1st Generation Specific (1979-1985) 1979-1985 Discussion including performance modifications and technical support sections

Transistor trick for 2GCDFIS.

Thread Tools
 
Search this Thread
 
Old 05-28-06, 03:09 PM
  #976  
Super Moderator

iTrader: (3)
 
gsl-se addict's Avatar
 
Join Date: Jan 2003
Location: Lynchburg, VA
Posts: 5,088
Likes: 0
Received 11 Likes on 11 Posts
Yes, I mean across each IC (pins 8 and 16 on the 4538b and between pins 7 and 14 on the NAND). Probably will use 0.1uF for these. Seems to be a pretty common value used. I tested without them, and everything appears to be fine.

Yeah, I put the diode on the 12v line (stripe pointing towards the TT). I just soldered inline. It will be covered with heatshirk on the final circuits. I think you could maybe use a small ring terimanl to connect to the tab. I stole the one in the pic from the old scope.

I am not sure on the bolt size. I think they are 10mm head, but I am unsure on the threads/length.

Kent
Old 05-28-06, 11:43 PM
  #977  
Banned. I got OWNED!!!
iTrader: (2)
 
dj55b's Avatar
 
Join Date: Oct 2004
Location: London, Ontario
Posts: 6,122
Likes: 0
Received 1 Like on 1 Post
ya most of the stuff under there is 10mm i just need the thread pitch and bolt diameter

Alright so hopefully I can do the testing tomorrow morning supplying that I have everything or that the electronics shop locally has everything that I need in stock which I would assume they would considering how basic they are.
Old 05-29-06, 09:15 AM
  #978  
Banned. I got OWNED!!!
iTrader: (2)
 
dj55b's Avatar
 
Join Date: Oct 2004
Location: London, Ontario
Posts: 6,122
Likes: 0
Received 1 Like on 1 Post
hey kent ... i've been trying to shop around for that 0.33uf cap but all that I could find is the electrolitic type. Should I be using that? Also i'll be picking up the 1N4007 just because they're slightly better and they're only 20 cents anyway.
Old 05-29-06, 09:32 AM
  #979  
Senior Member

 
toxic_d's Avatar
 
Join Date: Mar 2003
Location: NY
Posts: 604
Likes: 0
Received 0 Likes on 0 Posts
Kent how would this mod be applied to the first handbuild switching circuts? Should be easily done right?
Old 05-29-06, 11:18 AM
  #980  
Super Moderator

iTrader: (3)
 
gsl-se addict's Avatar
 
Join Date: Jan 2003
Location: Lynchburg, VA
Posts: 5,088
Likes: 0
Received 11 Likes on 11 Posts
Sam: Electrolytic would be fine.

Kyle: No, it would not be so easy. The previous circuits work correctly, but they don't have the ability to limit the pulse without a lot of modifications. Now, there is another possibility. Based on some numbers that I have seen on the coil chargeing time on the stock 1st gen system and those when the TT is connected, it appears that the J-109 may change its pulsewidth based on resistance. I would have to have someone experiment to find out for sure. You can do this by adjusting R4 (lowering). The problem is that as the resistance lowers, the heat given off by that resistor increases. Currently, R4 is 1k.

P=V^2/R, V is battery voltage (say 14v)

The stock coil is something like 2 Ohms. At 2 Ohms, you would need a resistor rated at more than 100W (has to be mounted outside the box). Now, Paul and Rob were measuring a coil charge time of about 26ms with the TT connected to the J-109. Based on numbers that Jeff20B posted before, I believe the stock setup has a charge time of about 6-7ms (near idle). Perhaps you could use a large power resistor in the 3-4 Ohm range, power rating of 100W. They are somewhat expensive, and would waste a lot of energy, though.

robs_seven on here is running the simple TT and has not experienced heating issues. Turns out that his stock leading was still connected (sparking into the air). This tells me that this change of coil charge time by the J-109 does depend on resistance. Perhaps you could trigger the TT off of the trailing - and run 0 split. Not sure if you want to do that, though.
Old 05-29-06, 11:01 PM
  #981  
Banned. I got OWNED!!!
iTrader: (2)
 
dj55b's Avatar
 
Join Date: Oct 2004
Location: London, Ontario
Posts: 6,122
Likes: 0
Received 1 Like on 1 Post
Hey,

