123 Ignition's 123/Tune Advanced Curves

[fiat142]

Hello,

Has anyone set up custom curves for the 123 Ignition distributor, and have a good starting point?

I've been searching all over the internet looking for a good set of curves for the 123/Tune bluetooth distributor. I got the bluetooth one because I liked the idea of the kill switch functionality, however it looks like it isn't really loaded up with a working curve or vacuum advance curve out of the box.

So, at the moment my engine runs, but I'm battling some intake backfiring

The best I've been able to find so far is off the non-bluetooth version's product page (http://www.123ignition.nl/product.phtml?id=29) where they describe (roughly) what it might look like:



I have emailed the company, but I figure this is probably a more veteran source of info - any help is much appreciated!

[18Fiatsandcounting]

I can't really answer your question as all the cars that I have any experience with were just the old school centrifugal advance. But, if it's any help, the static timing was usually set to 10 degrees BTDC at idle up to 1000 rpm, at which point the centrifugal advance took over to advance in a linear fashion from 1000 to 3500 rpm, where it leveled out at around 30 degrees BTDC (10 static plus 20 centrifugal). I usually set the timing so that the maximum total advance was around 35 BTDC, which would be around 15 BTDC at idle. Opinions vary, as does what's best for each model year.

Optimal ignition timing for a given engine speed is a combination of compression ratio, gasoline, engine load, carburetion system, intake/exhaust timing, shape of the combustion chamber, spark plug heat range, ambient temperature and humidity, desired power and fuel economy, emissions, engine age, and of course who is in the playoffs. In other words, it's usually a compromise.

-Bryan

[SteinOnkel]

Can't really figure out the ideal advance map without a chassis dyno. Main reason why these fancy ignitions don't really make sense, unless you go all the way and go to programmable EFI.

I would start by writing a map that's more of a curve based on what Bryan was saying. In fact, the haynes manuals for our cars have ignition curves in them. Basically the bins in one column would all be the same number. So from left to right you would start with 10 (or whatever you want idle to be), then go "all in" by xxxx rpm and just hold that and interpolate the numbers in between. I'm assuming the 123Junk software gives you the option to interpolate? That'll give you the same advance as with a distributor, because you are essentially ignoring the load axis. As you drive around you can flesh out and vary the numbers based on load. At light throttle cruising you can run more advance than at full load, for example.

Set your steps in the load/rpm axis to something that makes sense. You want more resolution down low, less so up top.

That's how I did it on my efi. It's all a moot point, though. I've got a few things to sort out and then it's time to drop some money and get it dyno tuned. Only way to truly unlock the engine's full potential.

Backfiring through the intake means you are too far advanced. The curves it comes with shouldn't be aggressive enough to go with that right out of the box, though. You sure your commanded advance is the actual advance on the engine? Check, double check and tripple check this with a timing light.

Post up what you have now, I'll give you some pointers.

[chrisg]

have you asked on Mirafiori? I think there are some 123 operators there & I do think the advance curve graphs are available in general to draw from also. I have the same distributor, but alas garage construction at home put it's implementation on about a 12 month delay & I've not had a chance to play with it yet.

[fiat142]

Thanks everyone for the great info!

I’ll let you know how it goes and where I end up after my next round of testing.

[fiat142]

So, after much mucking around I have a bit of a status update.

First off: a bit of annoying info on my setup… The PO removed the timing belt cover and I don't have any sort of timing mark bracket on my engine. Therefore, figuring out timing is a bit more annoying for me than what most people will probably deal with.

Here's what I did:

1) Verification of Mechanical Timing:
Removed the distributor, set the engine to TDC with the tried & true straw-down-cylinder-1 method. This was the like 5th time doing this but feels good to verify everything again, and having the engine at TDC is required to follow the distributor install instructions.

- Result: Cam holes line up perfectly with little pins at the top of the engine. The crank pulley mark is just turned to the point where the mark touches where the accessories belt meets up with it (about 2:15 o-clock). The accessory pulley holes are lined up with the belt tightener spring bolt. In other words I think my timing belt and TDC look correct.

2) Install Distributor & Get to 0°
Next I re-installed the distributor according to it's instructions – put it in, rotate counter clockwise until light turns on. This sets the distributor up as ready to fire on #4 but at 0° advance. Just perfectly at TDC...

- Result: I verified that the dist was setup to fire #4 right at TDC by using a timing light off the cams and the starter motor. Rechecked everything and put all the plugs back in, etc.

3a) Set Timing - Attempt 1
Now came time to actually set the timing. After setting that curve up in the app (based on Bryan's numbers mentioned above) I theoretically had it ready to go at 10° BTDC up to 1000rpm, and then linearly to 30° at 3500rpm.

