over the same section of road in the same conditions. If the time drops, you’ve made an improvement. If it doesn’t, you haven’t. It can be appreciated though, that getting the accuracy and repeatability required to get the spot on optimum setting with the road method is going to be very difficult. The variables run to run are probably more than the measured performance gain when you get very close to optimum. The trial and error method does risk damaging to the engine should you go too far.

[mgcjag]

Just to add......when useing a timeing lamp do not rely on the app dashboard for the revs......check it with the revcounter in your car.......the app has been known to give false revs.....a friends was displaying 500rpm under....Steve

[mgcjag]

For anyone looking for 3.8 vac curves its worth looking at this thread re confusion on vac units.....Steve http://forum.etypeuk.com/viewtopic.php? ... 43#p139175

[rfs1957]

I'm not sure quite where to post this.

It stems from my wanting to make a map for the MegaJolt that is as true a representation of the original 3.8 ignition as possible, reflecting the effects of both the centrifugal and the vacuum vacuum advance, as I haven't found anything that comes remotely close in the MegaJolt section.

However, the 123 ignition is a not dissimilar platform, and as some of the debate on this string has covered the topic of vaccum advance, I'm going to put it here, and add a link to this on the MegaJolt page.

This is the reasoning I followed ............

The Jag advance curve specifications (p.338 Bentley) for the DMBZ6A distributor (and others) are given in relation to manifold vacuum expressed in inches of mercury.

Higher suck (the higher mercury figure, equals closed throttle) produces a higher advance figure, whilst wide-open throttle (WOT) produces the lower mercury figure, which produces a lower advance.

So far so good.

However, the Jaguar figures are not absolute pressures as understood by a MAP sensor, but relative ones.

The Jaguar vacuum advance operates linearly across a range between 7" and 14" of relative mercury, which corresponds to 24 KPa (WOT) and 48 KPa (closed throttle).

These are absolute values.

Critically, as the MAP sensors used in the MegaJolt and the 123 consider 100 KPa to be the "zero condition", atmospheric, the range of operation of the vacuum advance therefore corresponds to MAP readings NOT of 24 KPa, but 76 KPa (100 - 24) at WOT, and NOT 48 KPa, but 52 KPa (100 - 48) at closed throttle.

Quite counter-intuitively, then, the MegaJolt and 123 map vacuum values, and break-points, are absolutely not comparable with the Jaguar values, because their numerical variations occur in diametrically opposed directions.

A MAP sensor applies a HIGHER advance figure when confronted with the SMALLER number, which it sees at closed-throttle, and applies a LOWER advance when confronted with a HIGHER number, which it sees at WOT.

The maximum advance correction (16° crank) is therefore to be applied for MAP values of up to 52KPa, and then diminishes to zero for MAP values of 76KPa and over.

I think this takes a bit of getting one's head around, but anyone trying to recreate original advance maps needs to understand these nuances.



For my money, the original 3.8 ignition looked like this - the HIGHER TOLERANCE and LOWER TOLERANCE reflect the circa 4° tolerances that Jaguar specified in the centrifugal advance, so the actual ignition probably wandered around a bit in the middle.

I've cancelled the effect of manifold vacuum for the 500 rpm bins as originally the vacuum advance was fed from ported manifold pressure, and so would have seen no suck with butterflies shut.

Please feel free to wade in and correct my calculations and reasoning, and there may be better ways of explaining this !

[rfs1957]

Here's an updated file with the suck parameters in both "Hg and KPa.

[mgcjag]

Great work Rory...thanks for posting......should clear up a lot of programing issues for 123 users....Steve

[rfs1957]

With help from the irrepressible and helpful Rob at BritishVacuumUnit.com

https://www.britishvacuumunit.com/ (they also do distributor rebuilds)

I now have this extract from a Lucas book that confirms the same figures, reassuringly, that Jaguar specified in period.



Rob said, and I quote

The graph and some spec information show the curve and some show the range.

Both units 54410415 and the later replacement 54412348 are the same code.