Alright so i've done some testing and some somewhat promising results i guess ... at first i had the 12v tapped into from the coil pack. Ground to the negative lead off the battery. When i cranked it on it sounded a bit better. without adjusting the timing. Check out the timing to my surprise i could actually see the line on the pulley now, I could kind of adjust it to the timing marks on there. By kinda i mean the output signal seemed to be really bouncing back and forth in uniform motion i would say about 20 degrees diference between each bounce back ... also there seems to be quite a pause between each one didn't seem like it was doing things right. Inside the car looking at the volt meter it was bouncing quite a bit at idle. So i decided to tap into the positive lead to the battery instead of the coil and much better signal. Volt meter inside was barely bouncing around now (must be noise from the coil packs?) Took the timing light out again ... seemed like all the pulses were there now without any pausing in the middle. revved it up a couple of time and she seemed to be going pretty good. I then tried to line up the marks on the pulley with the sparking and seemed to notice the same bouncyness on the timing marks. I can't remember wheter it was going to the right or left right now of the pulley. I'm thiking that might be a capacitor charging discharging causing this? I will try to post up some pictures of the setup also i will try to check for any shorts on the circuit board after all the modifications. I let it ran like this trying to find the prob and touched the igniter to see if it was getting hot or not. unfortunatly it was. Now i'm not sure if thats because i was mostly at idle all the time (idleing was around 600rpm so would of produced quite a long signal at idle). This is where my work ends ... the stupid neighbourghs had called the cops on my again :@ stupid a**holes ... anyhow but thats another story ... i will try to do some more testing tomorrow morning.

BTW this is what I have added:
All the capacitor except for that 10uf one which i have a 10nf one in there. r2 is still set at 56k ohms (maybe i will try to change that to a lower value). Let me know if you need to know anything else. Till tomorrwo i shall do moer testing.
Old 05-30-06, 01:42 AM
  #982  
Super Moderator

iTrader: (3)
 
gsl-se addict's Avatar
 
Join Date: Jan 2003
Location: Lynchburg, VA
Posts: 5,088
Likes: 0
Received 11 Likes on 11 Posts
Well, it sounds like good progress. So, you have the diodes in place and capacitors, then? (bypass caps, 10nF on the reg input).

Here is what I would try:
- increase the 10nF to 0.1uF or so. This should help with noise that gets picked up on the power line.

- try about 27k for R2. If you don't have a 27k, use something close or add another 56k in parallel.

- perhaps we need to go lower on R3. The reduction that we did seemed to help a lot. The value of this resistor is not very critical. On the very original version of the TT that a couple of us tested, I used 1k for R3. The reason that I changed it is that this resistor forms a voltage divider with the 1k pull-up resistor. Using 1k for R3 means that the C terminal on the J-109 sees about 7v in the high state (ignitor not firing). I don't think the J-109 cares what this voltage is as it just floats when not firing. I changed it from being 1k to 22k so the voltage would be closer to stock, just in case. Maybe try a 6.8k, 5.6k, or 4.7k. We can work our way lower and see what works.

One question: do you remember the direction that the timing marks moved as you reved? They should have moved towards the passenger side of the car if the timing was advacing properly.

Thanks for the help testing. Don't get yourself into trouble with the police, though.

Kent
Old 05-30-06, 12:26 PM
  #983  
Banned. I got OWNED!!!
iTrader: (2)
 
dj55b's Avatar
 
Join Date: Oct 2004
Location: London, Ontario
Posts: 6,122
Likes: 0
Received 1 Like on 1 Post
Hey,

I'm trying not to get in trouble with the cops but its just that my neighbourgh is just an *** about everything.

Anyhow i'll be going out there today to try things again. Across the voltage regulator I have the 0.33uf cap not the 10nf ... the 10nf is right at the wires ... i will pull it out and change it maybe to a .1uf.

I will also add the 27k instead of the 56k of r2.

As far as r3 goes ... I will try to place a 5.6k in there instead of the 10k that I replaced it with.

I will try to remember which way the mark moves to when reving ... its kinda hard doing stuff all by yourself. I could use an extra pair of hands on my body. Anyhow give me a few hours and i'll prob come back with some results.
Old 05-30-06, 01:17 PM
  #984  
Super Moderator

iTrader: (3)
 
gsl-se addict's Avatar
 
Join Date: Jan 2003
Location: Lynchburg, VA
Posts: 5,088
Likes: 0
Received 11 Likes on 11 Posts
Okay. So the 0.33uF is where I have the capaciotr in the pic, then? The 10nF is probably not needed, but wouldn't hurt anything. Sometimes you see in designs that they will put a larger cap in parallel with a smaller one. The smaller helps with high frequency noise, the large helps with the low frequency stuff.

