2eighTZ4me Posted February 10, 2011 Share Posted February 10, 2011 The folks at DIY just called me to let me know that they get a crap VR signal over 6100rpm's. (They're tuning the car now) Either the wheel is out of round, or is warped slightly. OK - no problem. The forged motor I have ready to drop in has a brand new Rebello BHJ damper with a(nother) 60 tooth wheel mounted on the inside of the damper. I mentioned this and they stated that the 60 tooth wheel is pretty much "too much" for MS. Said the wheel was too small to boot. Too many sensor events per revolution. I really wonder how Rebello is getting 7500+ rpm's out of their cars on a 60 tooth wheel - based on this observed limitation.....I'm gonna have to give Dave a call.... Nonetheless, to comply, I need to come up with another triggering solution. They recommended a 36 tooth wheel. The problem I have with that is mounting it on the outside of the damper. This will cause me to have to fab up aonther bracket to hold the sensor, not to mention probably re-engineering my (already tight with IC piping) engine bay. I like the idea of doing the marks on the back of the flywheel, but they stated that 6 holes is not nearly enough to be accurate. So - I'm curious what you MS'ers are using for distributorless ignition - i.e. wheel size/resolution etc. and what kind of performance pitfalls (if any) are you seeing? Quote Link to comment Share on other sites More sharing options...
BLOZ UP Posted February 10, 2011 Share Posted February 10, 2011 The CAS in the dizzy. Unless you mean really, really distributor-less. Quote Link to comment Share on other sites More sharing options...
2eighTZ4me Posted February 10, 2011 Author Share Posted February 10, 2011 I mean - 60 tooth (-2) wheel mounted to the crank with a sensor right down there on it. No dizzy at all. I want to avoid a distributor for several reasons altogether. Just spoke with Brian at Rebello racing and he suggested with the MS to drop down to a 36 tooth wheel. He stated they have had good luck with them up to 9K with MS. I'm awaiting a call back from Electromotive to find out if any of their new sensors would provide a better signal. All signs are pointing to dropping (and modifying) to a 36 tooth wheel though. Quote Link to comment Share on other sites More sharing options...
TimZ Posted February 11, 2011 Share Posted February 11, 2011 I mean - 60 tooth (-2) wheel mounted to the crank with a sensor right down there on it. No dizzy at all. I want to avoid a distributor for several reasons altogether. Just spoke with Brian at Rebello racing and he suggested with the MS to drop down to a 36 tooth wheel. He stated they have had good luck with them up to 9K with MS. I'm awaiting a call back from Electromotive to find out if any of their new sensors would provide a better signal. All signs are pointing to dropping (and modifying) to a 36 tooth wheel though. I'm having trouble believing that this is a sensor/wheel problem. As I recall it's a 7-inch wheel on the BHJ which should be plenty big. I'm using this (BHJ with a 60 -2 wheel) setup with a TEC3r and the small (3/8" od) Electromotive pickup with no issues whatsoever out to 8000 RPM. Are you using their newer larger sensor maybe? It's supposed to be more durable, but I think it's physically larger too. Quote Link to comment Share on other sites More sharing options...
2eighTZ4me Posted February 11, 2011 Author Share Posted February 11, 2011 (edited) I don't have the BHJ on there - yet. The balancer that's on the mule motor is a euro balancer of some sort that has had the toothed wheel welded on.....and not very good to boot. Based on the scope output that we picked up yesterday, its both out of round, AND a few of the teeth are screwed up. Kind of hard to explain, but the out of round is obvious - the scope output gets higher and lower in amplitude based on how close the wheel is to the sensor. Also - some of the teeth are showing "spikes". When the signal is running over a tooth, it should be a perfect square wave. We're seeing LARGE dips in about 8-10 teeth, that really gets exacerbated at higher RPM's. The bad teeth are so bad that - when the sensor is reading the tooth, it will make a large enough spike to cause the VR conditioner board to breach the threshold and drop ignition (it thinks there is no tooth there, when in reality, it's a valley within the tooth that it's picking up). I'm fairly confident that with the new BHJ damper and the new trigger wheel - it will be able to hit 8K. Problem is - I've got a bunch of $$ tied up in the current setup with the mule motor, such that I'm just having them finish the tune, and I'll drive it as-is with the rev limiter set to 6200 and thrash the holy hell out of it until it a) breaks or I feel like tearing the motor out of there. The mule only cost me $150, the forged motor, about $6000. The other major difference is the TEC3 that you're using. That thing is a hoss. I went MS due to cost, and it looks like it will be sufficient for what I want to do with it. FWIW - I spoke to Uwe at Electromotive at length yesterday, and he stated that there is no difference between ANY of their sensors. They just used the smaller 3/8" sensors for "tighter" applications where room was at a premium. It's nothing more than a magnet. I was asking for the very same reason - is it stronger? Will it give better resolution? Has the inductive "technology" come a long way since my original sensor was produced? No, no, and no. He has got me drooling over their XDI ignition setup though.... I may eventually go that route - if I can scrape together $900 for the box.... Just damn. Edited February 11, 2011 by 2eighTZ4me Quote Link to comment Share on other sites More sharing options...