In real life the Lucas test sheet may be the 7 hg can have a start test range of 6 to 8, full advance @ 14 hg may be 13 to 16 and the degrees may be 7 to 9.

When I build a unit I try to get them exact to the spec usually much closer than when they were new from Lucas.

If you're talking about the early XKE with the DMBZ6 40617 distributor, The XKE and XK150 cars were supplied when new at the dealer with the 54410415 early Hex type unit.

I have a 1962 CVA Lucas replacement parts book that has the 54412348 as the replacement unit. That was when they changed production to the cheaper cup chamber with the small threaded line fitting.

Remember or note that in real life all engine are not identical. If the cam timing, carb mixture, or exhaust etc is slightly off, it can change the vacuum that effects vacuum unit advance.

I've re-plotted that relative vacuum data, drawn a putative curve that satisfies all the conditions, and added the corresponding absolute vacuum points in green at the top



Based on all of that, this is what I now believe Jaguar - had they had a 123 or a MegaJolt - intended the advance behaviour to look like.

I am NOT suggesting that is what is best with today's petrol, but it is going to be my starting point.



Anyone like to check my advance figures ?

[mgcjag]

Hi Rory...figures look correct to me......so you have actually taken the approx mid point range of figures as shown for vacume in the distributor section of the Jag service manual........One point to note for anyone programing a 123 is that all info Rory has given is related to a 3.8 with DMBZ6A dizzy........however if you have a 4.2 that used a 22D6 dizzy the vacume figures shown in the service manual are exactly the same as the DMBZ...but the centrifugal advance is different......so just re calculate useing 22D6 centrifugal figures.......Steve

[rfs1957]

Yes, I just tried to plot a credible curve that lay within the vacuum spec boundaries.

I'm still not happy with the point at which I'm exposing the advance to the vacuum effect though.

Originally (just been testing with a vacuum gauge) the ported-vacuum only starts to read at all at around 1700 rpm, and is only at 5" of suck by 2000 rpm.

Whereas the absolute manifold vacuum at that point is at about 30 KPa, which is 20" of relative suck.

So the position of the ported take-off, right next to the butterflies, makes for completely different behaviour.

Difficult to know how and at what point to tie the two, ported and manifold, together.

Currently my map gives 24° of advance at 900rpm - the engine is sweet as a nut, but that wasn't what Jaguar had intended, it should be about 13°.

So a bit stumped right now.

EDIT - I think the question of ported vacuum as opposed to manifold vacuum has a far greater effect on the mapping than I've grasped until now, and puts mapping in this way into question ; so better considered Work In Progress. My apologies if I've given anyone sleepless nights. Will update when I've made more progress.

[Ltrom]

People here on the other side of the pond tend to set to curve 2 after setting dynamic timing to 10 degrees. That gives a good all around performance and fuel economy.

[malcolm]

People here on the other side of the pond tend to set to curve 2 after setting dynamic timing to 10 degrees. That gives a good all around performance and fuel economy.

Thats the curve I use on the non-tune 123 with the 16 selectable curves, although much of the above discussions refer to people using the fully programmable version

[John ball]

Did anyone ever come to a conclusion about the comments from Rory about manifold or ported vacuum ?
I have tried sending a PM but no answer. I would be very interested to hear a conclusion for programming the vacuum setting on my 123 Bluetooth version.

[mgcjag]

Hi John.......Rory isnt actuall useing a 123.....he,s on edis.........i think its very much a matter of measureing the vacume on your own set up...cant remember what set up you have or if your carbs are ported by the butterfly.....best to just measure the advance your getting with a timing lamp and if you have guages measure your vacume.......steve

[John ball]

Two years later !
I just wanted to say that I have been using a curve very similar to to Rory’s hard work converting the figures to Absolute on the vacuum curve for the Bluetooth 123 system. Although my car is not either a 3.8 or 4.2 E type, but an XK120FHC with a slightly tuned engine - higher lift cams, 2xHD8’s, 9/1 pistons. The car has never gone so well, so I suggest Rory’s doubts about Ported vacuum or not may be unfounded. You will note the static timing is set at TDC with the advance curve starting at 10 BTDC, not the static set at 10 BTDC then the curve taking that into consideration with lower settings. It is not exactly the Original Jag curve and vacuum as suggested on page 338 of the Bentley book, but seems to work very well with lots of pull and no pinking.