The other stuff sounds good. Hopefully that will help things. I know it can be difficult to test by yourself. We all appreciate your help, though. I would do it if I could. Good luck.

Kent
Old 05-30-06, 01:20 PM
  #985  
MattG FTW!!!!!

iTrader: (2)
 
MattG's Avatar
 
Join Date: Feb 2005
Location: South Windsor, CT
Posts: 2,733
Likes: 0
Received 0 Likes on 0 Posts
Why they call the cops on you? If I may ask?

Also good job guys keep up the good work.
Old 05-30-06, 03:48 PM
  #986  
Banned. I got OWNED!!!
iTrader: (2)
 
dj55b's Avatar
 
Join Date: Oct 2004
Location: London, Ontario
Posts: 6,122
Likes: 0
Received 1 Like on 1 Post
well i thought that questions didn't need any explaintion the beautiful sound that comes out of our cars of course. Anyhow kent:

I went out there and changed the resistors to what i said I would. TOday nothing was sparking, Yesterday when i was pulling the wires off and was quite ticked ... i thought that i saw a spark coming off the board somewhere but didn't care much at the point. I did some checking around with my volt meter good thing that i brought it. The voltage coming off the blue wire was now about 0.340V hence the red light not blinking but the green one was. I'm not sure what the output voltage really should be there but i know that for a fact that this is too low to give any signal to any electronic component. From there i looked at the voltage regular voltages ... 14v in and 5 volt out. then I looked at the transistor and thats what I think I might of blown. The one thats under the red light one more specifically. Sorry i didn't have a very informative day today but maybe tomorrow morning I'll do some more testing. Let me know what you think this is Kent and I will have a look at it tomorrow. I haven't touched anything else but those 2 resistors. I'm off to work now so hopefully by the time I come back I will let you know what happens.
Old 05-31-06, 07:06 AM
  #987  
Super Moderator

iTrader: (3)
 
gsl-se addict's Avatar
 
Join Date: Jan 2003
Location: Lynchburg, VA
Posts: 5,088
Likes: 0
Received 11 Likes on 11 Posts
Hmm. Check for shorts first. I would look at the area where the wires meet the boad and around where you soldered the replacement resistors. It sounds like maybe there could have been a short when your voltmeter was bouncing.

When you measured the voltage on the blue wire, was the car running? If it was running the voltage measured will be an average and much less than the normal 4.3V even with the board working properly. Once you look the board over for shorts, you can do some tests before installing the board. For resistances, they should be:

- red to red wire = 0
- blue to black 1k
- red to yellow 1k
- red to black 6k or more

If those all look good, you can power the board and then ground the yellow wire. The green LED should go out and the red LED should come on (one rapid blink). If the red one doesn't, start measuring voltages at the output and work your way back. This would have to be done without R2 connected. Otherwise the pulse will be too fast to measure on your meter.

- output should be about 4.3v
- the E terminal on Q2 should be 4.3v
- the C terminal on Q2 should be 5v
- the B terminal 5V

You can also test Q1 and Q2 if your meter has a diode check function. This test you run without power connected.

I wish there wasn't such a time difference between us. It would be easier as we could chat during the testing and figure things out. I will also give R3=6.8K a try here to make sure it works okay on the bench. Good luck. Let me know if I can help you in any way.

Kent
Old 05-31-06, 09:42 AM
  #988  
Banned. I got OWNED!!!
iTrader: (2)
 
dj55b's Avatar
 
Join Date: Oct 2004
Location: London, Ontario
Posts: 6,122
Likes: 0
Received 1 Like on 1 Post
Hey, I'll check those in a few minutes ...if you have a 5.6k to try with as I have that would be great. Thanks
Old 05-31-06, 10:34 AM
  #989  
Banned. I got OWNED!!!
iTrader: (2)
 
dj55b's Avatar
 
Join Date: Oct 2004
Location: London, Ontario
Posts: 6,122
Likes: 0
Received 1 Like on 1 Post
Alright so looking at resistances first .. everything check out I think ... A couple of thing thous ... between red and black resistance is about 168 kohm ... but I don't think that matters all that much .. between red and red though with the diode in place the it shows infinity for resistance but i presume thats from the diode ... voltage does get through as I get about 12 volts to the end of the igniter lead.

