Derek Posted January 9, 2009 Share Posted January 9, 2009 Divorced from Derek's cast intake manifold thread http://forums.hybridz.org/showthread.php?t=129941 --RTz Thanks guys. Sounds like I'm heading in the right direction. I'll pull the vacuum for the map sensor from the log. Since this is a test mule of sorts would it behoove me to weld in 6 O2 sensor bungs in the header while I have it off. This would allow me to read each individual cylinder. I'd have to move an O2 sensor from hole to hole but it's doable. I can make the bungs up on the lathe so it's not a big deal. Any thoughts? Derek Quote Link to comment Share on other sites More sharing options...
garvice Posted January 9, 2009 Share Posted January 9, 2009 Excuse my ignorance, but I thought you put the O2 sensors in the exhaust? Isn't it to determine the air fuel mixture after it is burnt? So that you know if you are running rich or lean. Quote Link to comment Share on other sites More sharing options...
naviathan Posted January 9, 2009 Share Posted January 9, 2009 I'm curious how you will adjust the AFR for the individual cylinder as most injection controls do batch fire which wouldn't give you a cylinder by cylinder adjustment? With Harleys, because of the way the cylinders are lined up the rear cylinder has a tendacy to run hotter, therefore leaner than the front. When I do my carb adjustments I typically adjust for the rear cylinder and let it run at that. The front plug will run a bit rich, but I pull the plugs every so often, clean and switch them around to keep things from getting fouled. I suppose another option would be to identify the cylinders that are running hotter and put a cooler plug in and vise versa? Quote Link to comment Share on other sites More sharing options...
garvice Posted January 9, 2009 Share Posted January 9, 2009 Just get yourself a decent computer and run full sequential ignition and injection. That way (with an O2 bung in each exhaust) you can tune each cylinder precisely. Quote Link to comment Share on other sites More sharing options...
naviathan Posted January 9, 2009 Share Posted January 9, 2009 Just get yourself a decent computer and run full sequential ignition and injection. That way (with an O2 bung in each exhaust) you can tune each cylinder precisely. What systems, that don't cost a crap load, can do that? Quote Link to comment Share on other sites More sharing options...
garvice Posted January 9, 2009 Share Posted January 9, 2009 Well, that depends what you consider a crap load. We are definitely not talking Megasquirt type costs here. I've got sequential ignition and injection with my Motec (Haven't gone to the trouble of custom tables for each cylinder as I can't measure the exhaust from each cylinder), Haltech E11 will do sequential ignition and injection, but not sure if you can customise for cylinders. I would think they would have a trim on each cylinder that you could activate. And with the Aussie dollar being so bad at the moment, either of these brands should be looking a bit cheaper for you guys at the moment. Quote Link to comment Share on other sites More sharing options...
naviathan Posted January 9, 2009 Share Posted January 9, 2009 The USD isn't exactly spanking other currencies right now either and I see it getting worse very soon. Quote Link to comment Share on other sites More sharing options...
garvice Posted January 9, 2009 Share Posted January 9, 2009 The Aussie dollar is most often referrenced to the USA dollar. When I said our dollar was not doing well, I meant that it wasn't doing well to the USA dollar. Haltech and Motec are both Australian Computers. Hence the comment. Sorry Derek, it is late over here, I will stop taking up all your thread. Quote Link to comment Share on other sites More sharing options...
Derek Posted January 9, 2009 Author Share Posted January 9, 2009 The Aussie dollar is most often referrenced to the USA dollar. When I said our dollar was not doing well, I meant that it wasn't doing well to the USA dollar. Haltech and Motec are both Australian Computers. Hence the comment. Sorry Derek, it is late over here, I will stop taking up all your thread. No worries (in my best yank does Australian accent)! The individual bungs are in each tube of the header so that after I have the car running and tuned I can then analyze the mixture on each cylinder by screwing in an 02 sensor and hooking it to megasquirt (or a standalone reader if that's problematic). That way if there's a glaring problem in the design I'll be able to gauge weather changes I make are making a difference. Derek Quote Link to comment Share on other sites More sharing options...
