2560 on a 99 motor at 14psi with MSPNP 300hp - 241ft/lbs
01-16-2008, 11:53 AM
#61
Elite Member
iTrader: (9)
Join Date: Jun 2006
Location: Chesterfield, NJ
Posts: 6,893
Total Cats: 399
Yep exactly.
Maybe I'll accidentally lose his manifold while welding it until he hooks the VICS up on his current manifold. It's laying under my desk and I keep kicking it. The mexicans may think it's garbage and take it.
Maybe I'll accidentally lose his manifold while welding it until he hooks the VICS up on his current manifold. It's laying under my desk and I keep kicking it. The mexicans may think it's garbage and take it.
Reply
0
0
01-16-2008, 05:14 PM
#63
Elite Member
Thread Starter
iTrader: (21)
Join Date: Jun 2006
Location: Point Pleasant, NJ
Posts: 2,957
Total Cats: 2
ok ok ok. i'll do it. maybe for the hell of it we can just hook a cable to vics and activate it manually on the dyno at 5000 or 5200 or whatever rpm people agree on that week.
first i gotta finish this one. install new injectors. tune car with both intakes. then dyno. who's helping me pay in the name of research? since swapping manifolds is gonna take some time i'll probably have to pay for 2 sessions
first i gotta finish this one. install new injectors. tune car with both intakes. then dyno. who's helping me pay in the name of research? since swapping manifolds is gonna take some time i'll probably have to pay for 2 sessions
Reply
0
0
03-10-2008, 12:34 PM
#64
Junior Member
Join Date: Jan 2007
Location: EXTREME Southern Illinois
Posts: 281
Total Cats: 0
Looking at your timing map on the first page, yours is FAR more agressive than what I'm running. Nearly 10* more agressive in the full-boost area. Did you make/test that map?
Reply
0
0
03-17-2008, 03:55 PM
#69
ok ok ok. i'll do it. maybe for the hell of it we can just hook a cable to vics and activate it manually on the dyno at 5000 or 5200 or whatever rpm people agree on that week.
first i gotta finish this one. install new injectors. tune car with both intakes. then dyno. who's helping me pay in the name of research? since swapping manifolds is gonna take some time i'll probably have to pay for 2 sessions
first i gotta finish this one. install new injectors. tune car with both intakes. then dyno. who's helping me pay in the name of research? since swapping manifolds is gonna take some time i'll probably have to pay for 2 sessions
The one thing I noticed, shutting off the accel enrichment makes the car so god-damn snappy my neck hurts.
I guess there's still no good box for knock sensing, is there? Something where you can get an analog voltage proportional to knock? I want to tune up my car a bit....
Reply
0
0
03-17-2008, 04:44 PM
#71
Junior Member
Join Date: Jan 2007
Location: EXTREME Southern Illinois
Posts: 281
Total Cats: 0
You COULD just use a simple RPM switch to run your VICS.
http://store.summitracing.com/partde...5&autoview=sku
And I've been very happy with my J&S for knock sensing. I have it's output running to my standalone so that I can log it and use it to adjust timing/fuel/boost in case of knock.
http://store.summitracing.com/partde...5&autoview=sku
And I've been very happy with my J&S for knock sensing. I have it's output running to my standalone so that I can log it and use it to adjust timing/fuel/boost in case of knock.
Reply
0
0
03-17-2008, 07:46 PM
#73
Junior Member
Join Date: Jan 2007
Location: EXTREME Southern Illinois
Posts: 281
Total Cats: 0
This is going to be a long post since I never do anything simple....
Both of my cars with standalone are using J&S units.
The J&S needs to know when the plugs fire to know when to listen for knock. It's intended to wire inline with the coils, and if it detects knock it automatically retards the timing on the cylinder that knocked, or all cylinders if you prefer. There are several different configurations depending on which model you get.