Now for voltage checks ... so you know i'm using the 12v off my computer as my power source as its easier to type here and do measurements.

With just 12v and ground connected the green light stay on.

Connecting the yellow wire to ground ... red turn on but very very low, I have to look straight at it to actually see it turn on.

Checking the output of the blue wire ... I get 0v from there.

Looking at q2 now ... i get 0 V at the base and emmiter. but i get 5 volts at the collector.

looking at q1 i get 0.4v at the vollector ... 0.69 at the base ... and 0v at the emitter.

Looking at the nand gate ... pin 2,3,6,8,12,13,14 I have 5 volts ... pin 1 = .04v ... pin 4,5,7 ,9,10, I get 0 volts .... and pin 11 sees 0.64volts ... from those numbers i can say that the nand gate is working properly but the inputs seem a little messed up.

That is as far as my testing goes for now ... I knew it was something wrong with the output on my board ... I've looked for shorts but couldn't find anything but i did it quickly i'll recheck things again tonight after work ... I have to go right now but if you can find out what is wrong with that ... please let me know thanks.

Sam
Old 05-31-06, 11:17 AM
  #990  
Super Moderator

iTrader: (3)
 
gsl-se addict's Avatar
 
Join Date: Jan 2003
Location: Lynchburg, VA
Posts: 5,088
Likes: 0
Received 11 Likes on 11 Posts
Hey Sam,

I took a look at mine. I used 5.6k for R3, seems to be working just fine. I wonder if there is anything wrong with your board. With R2=27k, the red LED will be very faint at idle. Not nearly as bright as when you had the 56k connected. I also measured my output voltage. With the 27k for R2 and 35Hz, the output is 0.40v measured by a DMM. Now idle, is something like 23Hz, so the voltage should be lower. I can only get down to 35Hz with the simple square wave generator that I built.

I think everything may be okay. You could set your meter on amps and check the current draw on the circuit (inline with the power input, make sure to move the test lead on your meter over to the amps position). Mine shows about 20mA with this set of resistors, both LEDs, and firing the coil. The strange thing is that there is almost no difference in current used between coil connected or not (about 0.3mA).

Kent
Old 05-31-06, 07:26 PM
  #991  
Banned. I got OWNED!!!
iTrader: (2)
 
dj55b's Avatar
 
Join Date: Oct 2004
Location: London, Ontario
Posts: 6,122
Likes: 0
Received 1 Like on 1 Post
well its raining out there right now so i can't do much testing the numbers that I have giving you was all running off of the computer power supply as mentioned before ... only thing that i can think of right now is could of my 2ng gen ignitor of been blown now? Like I have mentioned before it did get prety warm running it, 2 days ago. Is there any way of testing that? It thats all that is ... i should be able to pull one off friday or so from another car. or if you would recommend changing i think it was the resistor in there that blows away? I have a pretty good electronics parts store nearby so it should be hard to order in anything if they don't have it in stock.
Old 05-31-06, 11:30 PM
  #992  
pwned

 
dean23's Avatar
 
Join Date: Sep 2005
Location: Northern Indiana
Posts: 1,296
Likes: 0
Received 0 Likes on 0 Posts
ok.. prolly stupid, but i got and 80' SA. im getting the 2nd gen lead coil.. and want to hook up the TT with it. anything else i need? and how much for the tt.. pm me please


dean
Old 06-01-06, 02:19 AM
  #993  
Super Moderator

iTrader: (3)
 
gsl-se addict's Avatar
 
Join Date: Jan 2003
Location: Lynchburg, VA
Posts: 5,088
Likes: 0
Received 11 Likes on 11 Posts
Sam: The 0.34V on the blue wire was in the car with the car running or with the car off? If running, it sounds about right. Let's not jump to concusions on the 2nd gen pack being bad, yet. Many guys had theirs running for months with the full pulse from the J-109 (like 26ms according to Paul/Rob's measurements). With the 56k in place, it would have been limited to 5.6ms.

