When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
I’ve been doing a lot of VE tuning trying to get my new ECU to make the same, or better, power that my 15 year old Link (RIP) did. Headed to the dyno thinking I’d lean out the fuel a bit but the operator (chief mechanic, not tuner) and I ended up adjusting ignition instead. Started off with the same disappointing 170ish whp and raised timing 2 degrees every pull until we ended up at 216, after which we pulled a couple of degrees and called it a day. I was ecstatic. I’d been chasing these numbers for a year.
But later that day thought, better check knock. Only problem was, I didn’t know how.
Did two things: Sent my remote tuner the final map and datalog from the sixth pull and posted up the following inquiry on the Beer Money Motorsports Facebook page. “Question - how to read knock sensor info in megalogviewer? .In megalogviewer if I select "Knock 1" I see readings of -40 to -16 (just added a lot of ign adv at high boost, going from 172 to 216 on the dyno in 8 pulls.) But I need to see what the knock sensor is saying and don't know what the numbers mean.”
Received a very useful answer from the page, “So that number is the intensity of the knock readings. [In MLV} Use the value knock: current level and make a scatter plot like this against RPM and whatever else you like (I like MAP or hits). basically, as the RPM goes up, so should the vibration that the knock sensor detects because it's basically a microphone. What you're looking for is a nice straight ish line of knock readings as the RPM goes up. That is normal for the engine. What you should be worried about is any data points significantly above this line as that likely indicates a knock. Click that data point switch to the log viewer tab where it will be highlighted when that occurred. If there's a big spike in knock sensor intensity at that point, that's likely a knock and you'll want to play with your tune to remove spark or add fuel."
This is what my scatter plot looked like at max timing.
So yeah, lots of knock. (Feeling lucky I didn't blow up the motor.)
Ignition map at this point was basically the following:
I combed the internet for any similar boosted map and only found two threads containing close numbers.. 1. https://www.miataturbo.net/megasquir...estions-71612/ (In a subsequent thread this member discovered that his crank pulley was off and he was rebuilding the motor. 2. https://www.miataturbo.net/megasquir...ap-help-91571/ These maps also didn't have that high boost, high rpm run up in timing that I had (because we were only working in that area.) Looking now, I know it looks wonky.
I also heard back from my tuner, saying that's “an insane amount of advance” at 28 degrees." (I assume he meant stupid.) He went on, "I needed only to look at the first run under boost. Your engine is knocking, a lot. Take a look at your log as I describe what is happening below.
"The units on the knock sensor do not matter. The sensor picks up sound in a very narrow frequency band. Put the cursor at 23.9 seconds. (I edited the log so the time is different but you'll see what he's talking about.) The green trace is sound from the knock sensor and what we see up to this point in normal. The sound getting louder as the revs and boost build. To the right of 23.9 seconds, the knock trace has spikes far higher than the slope of previous sound would predict. This is knock. The engine in knocking all the way to redline."
"Put the cursor at 24.149sec. The advance is 28 degrees. That is an insane amount of advance. The engine is at 5800RPM. With your Link, I had ignition advance at 5800RPM set to 17.5 degrees. The last log I can find from you on 5-7-2023 has the advance set to 11 degrees. (This must have been a large part of the low power issue.) This is not enough and I feel like we got up to 17 degrees with the new ECU."
So the tuner massaged the timing map, I've been back to the dyno, still had some knock and the map has been massaged again to the following. It's still significantly more that what I see with a lot of maps but knock appears under control and the AFRs and power are good. Taking a break for now. Will log some more in the spring. Posting this up so other newbs may learn from my experience.
Happy New Year all.
Last edited by ridewhencan; 01-10-2024 at 01:03 PM.
Reason: scatter plot didn't display
That’s still a **** ton of timing. You wanna be around 28 degrees at 100kpa, and drop around a degree per psi of boost, putting you at 16 degrees or less at 188. You can add a few degrees for a bit more power with a bit more risk, but that’s still 18 degrees. You’re at 22. I feel like I’ve answered or seen a ton of these timing map threads, none suggest timing this high, for good reason. Your first map, holy hell. Your timing stayed the same from 100 kpa up to 188. I could keep increasing timing on the dyno until power stops increasing, but you gotta know when to stop.
I was always able to run more timing supercharged than turbocharged for obvious reasons.