- Result: Engine ran very badly (as you can imagine) and did some backfiring through the intake. Seemed to be the same as last week and not have any advance at all while at idle. ME: ಠ_ಠ

Poking around the UI of the app I was noticing that on the dashboard is a dial indicator of "Crankshaft Advance" that was zeroed out at idle, however did jump up when the engine was revved. Oddly enough that dial read about 10° off from what I'd enter into the curve. For example at 1000 it would read 0 and at 3500rpm it would read about 20. (In hindsight the default curve of the distributor was very close to that of the fiat, so I'd bet I was just seeing that default curve in action. More on that thought in next steps section below...)

3b) Set Timing - Attempt 2
Playing around with things I quickly noticed that if I went and physically turned the distributor counter clockwise the Crankshaft Advance dial (attached below) would read out the increase... And, of course, the engine would be running a lot better.



So at that point I just went all old school on it, and turned the distributor until the engine was running decently. This happened to move the Crankshaft Advance readout in the app to around 8-10°. Not very scientific way to set advance of my fancy fully bluetooth distributor, but what ya gonna do.

- Result: The distributor seems to be set at around 10° BTDC physically instead of via software. Looking at the timing light against the cams the little dots are indeed about 10° off of the mark, and the crankshaft pulley is pointed more like 1 o-clock (which, doesn't really help me to know what's going on since I don't have any physical timing marks to go off of, but is theoretically in the right ballpark according to the diagrams and pictures/videos I've seen).

4) Dialing It In
OK, so at that point things seem good at idle. I re-edited the curve to remove the 10° advance from 0-1000rpm (attached below). Let the engine warm up fully and played with the carb a bit so idle speed and screw were reasonable.



- Result: At this point I theoretically should have a decent re-creation of just a mechanical distributor. My old one didn't have a vacuum advance on it, so I should be matching what I had before. Indeed the engine sounds very close to before this setup, and I can see the curve in action using the dial indicator in the app.

5) Closing Thoughts & Next Steps
A baseline seems to have been established. The engine idles nicely at 800 - 900 with 20lbs of fairly steady vacuum pressure on the manifold. Engine sounds quieter and smoother than it did last year on the old carb/dist setup. Still has a bit of an odd cadence to it, but I’ve verified all plugs working and all cylinders have similar pressure (~90). I think it might just be the particular lope of these engines. (If anyone could point me at a video of a perfectly tuned engine that’d be helpful. This one is pretty darn close to how mine currently sounds: https://www.youtube.com/watch?v=nt0dweedyag)

We have some freak snowy weather happening in Seattle so I can’t take it for a test drive to see how it does under load, but I’m hopeful that it's in the ballpark now.

That said, I kinda feel like I should go back to 3A and try that setup again. I’m a software dev, so my spidey sense is tingling that in order for the CrankShaft Advance dial in the UI of the app to work at all there must be a sensor in the device that reads the position of the arm when the car first powers on, and then compares that angle to what is happening with it every few milliseconds. In other words the app is showing me the results of a sensor and not just what it thinks it should be doing based on the curve. This really has to be the case because it changed the reading when I physically messed with it, and right now I know there’s 10° in the setup but the dial in the app still reads at 0 on idle and then about 10° off from the curve value should be when playing in the RPM range. Plus, if you were making a decent product you’d have to create a feedback loop within the device for it to know if it’s executing the curve ok, and for all your internal testing and error logging. Seems to be a decent product, so I bet that exists.

Based on the instructions of the device and how it’s “supposed” to work I think I did something wrong when editing/entering in the initial curve into the app. I think I somehow didn’t actually apply it to be used by the distributor... I probably should have tried verifying the execution of my programmed curve rather physically messing with things. Was just so satisfying taking the shortcut

So, I might end up trying again unless someone has some insight into that topic? Does the 123 software have any quirks in executing that curve or should it really just work as advertised?

Next missing piece of the puzzle that maybe someone has some ideas on:
Any clues how the vacuum advances messes with everything? I’ve attached the curve that was pre-programmed in just for reference. I'm not sure what distributor its emulating there, but maybe ours does something similar?

[18Fiatsandcounting]

Tim,
I gotta study your post a few more times to let it sink in, but here are a few thoughts:

1) Ignition timing depends on a bunch of different factors, so the "advance curve" is just meant to get you in the ballpark for any combination of engine operating parameters.

2) Advancing the ignition simply has to do with how fast gas burns. When the engine is spinning faster, you need to give it the spark somewhat sooner so that when the fuel really starts burning (expanding), you're at the top of the stroke or somewhere very slightly after. Google "brake mean effective pressure" (BMEP) or the like.

3) My (limited) understanding is that vacuum advance (or vacuum retard) has much more to do with emissions than it does engine power production. Most of my cars have no vacuum unit, and I simply rely on centrifugal (mechanical) advance in the distributor.

4) The exhaust note in that YouTube video you posted sure does sound like the drum intro for Van Halen's "Hot for Teacher".... I'm just saying... Rock on, dude!

-Bryan

[fiat142]

More info to report back... Snow melted so did some driving

First off: yes, I totally just didn't load in the curve I thought I had into the dist during step 3A of my post. Doh!