There are a couple ways we can check things out. You can 1st check the board by doing those measurements (computer PS is okay) with R2 not connected. With R2, the pulse is so short, it is hard to do any real measurements with a DMM. The other possibility is to assume the board is okay and reconnect. Check to see that you are getting power to the tan wire on the 2nd gen coil and make sure that it is grounded. You can test the coil/ignitor by touching the red input wire of the 2nd gen coil (TT not connected) against 12v. You should get spark from the plugs.

Anyway, I can test those same points on my board with the input just grounded and see what I get just to compare. I am sure that we will get this figured out. Good luck.

Dean: I will send you a pm about the circuit in a little while.

Kent
Old 06-01-06, 11:30 AM
  #994  
Super Moderator

iTrader: (3)
 
gsl-se addict's Avatar
 
Join Date: Jan 2003
Location: Lynchburg, VA
Posts: 5,088
Likes: 0
Received 11 Likes on 11 Posts
Hey Sam,

I took some measurements on my board. It appear that there is something worng on your board. I will look at your numbers more and see if I can figure out what is going on. Here is what I get:

Q1: E=0, B=0, C=5V
Q2: E=0, B=0, C=5V

NAND: 1,3, 6, 8, 12, 13, 14 = 5V
2, 4, 5, 7, 9, 10, 11 = 0V
I will let you know what to check on your board.

Kent
Old 06-01-06, 02:09 PM
  #995  
procrastination engineer

iTrader: (1)
 
Mills's Avatar
 
Join Date: Mar 2003
Location: up to my ass in alligators
Posts: 1,270
Received 1 Like on 1 Post
It took me three days to read this thread!

Please put me on the back-up list for one of these, but I would need a completely assembled/sorted out unit glued shut so I could never open it .
Old 06-01-06, 02:27 PM
  #996  
Super Moderator

iTrader: (3)
 
gsl-se addict's Avatar
 
Join Date: Jan 2003
Location: Lynchburg, VA
Posts: 5,088
Likes: 0
Received 11 Likes on 11 Posts
Sounds good, Mills. I'll add you to the wait list. It is a beast of a thread, isn't it?

Sam: Was your testing with the yellow wire grounded? One thing that appears strange is your readings on Q1. It appears that the input is perhaps floating (not grounded). If the yellow wire is gronded, I would take a look near R3. It could be that the resistor is not soldered good or something and not letting the signal though. Some of the other numbers look to off as well. Anyway,we will figure it out. If you have time tomorrow or the weekend,we could find a time to chat and go through everything and figure out what is going on.

Kent
Old 06-01-06, 03:35 PM
  #997  
Banned. I got OWNED!!!
iTrader: (2)
 
dj55b's Avatar
 
Join Date: Oct 2004
Location: London, Ontario
Posts: 6,122
Likes: 0
Received 1 Like on 1 Post
Hey kent ... i though something was wrong on the board ... I will make sure that all the new resistors are in place properly ... I didn't put them in all the way through the holes but i tried to pull on them and they wouldn't come one but i'll put them in all the way and see if that makes a difference. As far as time wise ...we can maybe work something out for saturday or sunday morning EST ... how many hours difference are you? If i'm not mistaken since you're just about under england I would assume about 6 hour (ahead?). Anyhow i'll fiddle with the board tomorrow and let you know what happend then hopefully.
Old 06-01-06, 10:20 PM
  #998  
pwned

 
dean23's Avatar
 
Join Date: Sep 2005
Location: Northern Indiana
Posts: 1,296
Likes: 0
Received 0 Likes on 0 Posts
hey would replacing the trailing coil with a 2nd gen trailing be any good? i had one offered to me and cant decided if i should take it or not.
Old 06-01-06, 11:16 PM
  #999  
Lives on the Forum

 
Kentetsu's Avatar
 
Join Date: Jun 2004
Location: Grand Rapids Michigan
Posts: 11,359
Received 14 Likes on 11 Posts
The 2nd gen trailing coil is worthless to us, at least at this point in time...
Old 06-01-06, 11:27 PM
  #1000  
MattG FTW!!!!!

iTrader: (2)
 
MattG's Avatar
 
Join Date: Feb 2005
Location: South Windsor, CT
Posts: 2,733
Likes: 0
Received 0 Likes on 0 Posts
Correct you are my friend, the best thing you can do is upgrade your trailing coil to a MSD blaster 2 coil, or other high output coil.

POST # 1000

WooT!!!


Quick Reply: Transistor trick for 2GCDFIS.



All times are GMT -5. The time now is 02:05 AM.