That said, I go by plugs for detonation. My experience at high boost on a BP, detonation shows up on the plug/plugs before I could see actionable info from a knock sensor. The plugs don't lie, knock sensors damn sure can lie (false knock, or not report real knock).
93 octane with NGK IX Iridium #7 heat range. I'll pull the plugs later today and take a look. I definitely got carried away with finally making some power with this setup.
Check them with a jeweler's loupe, something like 5-10x magnification and look for little ***** of aluminum stuck to the porcelain. Keep track of which plug goes in which cylinder. I've always had certain cylinders more prone to det than others.
Pulled the plugs. Look okay to me (without the magnification.) I did notice that the electrodes on cyl 1 & 4 are lighter than 2 &3 and that the porcelains are darker.
Since the plugs were out I took pics using a Depstech endoscope of pistons 1 and 4 too. 2 and 3 were too far up to get a shot.
Went out and logged a couple of 3rd gear pulls. Would like to upload them here but not sure how yet. Sent them off to tuner. Thinking Curly is right and I should lower advance some more. It's a track day car so spends a lot of time at WOT.
Is individual cylinder fuel/spark detection on and working properly? For some reason 1&3 register, but 2&4 are flat lined. You're well below your threshold, but your threshold seems really high? The log is really difficult to view, if this is a beer money ECU, it sucks. I'll take a regular MS over this all day, it is extremely difficult to find the variable I usually look at. Reminds me of a speeduino unit.
If this is indeed a track car, I'd take 4 degrees from the top row, 2 degrees from the 100kpa row, and interpolate between the two. Yes, it's going to rob some power, but your cylinder temperatures are already going to be sky high thanks to your supercharger, no need help your pistons melt faster by being ANYWHERE close to knocking.
I wish I could answer the question regarding the individual cylinder fuel/spark detection but I can’t. I know that when I bench tested the coils outside of the motor for spark, 1&4 fired and 2&3 fired. I’ll ask about it on the facebook page.
When I was retiring my 15yr old FM Link, I knew I’d need assistance with rewiring the modified harness and also with tuning. My local tuner didn't want to get involved with rewiring a modified harness, so they were out. My remote tuner, who has been a tremendous resource for me as a remote tuner/trouble shooter for the 9 yrs I’ve owned my car was interested in the Beer Money ECU. He got one for his turbo NB, but after trying it for a year, now has a new Link. He also thought it might be a good option for the handful of his his old Link clients, like myself, who needed a new ECU. We spent a lot of time getting to get to a fully functioning stage but I was disappointed with the power and independently went for the ignition advance, looking for the power I had with the Link.
If I had to do it over again, I’d probably go for the MSPro, or perhaps Link or Haltech. But I gotta dance with the gal I brought to the party and that’s the BMM. I’ll try your suggestion (thanks for making it) and see how she runs. I made a short vid of the log that I posted above. When it hiccups after the 4th gear shift, that’s a video splice and not the motor crapping out, lol.
Is individual cylinder fuel/spark detection on and working properly? For some reason 1&3 register, but 2&4 are flat lined. You're well below your threshold, but your threshold seems really high? The log is really difficult to view, if this is a beer money ECU, it sucks. I'll take a regular MS over this all day, it is extremely difficult to find the variable I usually look at. Reminds me of a speeduino unit.
.
For the BMM, the msl file type isnt really used for logs. That is why you are seeing the weird fields and info. What field in particular are you looking for? Here is a normal one where i was logging wideband heating. Had to zip it.
I don't typically load the .msl file for log viewing either. An example is AFR and AFR target. I click the menu button, press "A", and then look for AFR, it's the 11th item (at least AFR* is, whatever that is). On a MSPNPpro log (for example), it's the first option, AFR target is 3rd. If I look for AFR target, it's not in A, ok try T for Target, nope, ok try F for fuel, oh there it is, 36 items down. See what I mean? That's just me though. I am not a big fan of Megasquirt/speedunio/BMM ECUs, I'm a big fan of the new Link ECUs. That being said, MLV is awesome, I still miss it sometimes while tuning Links. But I've tuned over a hundred MS based cars on the dyno, and I look for a particular set of data points while looking at logs, and it was particularly hard to find them in a BMM log, to help OP in this case.