The one on my DSM is an old one that just has a sensitivity **** and an arm/disarm wire that was meant to be connected to a TPS sensor so that it only listens under load. I used an on/off output from my standalone connected to it's TPS wire to have more control over when it's enabled/disabled. All of the J&S units have an output intended for the status display. But since it outputs a voltage proportional to the knock it senses, it's ideal for feeding back into the standalone for logging. The SMC in my DSM doesn't have anymore spare inputs, so I put a switch and a resistor on the wideband input. Depending on the switch position, it logs AFR, or Knock. I have it set as sensitive as I can get it without causing any false alarms. Occasionally it will show some false knock when the engine is cold, but it's dead reliable when everything is up to temp. I thought it was falsing during a rallycross and pressed on anyway despite the annoyance and power loss. On the way home I ran the boost up and watched the gauges. By 5K at 20psi my AFR said 14:1 and the J&S was pulling timing like mad. Turns out my fuel pump was dying. I'm still not sure how the motor survived, but I can't help but feel that the J&S helped a lot.
My Miata uses a newer UltraSafeguard. I didn't want to give the safeguard direct control over my coils this time since Autronic is working on their own super-whammy knock control system and I didn't want to rewire stuff when it became available. This one referenced manifold pressure instead of having the arm/disarm wire. I spoke with John at J&S about using it that way, and although he didn't recommend it, he helped me out. I wanted to run it off my tach signal which pulses with the crank signal and not the coils. I also wanted an arm/disarm wire added if possible. He reprogrammed the knock detection software to work with my odd trigger location and also programmed one of the other inputs (VTEC Squelch since this was a Honda Unit) to work as an arm/disarm. The reprogramming charge was minimal considering the time he spent working with me, and it seems to work great. Since I had some extra inputs, I'm feeding the signal back into the SM4 and linked the input to my ignition, fuel, and boost tables. If knock is sensed it will first pull some timing, then if that doesn't work it will start cutting boost and adding fuel. The only downside to this setup is that there's probably some lag to the ECU responding to knock whereas it would be instant if I let the J&S run the coils. I never seem to get false indications on the miata at all, but I've never tried to induce knock beyond the wrench test.
However you set it up, it's important to have the unit disabled at low load (very noisy down there) so you can have the sensitivity cranked up. The units are expensive, but I'm convinced they save motors and they include peace of mind. It is a great company to work with. When you call or e-mail you're talking to the guy that makes them. Sometimes he's busy, but don't give up on contacting him.
Both of my cars with standalone are using J&S units.
The J&S needs to know when the plugs fire to know when to listen for knock. It's intended to wire inline with the coils, and if it detects knock it automatically retards the timing on the cylinder that knocked, or all cylinders if you prefer. There are several different configurations depending on which model you get.
The one on my DSM is an old one that just has a sensitivity **** and an arm/disarm wire that was meant to be connected to a TPS sensor so that it only listens under load. I used an on/off output from my standalone connected to it's TPS wire to have more control over when it's enabled/disabled. All of the J&S units have an output intended for the status display. But since it outputs a voltage proportional to the knock it senses, it's ideal for feeding back into the standalone for logging. The SMC in my DSM doesn't have anymore spare inputs, so I put a switch and a resistor on the wideband input. Depending on the switch position, it logs AFR, or Knock. I have it set as sensitive as I can get it without causing any false alarms. Occasionally it will show some false knock when the engine is cold, but it's dead reliable when everything is up to temp. I thought it was falsing during a rallycross and pressed on anyway despite the annoyance and power loss. On the way home I ran the boost up and watched the gauges. By 5K at 20psi my AFR said 14:1 and the J&S was pulling timing like mad. Turns out my fuel pump was dying. I'm still not sure how the motor survived, but I can't help but feel that the J&S helped a lot.
My Miata uses a newer UltraSafeguard. I didn't want to give the safeguard direct control over my coils this time since Autronic is working on their own super-whammy knock control system and I didn't want to rewire stuff when it became available. This one referenced manifold pressure instead of having the arm/disarm wire. I spoke with John at J&S about using it that way, and although he didn't recommend it, he helped me out. I wanted to run it off my tach signal which pulses with the crank signal and not the coils. I also wanted an arm/disarm wire added if possible. He reprogrammed the knock detection software to work with my odd trigger location and also programmed one of the other inputs (VTEC Squelch since this was a Honda Unit) to work as an arm/disarm. The reprogramming charge was minimal considering the time he spent working with me, and it seems to work great. Since I had some extra inputs, I'm feeding the signal back into the SM4 and linked the input to my ignition, fuel, and boost tables. If knock is sensed it will first pull some timing, then if that doesn't work it will start cutting boost and adding fuel. The only downside to this setup is that there's probably some lag to the ECU responding to knock whereas it would be instant if I let the J&S run the coils. I never seem to get false indications on the miata at all, but I've never tried to induce knock beyond the wrench test.
