gsl-se addict

iTrader: (3)

Nope. Didn't get it then. I usually answer PMs right away. So if I am online and haven't answered, may mean that I never got it. Glad you got your questions answered, though. If you have any others, let me know.

Whanrow

a few questions:

why did I have to move the trailing wires to the leading side of the dizzy; and:
do I have to set my timing now?
Quote:
Originally Posted by Whanrow
Yes somebody please explain both these things. I've always tried to understand them too. I know with conversion to direct fire you have to advance the timing around 10 degrees, but don't know why, not do I know why you switch to the leading terminal on the dizzy.

I left my timing stock and I honestly don't have the answer as to why you must switch to the leading side. But I know at first I left them stock and it sputtered and choked out anywhere above ~2500 rpm. I'd like that question answered too. The timing should remain normal however, so I dont think you need to advance it 10 degrees. Never heard that before...

The 3 comments above were in the other TT thread, but nobody answered them. I'm hoping somebody can explain....maybe gsl-se addict?

gsl-se addict

iTrader: (3)

1. You don't have to move them over to the leading side, but it does help. If you look inside the distributor cap, you will see the the center post (L) has a carbon button that contacts the center of the rotor. So, normally the spark from the leading would travel down this "carbon button" to the center of the rotor and then jump from the rotor to either the L1 or L2 output. On the trailing side, there isn't the carbon button so the spark has to jump to the rotor from the T terminal and then jump again to either T1 or T2. Moving the wires just lets the spark have to travel one less jump. That way you get a bit more spark to the trailing plugs. The difference is probably not much, but you might as well take advantage of it.

2. The timing change is only for the TT version. The normal 2GCDFIS just uses the 2nd gen coil in place of the 1st gen. It uses all the 1st gen igniton parts to fire it. It is just like replacing the coil with an aftermarket one. The timing doesn't change. On the TT, we are using the 1st gen ignition to fire the 2nd gen igntior and coil. There is a bit of delay in the circuit (very small) and the rest of the delay is in the 2nd gen igntior itself. This causes the timing to retard 8-10*. The advacing of the timing really is to reset timing back to where it was. It only takes a minute to do. You can set it to factory timing or run a little more advance if you like.

Hope this clears it up for you. If you have any other questions, feel free to ask.

Kent

Whanrow

Good answer. The gap jumping is simple common sense, however I couldn't figure out the timing change. I didn't realize there was ANY delay in the ignitor or circuitry. So the delay equates to approx. 10 degrees on a running engine.

I guess the $128,000 question is how come the 1st gen ignitor has no delay, and the 2nd gen does?

gsl-se addict - the above question is rhetorical. It doesn't really matter, its just kind of strange that the 2nd gen has a delay and the 1st gen doesn't....

gsl-se addict

iTrader: (3)

Well, I don't know that the 1st gen ignitor has no delay. When you set the timing any delay between the VR sensor and the ignitor gets canceled out. In this case, however, we are taking a signal that has already been timed. It then goes through the circuit and then the 2nd gen ignitor. I am sure the circuit delay is very small. I know the spec on some of the stuff in the circuit has delay in the 100 nanosecond region.

So anyway, there is some delay, which then retards the timing a bit. You just have to reset your timing. Takes only a couple minutes. It is a good idea to check timing every once in awhile anyway. It is something that many people ignore.

So, where do I pick up my money?

Whanrow

That's true. Since the advent of electronic ignitions, most people set em and forget em. I think a lot of people don't even own a timing light anymore.

When I pay you for the TT board! lol

teejs

Re:
"Okay, guys. I just got report of another possible coil failure. jayroc, one of the first people that tested the setup, let me know that his 2nd gen ignitor died the other day. He said that he noticed it when starting up and not during a hard run. This has me a bit worried. One failure may be coincidence, but having two failures suggests we may be driving the ignitor too hard. He was running the variable pulsewidth mode only where the pulsewidth is determined by the J-109."
GSL-SE Addict....
Do you have any info regarding where the coil-packs were/are mounted and if they used a seperate ground wire or strap for the Coil-Pack? Preferably using the tried and true toothed washers and such....Just bolting to a frame-type ground is 'ok' ....but just ok....I'd prefer to overdo as the ignition system in operating condition makes it so much easier to get home than the Flintstone method.
Tom

Zyrano

Actually, I don't know what I did wrong, but I blew my second coil. It's funny, because both coils blew when the car was idling (one while i was sitting in traffic, the other while i was waiting on grid between runs at a autox). I think for now I'm switching back to the stock setup. Two thoughts first though, both my coils were from junkyard cars, and I did not take the transistor apart and redo the heat sink, so it's possible that they are overheating? The second thought is that maybe we can hook up a scope and compare the signal between a stock second gen sensor with the regular cas, and the tt.