I've redone the install with it at 0° and am using software fully to control the timing (woot!). Works like a charm. I can see the actual advance when in idle now, and the app even allows for playing with it by 1° increments while underway. Nice surprise feature.

The car's timing sounds much better while driving around now. No backfiring through the intake and everything seems to work nicely through the rev range. I didn't want to brave any big hills after the snow to see about pinging, however a quick e-brake test didn't raise eyebrows.

Lastly, I've exchanged a few emails with Ed over at https://123ignitionusa.com, and he's independently come up with a very similar curve to what we arrived at here. He provided a vacuum advance curve as well so I have a starting place now on that end of things.

Below is what he provided and what I'm going to start with tomorrow (snow depending). I'll let everyone know how this next iteration of it goes.

Tune #2:

[SteinOnkel]

Pretty sure the youtube video was supposed to show off some hot cams. That's why it's so lopey at idle.

It should be much smoother on a stock setup.

The MAP advance makes sense to me, as long as it is adding 10*, not setting absolute advance to 10*.

Is there a way for the software to overlay both curves and give you a true map? Why is it in curves anyways, every digital anything is supposed to be a table with an optional 3D view. That's the whole point of electronic ignition...

[ward00]

Thanks for confirming why I decided not to buy this expensive paperweight - without a dyno playing with curves is a recipe for a burnt engine. Why they don't provide at least one of the curves that they do with the non bluetooth version is weird.

[aris64]

I had the same concern and I did exact the same steps setting my 123. I tried both ways, static 10° and automatic up to 26°, totalling to 36° @3600rpm. Also I had to experiment with vacuum curve because the max advance was up to 46°! and I modified the curve so to cut off the extra 10° from vacuum. The other way was static advance to 0° and then put +10° on every point of the curve, but engine was hard to start with 0° static advance! So I went back to the first way. What is your experience? How easy the engine starts at 0°?
Forgot to mention my engine is 125BC000 (1608 cc) with 10.4 calculated CR and use 95RON gasoline.

Thanks
Aris

[18Fiatsandcounting]

From what I understand, the optimal spark advance for any engine is about 36 or 37 degrees BTDC once it is up to speed (3000 rpm or so), and this is based on the combustion velocity of gasoline and piston speeds. In other words, it is the same for all typical engines.

So, the question is: How do you want to advance the spark from idle up to this max at 3000 rpm? Linear? Stepwise? A curve? Opinions differ, and it all depends on whether you want your power to be at 1500 to 3500 rpm, or 3500 rpm and above. For me, I set my advance to be 36 degrees BTDC at 3000 rpm and above, and I pay no attention to what it does below that. After all, I don't give a squat whether my engine is producing 20 HP or 30 HP at 1500 rpm. What I really care about is the power it is producing at 3000 to 6500 rpm, so that's what I focus on. Your mileage may vary.

No more than 36 or 37 degrees BTDC total advance.

-Bryan

[RRoller123]

That is very helpful. I think maybe take the crank pulley to a machine shop and have them accurately mark a small timing notch at 36? Hard to know by just visual estimation.

That being said, the advance at idle makes a huge difference in idle quality, so we can't completely ignore it down there at about 1000?

[18Fiatsandcounting]

That being said, the advance at idle makes a huge difference in idle quality, so we can't completely ignore it down there at about 1000?

I agree, and my recollection is that the engine starts to surge or otherwise not run smoothly if the spark is too far advanced at idle. What you definitely don't want is for the exploding (expanding) fuel to push hard on the piston as it's still on its upward stroke. Reaching maximum pressure exactly at TDC probably isn't good either, as all you're doing with that pressure is pushing straight down on your connecting rods through the bearings to the crankshaft. I used to have a book on engine theory (maybe still do) that mapped out the ideal pressure rise curve in terms of where the piston is in its stroke, which of course is related to crankshaft angle.

But, optimal spark advance at a given speed depends on many variables*, so ignition advance curves are by necessity a compromise. Modern cars that have the spark controlled by computers driven by sensors can do a much better job, but for the stock Fiat DOHC engine, the only tools we have at our disposal are the centrifugal advance and a vacuum advance if it has one. However, the original poster's question was about aftermarket ignition advance systems, and while these can get the advance closer to optimal for a given engine speed, it's tricky without a dyno or the like (as pointed out by other responses).

*Compression ratio, shape of the combustion chamber, A/F mixture, engine load, type of fuel (including if it has ethanol), emission requirements, engine temperature, etc.

-Bryan

[msrichmond]

I assume you are running a carb without ported vacuum, which normally eliminates the ability to run vacuum advance. But the 123 has "gearshift retard" which means when engine vac is > 18 inches, there is no vacuum advance. This should give you the effect of ported vacuum. I would try it, it might improve part-throttle response. It should make no difference at idle or wide open throttle.

p.s. I have the 123 switchable model, so check that the Bluetooth model has the gearshift retard option and how to enable it. I think it does.