MS2 Spark Hardware Latency.
10-25-2010, 05:36 PM
#1
Junior Member
Thread Starter
Join Date: Oct 2008
Location: Manchester, UK
Posts: 272
Total Cats: 0
MS2 Spark Hardware Latency.
Let’s make sure your spark hardware latency is configured. You can see by default its set to 0. You can verify by locking ignition timing 10°, which it should still be and then slowly running the engine from idle to 6,000 RPM while observing with a timing gun. If the spark tends to retard as RPM goes up, you need to add latency correction until it stays at 10*. This will prevent your timing from dropping in higher rpms.
It's the same scenario higher up the rev range where at 4000rpm the MS is commanding 21.8* but the spark cells are set to 26*, 27*, 30* and 32*.
When I installed the MS2, it was late and I didn't want to rev the engine up to 6000rpm
I'll have to drive to a more remote location and check it out. If anyone could advise, that would be cool.
Reply
0
0
10-25-2010, 05:52 PM
#3
Elite Member
iTrader: (1)
Join Date: Jun 2006
Location: Warrington/Birmingham
Posts: 2,642
Total Cats: 42
THats nothing to do with latency mate, latency is the difference between commanded and actual timing.
You must have IAT based retard or something along those lines in software which is reducing the commanded timing
You must have IAT based retard or something along those lines in software which is reducing the commanded timing
Reply
0
0
10-25-2010, 05:53 PM
#4
Junior Member
Thread Starter
Join Date: Oct 2008
Location: Manchester, UK
Posts: 272
Total Cats: 0
I haven't touched any knock sensitivity settings.
I've just applied KnockRet in MLV and it's pretty much sat at a value of 5* for the whole datalog. The figure drops back to 0* at idle and then ramps back up to 5* with any throttle input. Richy, I'll check out IAT related retard too. I get what your saying regarding latency, it wouldn't show in the log at all would it as it's visually inspected at the crank with a timing light and checked against the MS. (Hey I had a bad nights sleep last night so blame it on my tiredness. My 10 month old wanted to sing at 2am and 4am and then she was up at 6.30am!)
I think we have a winner...?
I've just applied KnockRet in MLV and it's pretty much sat at a value of 5* for the whole datalog. The figure drops back to 0* at idle and then ramps back up to 5* with any throttle input. Richy, I'll check out IAT related retard too. I get what your saying regarding latency, it wouldn't show in the log at all would it as it's visually inspected at the crank with a timing light and checked against the MS. (Hey I had a bad nights sleep last night so blame it on my tiredness. My 10 month old wanted to sing at 2am and 4am and then she was up at 6.30am!)
I think we have a winner...?
Last edited by CRAIGO; 10-25-2010 at 06:08 PM.
Reply
0
0
10-25-2010, 06:09 PM
#5
Elite Member
iTrader: (10)
Join Date: Jun 2006
Location: Athens, Greece
Posts: 5,977
Total Cats: 356
Yup, you need to adjust the knock sensivity on the MS2 board via a potentiometer. I think someone asked on the forum whether this unit support knock sensing, so I'll repeat here:
1) The DIYPNP has a basic onboard knock sensing, which isn't brilliant but does provide some basic functionality
2) While testing I discovered that the circuit was not designed properly and was ALWAYS detecting knock. DIYAutoTune suggested a fix which I applied to all units.
3) You will need to adjust the sensitivity of the circuit to your engine. Basically enable it even at idle (min rpm set to 800) and then adjust the blue potentiometer (there's only one the DIYPNP) until you get knock (or you don't get knock if you were getting it all the time) and then back off another 10 degrees CCW. That should be a good starting point.
4) ALWAYS listen for knock anyways - this circuit is really, really basic and has no guaranteed performance. Knock WILL kill your engine in seconds under boost!
1) The DIYPNP has a basic onboard knock sensing, which isn't brilliant but does provide some basic functionality
2) While testing I discovered that the circuit was not designed properly and was ALWAYS detecting knock. DIYAutoTune suggested a fix which I applied to all units.
3) You will need to adjust the sensitivity of the circuit to your engine. Basically enable it even at idle (min rpm set to 800) and then adjust the blue potentiometer (there's only one the DIYPNP) until you get knock (or you don't get knock if you were getting it all the time) and then back off another 10 degrees CCW. That should be a good starting point.
4) ALWAYS listen for knock anyways - this circuit is really, really basic and has no guaranteed performance. Knock WILL kill your engine in seconds under boost!
Reply
0
0
10-25-2010, 06:21 PM
#6
Junior Member
Thread Starter
Join Date: Oct 2008
Location: Manchester, UK
Posts: 272
Total Cats: 0
Cheers for that!
As it stands, I'm tuning fuel with caution and my spark map is around 10* more conservative in boost than my previous MS1 spark map.
For the time being, can I simply set knock control in TunerStudio to "disabled" until I have some time to set it up properly?
As it stands, I'm tuning fuel with caution and my spark map is around 10* more conservative in boost than my previous MS1 spark map.
For the time being, can I simply set knock control in TunerStudio to "disabled" until I have some time to set it up properly?
Reply
0
0
10-26-2010, 04:06 AM
#10
Junior Member
Thread Starter
Join Date: Oct 2008
Location: Manchester, UK
Posts: 272
Total Cats: 0
Yeah I got that once I had engaged my brain. I understand it has to be visually compensated for using a timing light and that the MS won't have a clue if there is any associated latency.
This mornings log shows true spark map figures. I'll also do the hardware latency test this weekend.
This mornings log shows true spark map figures. I'll also do the hardware latency test this weekend.
Last edited by CRAIGO; 10-26-2010 at 04:16 AM.
Reply
0
0
Thread
Thread Starter
Forum
Replies
Last Post
StratoBlue1109
Miata parts for sale/trade
21
09-30-2018 01:09 PM