Alvin

gsl-se addict

iTrader: (3)

Sorry to hear that, Alvin. jayroc's died when he when he went to start it. Maybe it actually died the night before just before shutting off. jayroc had an additional ground wire connected to his, but he did say that he noticed the crimp was loose. I think that the maybe the dwell is too long on from the J-109s at idle/low rpm. This is corrected in the fixed pulsewidth version, but that version tends to not have as smooth of an idle. The strange thing is that in the fixed PW version, my coil stayed cool during bench testing, but got hot when in the car (even at idle). I then thought that maybe grounding was to blame, but my grounding seemed fine from the testing I did.

I have been wanting to compare to the 2nd gen ECU/CAS for quite some time, but haven't found anyone with a scope and a stock 2nd gen to test.

I am open to any suggestions you guys may have. I will do my best to get this corrected.

If we get this figured out and corrected, I will do something to make this right with you guys. Maybe the updated circuit at cost? I want to stand behind the work. I don't want anyone feeling cheated.

Thanks.

Kent

Zyrano

Kent, Don't worry about it, i'm not blaming anyone on this, it's more a a science experiement from the beginning, and i understood the risk and all that's involved when i signed up. If anything, i appreicate you doing all of the work to begin with. I'll see if i can find anyone with a scope and a second gen. But for now, i'm going to hook my stock stuff back up. At this point though I'm also considering an msd box as well, so we'll see what happens...

Alvin

Zyrano

oh yeah, the nice thing was i had everything ready to switch back to stock, and i did it on the track in less then 5 minutes, I didn't have a timing light with me, but the distributer had a mark on where it used to be set to, so i just moved it there... no big deal!

Alvin

Kentetsu

Just a quick note here. My first attempt at the fixed p/w circuit resulted in a dead coil too. However, I tracked my failure down to a wiring error that I had made during the install. I ran the sytem for about a month after that with no issues prior to upgrading to the autoswitching version, which is still running strong. Hopefully this is just related to the use of 20 year old coil packs, and not the tt unit. Good luck guys!

ratboy

iTrader: (1)

I've been running the TT for about 2 months on my daily driver with a coilpack from a parts car (unknown age) with no problems. I'm w/ kentetsu and hoping that this just normal failures due to old coils because I still love this thing! Sounds like both failures were with the oringinal or manually switched versions..

gsl-se addict

iTrader: (3)

I hope it is just the coil packs, but there is getting to be enough to cause concern. Ken: on yours, the coil itself failed, correct? These others appear to be ignitor failures.

I think I do have an idea to fix this. The IC in the board to set the fixed pulsewidth has another half to it that isn't used. I bet we could use this to set a maximum pulsewidth. That way the actual pulsewidth would always be between that value and the value used in the fixed pulsewidth mode. All we need to do is to figure out what value this maximum should be. To set the value, we just need to add another capacitor and resistor to the circuit.

If we knew what the FC ECU did, that would help. jayroc posted his problem on the megasquirt sub-forum wondering if it was his wiring. The mod on there says he uses a 2.1 ms pw from his MegaSquirt at idle on his FC, This is basically what our fixed PW circuit delivers.

Based on the numbers Jeff20B posted early in this thread, I believe the J-109 puts out about a 6.5 ms PW at idle. It then drops off with rpm. Must be too hard on the ignitor under these conditions. Maybe set the range from like 2.2 ms to 4 ms or something. I will look this weekend to make sure the design would work.

This should give the benefits we have now, without the additional load on the ignitor/coil. I will definitely offer all you guys a new kit at whatever my cost is. If I was in the US, I would even build you a new one. I know it is pretty much an experimental design, and I thought it was tested well, but things happen. I guess this is how improvements are made.

Kent

Kentetsu

Kent- I was never able to verify whether it was the coil or the ignitor. I never had an issue with the TT itself though. Hope this helps...

H4Inf

Hi Kent and everyone, I have just spent 4 hours reading this thread - very inspirational stuff! When you get a chance Kent, if you could design a circuit and board/parts list for a unit incorporating both a minimum and maximum pulse width, a friend and I would love to build it and test it out here in Australia. Our local forums have no mention of any sort of mod like this, we're very keen to test it out! (I think people here tend to run aftermarket ECU with DirectFire - I'd rather save the $$). I've got a 13B N/A and friend has a 12A.

Great work everyone that contributed! Awesome stuff All the best with France Kent, I look forward to hearing from you!!

Happy New Year =)

gsl-se addict

iTrader: (3)

Thanks, glad you like the idea. I could probably post something up on the design. I am going to be doing a group buy here soon for kits/assembled units to build the newest design. These will be professionally built boards. I am willing to ship to Aus. Maybe some of you guys can get a group together and I can send all the kits together to save on shipping. As far as I know, no other forum has came up with a mod like this one. It works very well. I orignally didn't think it would make much difference, until I tried it out. I have the design with the max/min pulsewidth figured out. I don't know what we should limit the max pulsewidth to be. I am thinking like 4ms would be good. The circuit is built so you can change this value by swaping out 1 resistor.

