Baseline MSQ for '99-00, including Abe's Board
#122
Elite Member
Thread Starter
iTrader: (3)
Join Date: Aug 2006
Location: San Diego, CA
Posts: 3,047
Total Cats: 12
yeah, that's the idea, with everything warmed up. It's not the best solution in the world, but no knocksensor is the worst one in the world. :-)
It only takes a minute to tune, just turn the screw back and forth, the biggest challange is finding somewhere you won't **** off the world by revving your motor.
There's a fee to get in. $15 or $30 or something. Worth it for a day at the races. My friend's mini had a manquin in a chair on top: Mr. Bean. I mean, c'mon. :-)
It only takes a minute to tune, just turn the screw back and forth, the biggest challange is finding somewhere you won't **** off the world by revving your motor.
There's a fee to get in. $15 or $30 or something. Worth it for a day at the races. My friend's mini had a manquin in a chair on top: Mr. Bean. I mean, c'mon. :-)
#123
Ok, I finally got a datalog of the suspected sync drops. If my suspicion is correct, I may have a clue for us.
I hooked up the computer directly without leaving the car on in position. Low and behold, I got the spark drops. It seems to be more apparent when the car FIRST STARTS and the car isnt "on" for a moment. If I'm thinking right, im going to try to start my car again tonight with another factor in place.
THe last time i told you about when i started the car and it started up just fine without any hick, is because the car was left on the "ON" position while i hooked up my computer to datalog. It could be something related to capacitors.
And the datalogs are not only showing a loss of sync code 32, but it showed up 31 as well, meaning NO triggers were seen at that point. The timing error was the same, at 12.7%
I hooked up the computer directly without leaving the car on in position. Low and behold, I got the spark drops. It seems to be more apparent when the car FIRST STARTS and the car isnt "on" for a moment. If I'm thinking right, im going to try to start my car again tonight with another factor in place.
THe last time i told you about when i started the car and it started up just fine without any hick, is because the car was left on the "ON" position while i hooked up my computer to datalog. It could be something related to capacitors.
And the datalogs are not only showing a loss of sync code 32, but it showed up 31 as well, meaning NO triggers were seen at that point. The timing error was the same, at 12.7%
#124
Elite Member
Thread Starter
iTrader: (3)
Join Date: Aug 2006
Location: San Diego, CA
Posts: 3,047
Total Cats: 12
Well, I can shed one small bit of light: 12.7% is very likely the most it can show, meaning, once it hits 12.7% it just assumes it a miss, so it's no great insight, but at least it's consistent.
I don't know how cavalier you want to be with your board, but you could desolder one leg of the resistors if you think they are the issue. I don't have any in my car. They should make you more likely to miss a tooth and less likely to get a false one. Similarly we could add another resistor in parallel to make it more sensitive but the cap is the place to start.
But if you have some other ideas, try them first.
I don't know how cavalier you want to be with your board, but you could desolder one leg of the resistors if you think they are the issue. I don't have any in my car. They should make you more likely to miss a tooth and less likely to get a false one. Similarly we could add another resistor in parallel to make it more sensitive but the cap is the place to start.
But if you have some other ideas, try them first.
#125
Yup. As i suspected.
The only thing that I have an idea of it missing like that is if the car doesn't "charge" up and you just flip the switch right away.
So. I turned my key to on, and I sat there for about 30 seconds. Started the car on a cold start. Vroom. No misses.
It has to be something in the circuit making it miss like that. Unfortunately, thats as far as my trouble shooting can go, I am clueless on how your circuit, or the op-amp circuit works. All i know, is that there is another circuit that uses hex-schmitt triggers, but i have no idea how the resistors or the capacitors work in each one.
The only thing that I have an idea of it missing like that is if the car doesn't "charge" up and you just flip the switch right away.
So. I turned my key to on, and I sat there for about 30 seconds. Started the car on a cold start. Vroom. No misses.
It has to be something in the circuit making it miss like that. Unfortunately, thats as far as my trouble shooting can go, I am clueless on how your circuit, or the op-amp circuit works. All i know, is that there is another circuit that uses hex-schmitt triggers, but i have no idea how the resistors or the capacitors work in each one.
#126
Elite Member
Thread Starter
iTrader: (3)
Join Date: Aug 2006
Location: San Diego, CA
Posts: 3,047
Total Cats: 12
Yeah, there's two little caps, as I recall they should be close to the 8 pin IC. They DO charge, they are your noise filters, and they are behind the resistors. Don't hate me too much, that's mazda's design! :-P
My first instinct would be to say ignore them, so you miss when you start. :-)
My next would be to say either drop the value of the resistors charging them (make you somewhat more likely to miss), drop the value of the cap (probably the best idea), or possibly lower the resistance on the feed to the sensor (won't help a lot with them charging).
I would just take them out and call it good, but if you're really inspired, cut the value by, oh, 1/3 or so. I think they are 10 nf (or 1, I always get that one wrong. Therefore probably 1) so go buy something smaller, 2.2 or 3.3 nf (more likely 220 or 330pf) and see if it helps. Should charge in ~1/2 the time and still will offer good noise rejection.