Administrators RTz Posted February 11, 2011 Administrators Share Posted February 11, 2011 ...the scope output gets higher and lower in amplitude based on how close the wheel is to the sensor. VR's are supposed to do that. It's normal. When the signal is running over a tooth, it should be a perfect square wave. Not with a VR. It should resemble a sine wave. How well is the sensor shielded? Quote Link to comment Share on other sites More sharing options...
2eighTZ4me Posted February 11, 2011 Author Share Posted February 11, 2011 Sine wave is analog and a smooth curve. It would stand to reason that in the "digital" domain - which is what this information is being broken down in to (1's and 0's) - that you have an on or an off state. There are no "in betweens" which is what a sine wave represents in the audio world. A digital sine wave is merely a representation of bit depth coupled with a sampling rate. You never get a smooth curve, but rather a stepped curve at each sample location. I'm just letting you know what I saw on the scope output. It was clearly an on or an off state. There were no smooth transitions between peaks and valleys. Just 90 degree transitions. I'm an audio engineer on the side, so I know what you're talking about WRT a sine wave. If this were a sawtooth trigger wheel - I could understand that logic, but this is a square toothed wheel, and by all logic, it should mimic on the scope what each tooth physically looks like. OK - I know you know more about this stuff than I do - but I need to ask you to explain the differences in amplitude that I'm seeing on a recurrent basis as the wheel goes around. On an electric guitar pickup (which is exactly what the VR sensor is) - the closer the strings are to the pickup, the hotter the output signal. In a perfect world, every tooth the sensor reads "should" output a consistent +5v. I'm getting lows down in the 3's and then up to 5's and then back down again. If logic follows here with the guitar example - the closer the sensor is to the teeth, the hotter the signal it gets. Conversely, the further away from the pickup/sensor, the less voltage is going to be generated. I'm all ears though - I love to learn - but have been an audio engineer for over 25 years and know what a digital sinewave looks like compared to an analog one - and am just applying that logic to what I'm seeing on the scope. Could you please elaborate? I'm not challenging your response - I know what I saw on the scope (and maybe the whole shebang is wrong) - and I'd like to learn more. Quote Link to comment Share on other sites More sharing options...
mobythevan Posted February 11, 2011 Share Posted February 11, 2011 I have not run the calculations recently, but I do believe an MS1 will not sample fast enough to support a 60 tooth wheel. I would think MSII solves this problem, the guys at DIYautotune would know that off hand. Looking back at my megasquirt LS1 install notes I see this: NOTE: These instructions are designed for LS1/LS6 type engines with a 24-tooth wheel, which can be identified by its black harness connector. This will not work with the LS2/LS7 and its 58-tooth wheel, which can be easily recognized by its gray harness connector. I am sure I said that 58 tooth will not work because the tooth events at high rpm become too fast for the MS1 processor to sample. Like I said, I think the MSII processor would solve this because it runs much faster. Someone should verify so we know it is only an MS1 limitation. Quote Link to comment Share on other sites More sharing options...
NewZed Posted February 11, 2011 Share Posted February 11, 2011 (edited) The variable reluctor generates its own voltage. There is no +/- 5 being turned on and off. (Edit - on the oscilloscope) The Megamanual does a pretty good job of explaining it. http://www.megamanual.com/ms2/pickups.htm Edited February 11, 2011 by NewZed Quote Link to comment Share on other sites More sharing options...
Administrators RTz Posted February 11, 2011 Administrators Share Posted February 11, 2011 The variable reluctor generates its own voltage. There is no +/- 5 being turned on and off. Exactly. A VR is not a digital sensor. Quote Link to comment Share on other sites More sharing options...
2eighTZ4me Posted February 11, 2011 Author Share Posted February 11, 2011 (edited) It's rather interesting that this is the way it was explained to me (perhaps I'm ignorant and didn't pick up the "whole" gist) by the guys at DIY AutoTune. They're the ones tuning my car - and explaining this to me and showing me the scope output. Jesus - I just don't know who to believe anymore. I'll read over the manual again and try to digest - and then question them again when I go to pick up the car. It's probably a "detail" I missed in passing. Moby - the car is running MS II now. Most recent firmware as well. Regardless - if you look at a peak of the wave on the teeth in question - there's a huge downward spike, then back up again - and the curve continues. It's far from smooth, and extremely erratic. Edited February 11, 2011 by 2eighTZ4me Quote Link to comment Share on other sites More sharing options...