Kent

H4Inf

We might just go with your autoswitching circuit as posted earlier.. and see how it goes Is is possible to measure the pulse widths from the stock system at low rpm so we can work out a reasonable time to limit it at?

Based on results (hopefully get some dyno figures too), we will recommend it to people on our local forums then get back to you about a number to order ;-)

Many thanks!!

Paul.

gsl-se addict

iTrader: (3)

Hey Paul,

You might want to give the simplified version a try if you just want to see how you like it.

https://www.rx7club.com/1st-gen-archive-71/ignition-2gcdfis-w-tt-transistor-trick-write-up-simplified-design-484176/

The auto-switching version is mostly the same, but has stronger performance in the top end. The problem is that there have been a couple ignitor failures with this setup (current auto-switching has the same problem). The 1st gen dizzy/J-109 uses too long of a pulse for the 2nd gen coil. The design for the GB should eliminate this problem.

Some measurements on a 2nd gen car would be helpful. That way we know what it sees stock so we know what a safe limit would be. I think without further measurement, I will go with a 4ms pulse as the max. I think the J-109 uses around 6ms in the lower RPM range. I have tested a contant 2.2ms, and it was a little weak on in the low end. I think around 4ms would be a good compromise. As I said, 1 resistor changes this, so you could always experiment to see what works best. We want it long enough to get the stongest spark, but not so long that we strain the ignitor.

If you have any questions, let me know.

Kent

H4Inf

Hi Kent,

We're willing to give the autoswitcher a go, just wondering which resistor would need to be changed, and to what value to achieve a 4ms maximum?

I don't have the equipment to measure on hand so perhaps someone else would be able to get a measurement some time - I think 4ms would be a safe test to begin with.

We're also thinking of putting together a simple heatsink made up of 3 or 4 pieces of aluminium rectangles bolted ot the base of the 2nd gen coil, then with the edges that stick out the ends, bend the different sheets out to different angles to help remove some of the heat generated.

Cheers,

Paul.

gsl-se addict

iTrader: (3)

The current auto-switcher posted doesn't allow for setting a max pulsewidth (only a minimum). I was referring to the newer design. I can probably post something later today about it. It is nearly the same as the current auto-switcher, though. One thing I am trying to figure out is that we need an AND gate and an OR gate with this setup. Most integrated circuits have one or the other, but usually not both. I will see if I can find an IC that contains both (easiest option). If not, we can build the gates from transistors.

The heatsink is a good idea. Even if it doesn't really need it, electronic components almost always benefit from running cooler (longer life and so on).

Quail Squasher

Hi Kent, firstly this is fantastic stuff! Judging by the positive responses from those who have trialed various adaptions of the original Transistor Trick, this mod looks very promising. This whole thread has been quite educational and inspirational and worth the ~4hr read. As Paul has suggested, we would be very interested to test the latest incarnation of your circuit on a dyno! (comparo with stock ignition). The wider ausrotary community doesnt seem to be aware of these highly promising developments, so to get some hard numbers and report any useful findings should generate some excitement in the local scene, as well as add some numbers to a group buy scheme.

Just fantastic!

Rob

gsl-se addict

iTrader: (3)

Thanks, Rob. I would be interested in a dyno comparision as well. I will try to post up the lastest design tonight sometime. It would be cool to get some Aus guys trying it. You guys always do such a great job with your cars. Welcome to the forum.

Kent

Kentetsu

Dyno testing.....Yummm!

gsl-se addict

iTrader: (3)

Okay, guys. I got it all figured out. All the parts are the same as what was listed for the auto-switching design. You will just need 1 more 0.1 uF cap and a 40kOhm resistor (for 4 ms max width). The time is set by the cap x resistance, so 0.1 x resistance (in K Ohms) = pulsewidth in milliseconds. You may need to go 1% or 2% tolerance on the resistor to get a value near what you want. I will try to post the design when I get a chance.

Also:

I have decided to do group buy for the new boards through www.rotaryx.com . It is a new auction site for rotary related stuff. This way we don't have to wait longer for GB approval. I will post an auction real soon.

Guys that have purchased older circuits from me: pm me if you want the latest/greatest design. I will give you guys a nice discount. I will probably only be able to hook you guys up with a kit as I am not in the US to do the assembly myself. I want you guys to have something more reliable. I don't want any more killed ignitors if we can avoid it.

I plan to take orders until probably the end of the month for this batch. I will then oder everything and we can start the distribution. That way we can get you guys the circuits by Feb or early March. In time for taking the 7s out of storage.

Thanks.

Kent