My first instinct would be to say ignore them, so you miss when you start. :-)
My next would be to say either drop the value of the resistors charging them (make you somewhat more likely to miss), drop the value of the cap (probably the best idea), or possibly lower the resistance on the feed to the sensor (won't help a lot with them charging).
I would just take them out and call it good, but if you're really inspired, cut the value by, oh, 1/3 or so. I think they are 10 nf (or 1, I always get that one wrong. Therefore probably 1) so go buy something smaller, 2.2 or 3.3 nf (more likely 220 or 330pf) and see if it helps. Should charge in ~1/2 the time and still will offer good noise rejection.
#128
Elite Member
Thread Starter
iTrader: (3)
Join Date: Aug 2006
Location: San Diego, CA
Posts: 3,047
Total Cats: 12
Play with the caps first. Since we know we can leave them out, and you don't have an oscope, it's the safest route.
They should say something on them, 105 or similar. Subtract one from the last number and multiply the 10 by ~3 to get what you want. so, 105 becomes 334 or 224, 106 becomes 335. Make sense?
They should say something on them, 105 or similar. Subtract one from the last number and multiply the 10 by ~3 to get what you want. so, 105 becomes 334 or 224, 106 becomes 335. Make sense?
#133
I replaced the transistors. Hooked up the MS, IAT sensor wire to pin #3, vaccumm hose. Turned the ignition on, ran MT on my laptop, tested communication, loaded Marc's file, entered the fuel settings (i.e., eng displacement, injector size) calibrated TPS, saved the file and burned it to MS. Started the car, no luck. The car did not start. Am I missing something? I am a complete noob to MS and MT. BTW, its a stock NB. I haven't done the turbo install yet. My friend Craig (dvcn) said there's no spark or fuel. Anything I should set on MT, to change the spark and/or fuel to work on a stock NB?
#134
Elite Member
Thread Starter
iTrader: (3)
Join Date: Aug 2006
Location: San Diego, CA
Posts: 3,047
Total Cats: 12
Sure, let's see.
First off, check all the sensors, try "real time display". What is the CLT, AIT (might be MAT), etc. You want to make sure everything makes sense.
Also, any warnings when you loaded the MSQ? The MSQ and the firmware version MUST match. Make sure they are all the same.
Then, if all that is good, make sure the RPM are showing while cranking. You have to crank longer than you think, i.e. several seconds, not just "bump the key"
First off, check all the sensors, try "real time display". What is the CLT, AIT (might be MAT), etc. You want to make sure everything makes sense.
Also, any warnings when you loaded the MSQ? The MSQ and the firmware version MUST match. Make sure they are all the same.
Then, if all that is good, make sure the RPM are showing while cranking. You have to crank longer than you think, i.e. several seconds, not just "bump the key"
#138
Elite Member
Thread Starter
iTrader: (3)
Join Date: Aug 2006
Location: San Diego, CA
Posts: 3,047
Total Cats: 12
There really aren't much, it's all in the build thread, here: https://www.miataturbo.net/forum/t38342/
#139
I put a guide here on setting up TunerStudio for your computer
https://www.miataturbo.net/forum/t40159/
Abe, I haven't gotten around to installing the capacitors yet. I went reading through the old thread about your circuit to find out something about the capacitors.
If you put 10nf on the capacitor, I remember jason saying it was too big for a cap and the capacitors used should be X7R or NPO type capacitors. I ordered some of these from digikey, and i bought 4700pf and 2200pf capacitors. They roughly mean 4.7nf and 2.2nf. Ill try these out, and ill let you know the results.
Link to posts:
https://www.miataturbo.net/forum/t25789-2/#post306451
https://www.miataturbo.net/forum/t25789-2/#post306654
https://www.miataturbo.net/forum/t40159/
Abe, I haven't gotten around to installing the capacitors yet. I went reading through the old thread about your circuit to find out something about the capacitors.
If you put 10nf on the capacitor, I remember jason saying it was too big for a cap and the capacitors used should be X7R or NPO type capacitors. I ordered some of these from digikey, and i bought 4700pf and 2200pf capacitors. They roughly mean 4.7nf and 2.2nf. Ill try these out, and ill let you know the results.
Link to posts:
https://www.miataturbo.net/forum/t25789-2/#post306451
https://www.miataturbo.net/forum/t25789-2/#post306654
Last edited by Marc D; 10-14-2009 at 12:51 AM.
#140
Elite Member
Thread Starter
iTrader: (3)
Join Date: Aug 2006
Location: San Diego, CA
Posts: 3,047
Total Cats: 12
I believe what's on the board should be 1.0 nf, so 470 or 220 (I wouldn't bother going much bigger than 220 or there won't be enough difference to notice)... Again, I just have no caps and it works well for me.
I need to try my logfile on windows, but my linux version of MLV won't open long logs. It gets most of the way and freezes. Has anyone else seen this?
I need to try my logfile on windows, but my linux version of MLV won't open long logs. It gets most of the way and freezes. Has anyone else seen this?