Administrators RTz Posted February 11, 2011 Administrators Share Posted February 11, 2011 It's rather interesting that this is the way it was explained to me (perhaps I'm ignorant and didn't pick up the "whole" gist) by the guys at DIY AutoTune. They're the ones tuning my car - and explaining this to me and showing me the scope output. Jesus - I just don't know who to believe anymore. In general, I trust Matt and Jerry. It's probably a simple misunderstanding. Quote Link to comment Share on other sites More sharing options...
NewZed Posted February 11, 2011 Share Posted February 11, 2011 Maybe they were monitoring a converted signal from the Megasquirt board, not the actual VR output. Quote Link to comment Share on other sites More sharing options...
2eighTZ4me Posted February 11, 2011 Author Share Posted February 11, 2011 They had to install a VR conditioning board to even get it to function. Car wouldn't run at all without it. They were showing me both raw data right off the sensor, AND the converted/conditioned data. Raw data was in blue and the conditioned data was in red. The conditioned data was a perfect +5v all the way across - amplitudes dead even. The raw signal looked like a seesaw.. Quote Link to comment Share on other sites More sharing options...
BLOZ UP Posted February 11, 2011 Share Posted February 11, 2011 (edited) OK - I know you know more about this stuff than I do - but I need to ask you to explain the differences in amplitude that I'm seeing on a recurrent basis as the wheel goes around. On an electric guitar pickup (which is exactly what the VR sensor is) - the closer the strings are to the pickup, the hotter the output signal. In a perfect world, every tooth the sensor reads "should" output a consistent +5v. I'm getting lows down in the 3's and then up to 5's and then back down again. If logic follows here with the guitar example - the closer the sensor is to the teeth, the hotter the signal it gets. Conversely, the further away from the pickup/sensor, the less voltage is going to be generated. Your engine speed would have to be very, very, consistent for the amplitude to stay even. Changes the voltage depend on the change in magnetic flux. The faster the reluctor spins, the bigger change in flux (over time), and then the bigger change in electramotive force. I'm not sure whether the range your seeing is or isn't normal, but it explains why it is different. Edited February 11, 2011 by BLOZ UP Quote Link to comment Share on other sites More sharing options...
BLOZ UP Posted February 11, 2011 Share Posted February 11, 2011 Did you scope the sensor itself or the circuit output? The output would be a squareish wave for MS, the sensor itself should be sinish. Quote Link to comment Share on other sites More sharing options...
2eighTZ4me Posted February 11, 2011 Author Share Posted February 11, 2011 I was "told" that it was raw data - but now it makes sense what you say about the circuit output. All I know is they said they took one sample before the VR conditioning board, and one sample after it. I've got a bunch of questions for those boys!! Hopefully, I'll get the car back tonight. Yesterday - without timing completely dialed in, at 15psi, it was making 300rwhp and 320 torque on a bone stock 80 ZX motor with 150K mi. on it. Curious to see what they were able to squeeze out of it at 20psi and with all the timing dialed in.... I gots me some learnin' to do..... Quote Link to comment Share on other sites More sharing options...
mobythevan Posted February 12, 2011 Share Posted February 12, 2011 I did some checking in the megamanual and for MSII it says: The missing tooth wheel decoder has been tested up to at least 18000 rpm for 36-1 and 60-2 wheels (yes, 18 thousand rpm) on the bench, and works fine. So, MSII easily supports the 60 tooth wheel. Maybe the VR conditioner circuit does not work, but the processor is capable. Quote Link to comment Share on other sites More sharing options...
madkaw Posted February 12, 2011 Share Posted February 12, 2011 Maybe you could try something like this; http://forums.hybridz.org/index.php/topic/90889-wasted-spark-ignition-no-edis/ but drill 36 holes, that's what I am going for. The hall effect sensors are 20$ and very accurate to very high RPM's. Quote Link to comment Share on other sites More sharing options...
Isaiah Posted March 11, 2011 Share Posted March 11, 2011 I'm doing the same on a flywheel; 36-1 ideally but 12-1 minimum. Plan to run it over to a machine shop next week and maybe shave off a # or two also. The other solutions: Derek's wheel + AC pulley || BJH -($$$$) were not as elegant to me as X64v's solution. Its been done by the OEM for EDIS elsewhere http://www.kolumbus.fi/ulpu.tuominen/JT_kuvat/flywheel.jpg Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.