Here's a log from a dyno pull on my car, back when it had a MS. These are a few of the variables I look at during a log, I'll change them depending on what I'm concerned with at the moment. But here I'd be monitoring fuel, fuel target, knock, MAP, CLT/MAT temps, battery voltage, and fuel pressure. I opened this log and set all these parameters within a minute. I'd probably add TPS, spark advance, and a few others normally, but this is just for an example.
Here's your BMM log, granted without the same change in RPM, but I was trying to set up the same parameters. Knock was fairly easy to find, although I'm not sure why I need to see all 12 cylinders of knock. I only enabled 4 in my MS, so it only logged 4. Was a little annoying that MAT was IAT, AFR is AFR*, and down the list there's air/fuel ratio 1 and 2, AFR target was buried in fuel, battery voltage was under V for Vbatt, couldn't for the life of me find fuel pressure, maybe you don't have it enabled? Not sure. There's roughly 160 logged parameters in my MS log. I used maybe 10-20 consistently. Your log has 550ish logged items, and I still just want the same 10-20 parameters. Just a **** ton to scroll through.
For some reason when I edit a log to reduce the size to the target area, it saves as an .msl file. In the interest of science (and learning) I tried to attach two .mlg logs. The first is a fourth gear pull, which led to the reduced timing 12-20 map above. I wasn't able to attach second log since at 35mb it exceeds the 20mb limit. It was a much longer log using the 12-20 map and showed the pull in the video above. That log lead to a further reduction in timing and fattening up the fuel a bit mid range. I was trying to look at Curly's parameters above but couldn't find IAT. I found Ignition IAT correction and Fuel IAT correction. I've not yet logged the 1-3 map. I know it still has more than the suggested advance. East coast 93 octane and supercharged, not turbo? Perhaps, need to see what the logs say. But each of the maps are a further reduction from that ridiculous max advance I had going. (Now to see if I can modify the log to something other than .msl.)
Got it. Just had to change the ext from mls to mlg. Scheesh. (Edit. Okay, seems the modification didn't work. And speaking of work, I'd better get going.)
Last edited by ridewhencan; 01-11-2024 at 08:40 AM.
Reason: Modified log didn't work.
I don't typically load the .msl file for log viewing either. An example is AFR and AFR target. I click the menu button, press "A", and then look for AFR, it's the 11th item (at least AFR* is, whatever that is). On a MSPNPpro log (for example), it's the first option, AFR target is 3rd. If I look for AFR target, it's not in A, ok try T for Target, nope, ok try F for fuel, oh there it is, 36 items down. See what I mean? That's just me though. I am not a big fan of Megasquirt/speedunio/BMM ECUs, I'm a big fan of the new Link ECUs. That being said, MLV is awesome, I still miss it sometimes while tuning Links. But I've tuned over a hundred MS based cars on the dyno, and I look for a particular set of data points while looking at logs, and it was particularly hard to find them in a BMM log, to help OP in this case.
Here's a log from a dyno pull on my car, back when it had a MS. These are a few of the variables I look at during a log, I'll change them depending on what I'm concerned with at the moment. But here I'd be monitoring fuel, fuel target, knock, MAP, CLT/MAT temps, battery voltage, and fuel pressure. I opened this log and set all these parameters within a minute. I'd probably add TPS, spark advance, and a few others normally, but this is just for an example.
Here's your BMM log, granted without the same change in RPM, but I was trying to set up the same parameters. Knock was fairly easy to find, although I'm not sure why I need to see all 12 cylinders of knock. I only enabled 4 in my MS, so it only logged 4. Was a little annoying that MAT was IAT, AFR is AFR*, and down the list there's air/fuel ratio 1 and 2, AFR target was buried in fuel, battery voltage was under V for Vbatt, couldn't for the life of me find fuel pressure, maybe you don't have it enabled? Not sure. There's roughly 160 logged parameters in my MS log. I used maybe 10-20 consistently. Your log has 550ish logged items, and I still just want the same 10-20 parameters. Just a **** ton to scroll through.
Thank you for the input. You do raise a good point. We need to filter unused fields to clean it up a bit. Right now, you just get everything possible. Since the brains of the ecu are capable of 12 cylinders of fuel and ignition, you see all of those fields in the log. Ill take that back to the firmware guys and see what can be done.
This. I'm only running a naturally aspirated high comp BP but 30° was the most we ever dyno'ed at 100kpa. Power difference was 2hp, so negligible and I'm running 28° at most.
12-31-2023, 03:45 PM
Ign adv adventure, or what I learned about knock.
0