However you set it up, it's important to have the unit disabled at low load (very noisy down there) so you can have the sensitivity cranked up. The units are expensive, but I'm convinced they save motors and they include peace of mind. It is a great company to work with. When you call or e-mail you're talking to the guy that makes them. Sometimes he's busy, but don't give up on contacting him.
Reply
0
0
03-17-2008, 07:59 PM
#74
If it's only being used as an input, then you can leave sensitivity pretty high up - let the MS or whatever computer you're running deal with it.
I'd be curious to see what the signal looks like if ALL you do is restrict it to be in a narrow time window after firing....
What does one of those cost - ballpark? Is there a less/cheaper version without the retard?
I'd be curious to see what the signal looks like if ALL you do is restrict it to be in a narrow time window after firing....
What does one of those cost - ballpark? Is there a less/cheaper version without the retard?
Reply
0
0
03-17-2008, 09:37 PM
#75
Junior Member
Join Date: Jan 2007
Location: EXTREME Southern Illinois
Posts: 281
Total Cats: 0
It appears that it uses more of a decaying knock sum rather than a noise level. For each knock event it pulls more timing, then when no more is detected it decays back to nothing. If you make it too sensitive then it's just pegged all the time.
The output is not the actual knock sensor signal, rather it's the amount of knock the device sees in terms of ten ~0.3v steps.
There's lots of info including pricing on the website: http://www.jandssafeguard.com
The output is not the actual knock sensor signal, rather it's the amount of knock the device sees in terms of ten ~0.3v steps.
There's lots of info including pricing on the website: http://www.jandssafeguard.com
Reply
0
0
03-18-2008, 10:04 AM
#77
Elite Member
iTrader: (22)
Join Date: Dec 2006
Location: Sunny Spanish speaking Non US Caribbean
Posts: 3,224
Total Cats: 3
It appears that it uses more of a decaying knock sum rather than a noise level. For each knock event it pulls more timing, then when no more is detected it decays back to nothing. If you make it too sensitive then it's just pegged all the time.
The output is not the actual knock sensor signal, rather it's the amount of knock the device sees in terms of ten ~0.3v steps.
There's lots of info including pricing on the website: http://www.jandssafeguard.com
The output is not the actual knock sensor signal, rather it's the amount of knock the device sees in terms of ten ~0.3v steps.
There's lots of info including pricing on the website: http://www.jandssafeguard.com
very informative indeed. Now on to my question: you mention hooking it up to the coils; what if anyone is running COPS?
Thanks,
Rafa
Edit: I'm an idiot. I just checked your above link. Thanks.
Last edited by Rafa; 03-18-2008 at 10:05 AM. Reason: Additional info garnered
Reply
0
0
03-18-2008, 10:22 AM
#78
Junior Member
Join Date: Jan 2007
Location: EXTREME Southern Illinois
Posts: 281
Total Cats: 0
Well, most people are installing COP while keeping the wasted spark configuration (only two coil wires from the ECU). In that case it would work just the same. If you wanted a true 4-channel setup then you'd need to buy their 4-channel system instead of a 2-channel.
Reply
0
0
03-18-2008, 10:32 AM
#79
Elite Member
iTrader: (22)
Join Date: Dec 2006
Location: Sunny Spanish speaking Non US Caribbean
Posts: 3,224
Total Cats: 3
Well, most people are installing COP while keeping the wasted spark configuration (only two coil wires from the ECU). In that case it would work just the same. If you wanted a true 4-channel setup then you'd need to buy their 4-channel system instead of a 2-channel.
Sorry to hijack Paul but I thought this would be useful information for all of us running COPS.
Thanks
Reply
0
0
03-18-2008, 10:42 AM
#80
Junior Member
Join Date: Jan 2007
Location: EXTREME Southern Illinois
Posts: 281
Total Cats: 0
It's all personal preference. The stock coils are fairly strong, so from a 'performance per dollar' standpoint you shouldn't change anything. If I were going to mess with it, I'd rather go CDI than COP. The 4-channel system can be used for 2-channel if you don't know which direction you are going to go. But I don't think the 4-channel system has the arm/disarm ability, only the MAP sensor (which would work fine really, I'm just a control freak).
Reply
0
0