[rfs1957]

It speaks volumes for The Decline Of Man that, only two years on, most of what I wrote is completely unintelligible to me

I am immensely reassured, therefore, that John has found it to be perhaps of some help !

It doesn’t help that my (finished) engine has been under a dust sheet for that time, as I await the news that the shell is ready for collection from the painters in the UK, over a year since I trailered it north.

Maybe I can now post questions to help me understand my own élucubrations ?

Decline indeed …….

[Tom W]

John, how easy is the car to start with the static timing set at TDC and all the timing at idle added in the app?

The reason I ask is I’ve recently swapped my pre-programmed 123 for the Bluetooth tube version and initially I set it at TDC and added the idle timing value into the app. The problem is the car now won’t idle until it’s warm. My car has Webers and so no choke or fast idle setting. From cold it wants to idle around 500rpm or lower. With around 12 degrees advance it will do this. However, below 500rpm it appears the 123 applies no advance, despite what is programmed in the app. So with the distributor set at TDC I get no advance, and with 0 degrees advance the car won’t run. Looking at the app with the car running it appears any rpm below 500 is displayed as 500, but it applies no advance. If I rev it it looks like it’s behaving properly.

The initial solution is to set the distributor at the minimum advance figure required (I picked 10 degrees) and take 10 degrees off the mechanical advance values in the app. Now whatever the distributor does, I don’t drop below 10 degrees and the car will idle. This isn’t ideal though as now the ‘crankshaft degrees’ reading on the app isn’t total crankshaft degrees, it’s crankshaft degrees beyond 10 degrees.

I guess you don’t encounter this as with SUs the fast idle mechanism raises the RPM into the range when the 123 works properly. It worked properly when fitted on a friend’s car with Strombergs, just seems like it doesn’t work so well with Webers. Disappointing, as I specifically wanted to upgrade to the tune version to draw a suitable curve for the Webers.

I guess in the long run it doesn’t matter. Once I’ve programmed the curve on a rolling road, I’ll likely not look at the app again, but it’s frustrating the thing can’t be set up like it’s supposed to.

[John ball]

Tom, I have tried it both ways. Setting it at TDC and programming app to start at 10, also as you suggested, setting at 10 BTDC then lower curves. I cannot see any difference in the starting, it seems the same when I have tried both. But interestingly the 123 instructions say set at 10 BTDC, but when I emailed 123 in Holland they asked why not set at TDC and let app do the full curve. It is interesting that on many prewar cars you retarded the timing to start on the steering wheel lever. My car is in fact fairly had to start, it does not have your E type pull down jets vis choke lever. I have a manual override electric choke hisser with six individual tubes into the bottom of the manifold. Also I have only two HD8’s not your three. Also two H6’s were standard on XK120’s being 1 3/4” not 2”. - so, yes my engine is a bit of mongrel.

Certainly, when the engine is hot, it idles smoothly at 700rpm and with Rory’s curves or I should say my slightly more conservative version it pulls better than ever. It will not idle smoothly when off the choke before hot, but I think this is due to the cams I have that are slightly higher lift and possibly more duration ???

It would be interesting to know why the graphs start at 500rpm, maybe I should contact 123 and ask the question ? I have a modern hi torque starter, but can’t really see what speed it turn over the engine.

It would be interesting to hear from E type owners who may have tried Rory’s curves or even try mine.

The reason I am using similar curve to the E type is because my 3.4 in its state of tune is nearer to a 3.8 and both of those are freer revving than a 4.2.

My rolling road tuners tell me that they get many E types to tune both for road and racing. In standard form they never produce the original stated by Jaguar 265bhp actually nearer 220bhp.

Let’s keep this discussion, opinions, arguments going as the 123 system does need understanding being quite different to a Lucas distributor, also the 123 instructions and videos are not really the most informative!