Engine tuning DIY style - ignition

[Phil H]

Now I'm pretty conversant with the general theory of engine tuning (assume here a random selection of parts assembled from those eBay purchases we all have gathering dust..or is it just me then?) - whilst "perfect" combustion happens at a 14.7:1 air/fuel ratio, I believe that best mpg for cruising is approaching the 16/17:1 ratio and max power is normally around the 12.5/13:1 area.

One bit I've not quite got my head around is ignition timing..or rather "how" to do it yourself. I know what a knock sensor is and what it detects and if your timing is too early you get pre-detonation (bad - especially with forced induction) and if it's too late, the main "bang" of combustion is still happening whilst the piston is already on its merry way down the bore, so the energy isn't being fully utilised but how does the average DIY tuner find the right advance settings without sitting on a dyno for hours? Can you utilise, say, EGT as your "feedback" in the same way a wideband Lambda sensor tells you whats going on with the mixture?

I realise there isn't a "one size fits all" answer - combustion chamber shape seems to be a deciding factor for starters and aside from trying to find what someone else did, I'm not sure how to get to the "best" settings for an ignition map. Without knowing what I'm aiming for, will I really be able to make it better than a clockwork (OEM) dizzy? Surely that's the point of going mapped ignition - certainly is in my case.

[slater]

You can do it on a dyno in 5mins with an ecu. You cant really do it properly without a dyno tho i'm afraid. You can kind of guess by the feel of the engine on the road but it's not going to get it spot on and it's not as easy to keep an eye in what's going on and avoid making mistakes.

Generally tho on a dyno you just adjust it and see what happens. On an NA engine it will generally make best power before it gets into the danger zone. On a turbo it might not be quite as simple as that.

[ChasR]

Sept 30, 2019 20:03:33 GMT slater said:

You can do it on a dyno in 5mins with an ecu. You cant really do it properly without a dyno tho i'm afraid. You can kind of guess by the feel of the engine on the road but it's not going to get it spot on and it's not as easy to keep an eye in what's going on and avoid making mistakes.

Generally tho on a dyno you just adjust it and see what happens. On an NA engine it will generally make best power before it gets into the danger zone. On a turbo it might not be quite as simple as that.

I agree with you, but a few people here I mentioned this to last time will put us in the nuns and kittens category, claiming a man off the internet said to do it and to listen out for knock. I wonder if said man would ever pay for the damage...

But, back to the OP.

Ignition timing plays more of a role in power than the mixture. On a few cars, going from 11 to 12.5:1 won't make alot of difference. It's talked about more now as AFR gauges have hit the scene, and that makes people seem they know what they are on about. Make no mistake ; fuel still plays an important role.

Ignition though is interesting. Engine Masters did a test on a 460BHP engine. At full whack, total timing made the following difference:
-11 : 320 complete with flames from the exhaust!
-21 : 409
-31: 461
-35: 466

Thus proving there is a margin. As stated, the margin is much much smaller on a turbo engine, with the problems having more serious implications.

Most dizzys on cars I've bought don't tend to advance as much as they should. Cars can also be very hard to start with too much initial timing as well. This is the main drawback of a normal dizzy. You can put in springs to compensate for this, but it is a hell of alot of fiddling and trial and error. A dizzyless system like Nodiz or Megajolt can do this easier.

Also, getting the rate at which the dizzy goes all in will affect things, as shown above.


On the 2CV I did the following:
-Starting: 6 degrees : This cured an issue with the cheapo starter motor the last guy fitted
-Idle/Initial: 14 or 15 degrees C. It seemed to work very well here
-Total : 32 degrees C.

Compared to when the last guy simply turned the dizzy to 13 intiial and then it was advancing all out at 25, it did drive and start better for it. And this is why a fresh dizzy can often make an engine seem much more alive.

If you want to know more about timing, I recommend you read Des Hammil's Ignition book. He claims it can't be used to do ECU systems, but the principles are much the same. Instead of filing (or building up) weights, and changing springs, you let your phone/laptop dial that in.

[Phil H]

Sept 30, 2019 20:03:33 GMT slater said:

You can do it on a dyno in 5mins with an ecu.

If you know of a dyno that can do a Nodiz/Megajolt map in 5 minutes, give me their number and I’ll book it in straight away!

[ChasR]

I think we know what he meant. It will be significantly quicker and result in a better drive for less effort.

[slater]

I probably can do it in 5mins if everything goes well but that doesnt include getting it on and off the dyno and all the other stuff that goes with it.. Still, an hour session on a rolling road should be enough to just do the ignition map, it shouldn't cost too much in the grand scheme of building cars..

[Phil H]

Very true..the question was more "can you do it without a dyno" - sounds like the answer is "no".

I'll rephrase it - "can you get close without a dyno and if so, how do you know you're close"?

[slater]

Realistically 'it feels faster' is the only way I can see you knowing really. I mean you could use a knock sensor and data log it all and try to back it off from where the sensor is telling you its too advanced. I'm not sure I really like that idea tho! It's a bit like advancing until it 'pinks' but like I've said before it usually will loose power as you over advance it before you hear it pink or sense knock.

The big thing the dyno let's you do is brake the engine at whatever load/rpm site you want. You cant do that on the road so you would have to data log it as you do acceleration runs at different throttle openings I guess and try and look back at what the data is saying.

You mention EGT but while it will change with advance I cant see how you could determine what was 'optimal egt' as such

[stealthstylz]

The old way to do it was to learn how to read the spark plugs - the ground strap was the main thing to look at with ignition timing, at the perfect amount of advance you should get the blue-ing/darkening on the earth strap somewhere around the 90 degree bend in it. It's a ball ache to do though as you have to shut the engine straight off after your acceleration run to get a correct reading. I use it as a diagnostic sample to show customers why they've blown their engines up.

[pxr5]

With a mic' strapped to the block & head phones, along with a good A/F gauge you can dial in a good tune on the road,

[melle]

I found Des Hammill's book How to Build & Power Tune Distributor-type Ignition Systems very helpful when I wanted to learn more about ignition.

[ChasR]

Oct 3, 2019 19:29:52 GMT pxr5 said:

With a mic' strapped to the block & head phones, along with a good A/F gauge you can dial in a good tune on the road,

Most tuners won't need to hear it pinking; Generally, the power will stop increasing over roughly a 5 degree window. That window is alot less with forced induction cars.

That and some engines don't take pinging well at all. Scoobys are known for their flat fours being ruined. Do the pinking method on an Wankel engine and the motor will be history, as a few here know and have told the tale

Des Hammil's book on this is very good.