Pharaohabq Posted January 9, 2009 Share Posted January 9, 2009 ... The individual bungs are in each tube of the header so that after I have the car running and tuned I can then analyze the mixture on each cylinder by screwing in an 02 sensor and hooking it to megasquirt (or a standalone reader if that's problematic). That way if there's a glaring problem in the design I'll be able to gauge weather changes I make are making a difference. Derek Ahh, okay. This is more for a test bed, you're not going to be using 6 individual O2 sensors (read expensive) but that you'll have bungs in your header (i'd imagine brass plugs in the unused bungs) so you can move the O2 sensor from cylinder to cylinder and thus read how well each is firing.. right?? only problem I see with that, is that megasquirt will adjust your mixture based on what it reads from the O2 sensor. You could read the voltages manually (separate from MS) if you got a meter or Oscope and used an additional O2 sensor, and that might tell you exactly how each cylinder is firing, but you'll see differences based on a number of factors including the compression of each cyl, the engine temp at each measurment as well as I'm sure a few unknowns. Yes you'll see how well each cyl is burning it's fuel, but it will be harder to note if it's the manifold, or if it's the engine making the difference. I guess if you test each cyl like that before you swap then after you'll get a better idea of the efficiency. Sounds like a lot of extra work to me. As he said above the O2 sensor usually goes furthor down the line (OBDII uses 2, one befor the cat and one after) to measure efficiency and adjust mixture. Sequential vs batch firing EFI is a whole different animal. You get better mileage with sequential. but MS doesn't do that yet. You could swap to a GM engine computer and get an aftermarket harness, n it could possibly work. but since you've already started with MS, you probably shouldn't change horses so to speak. (tho I'm a fan of the GM Efi system) Phar Quote Link to comment Share on other sites More sharing options...
Tony D Posted January 9, 2009 Share Posted January 9, 2009 Testing the O2 out of each cylinder is kind of advanced tuning. For what you are after I think you won't need it. Realistically, a good EGT Probe and scanner like they use in light aircraft (Aircraft Spruce has them for Cessnas that run on 13.8VDC...) can monitor and give you a graphic readout of individual cylinder temperatures and differential. Before O2 sensors were commonly available, that was THE way to tune the mixture and test airflow through each cylinder. If you can see less than 25 degrees differential cylinder-to-cylinder during a steady load (like climbing a tall hill in 4th gear at 3/4 throttle or WOT...your airflow is balanced and no tweaking of your horns would be required. The probes and kits for simple rotary switch setups for EGT reading are available pretty reasonably (far less cost than the aircraft stuff) and if you have a buddy reading and writing numbers as he flips the switch from cylinder to cylinder you will get a darned good idea of variation. Six probes, leads, rotary switch and Decently-Sized DIGITAL READOUT (don't even WASTE your time with an analog gauge in this application if you are serious about analysis, you need ACCURATE numbers like 1285F versus 1320F which on most analog gauges will be almost identical...) The holes in your header would be less than 1/8" and the probes would go in VERY close to the mating flange of the head---you could even get longer thermocouples and put them so they go into the exhaust port and get even closer to the exhaust valve for even more accurate readings. There is an integrated thermocouple/stainless steel hose clamp arrangement that lets you drill the small holes in the header, then install the assembly over the hole tightenting the hose clamp to seal the whole thing back up again for driving. Small holes are good. I like small holes. You would be amazed to see what misfires are occurring and can be read with such a 'primitive' setup. Sequential? Let's not even go there! Quote Link to comment Share on other sites More sharing options...
Daeron Posted January 9, 2009 Share Posted January 9, 2009 This car wants EGT sensors, not O2. Thats my gut feel at least. And not to contradict what tonyd said about analog gauges, but to me one of the [pinnacles of having a nice, old skool setup is a dual EGT gauge with needles that sweep against each other : http://www.aircraftspruce.com/catalog/inpages/westduals2.php (you can see what he meant about the difference between 1285 and 1320) Quote Link to comment Share on other sites More sharing options...
Tony D Posted January 9, 2009 Share Posted January 9, 2009 That gauge is suitable for TIT TOT readout only. Someplace where you are looking for a 400 degree difference in steady-state loading. Look at Aircraft Spruces "Engine Analyzers" with 6 EGTs, plus TIT and TOT...they alert you to hot pairs, which cylinders come to temp first, you can set a differential alarm. Pricey, but if you are into analysis longterm (they can datalog onboard for two+ hours in some cases!) this is the thing you want. As for Analog---you show me how you will see 1280 and 1320 on that short sweep.... Not happening. For analysis you want hard numbers and a digital readout. It's not long term, it's developmental. I have one of those very gauges, and as I recalled, that gauge is even specified for TIT / TOT. I'm not a fan of the short-sweep Westach instruments. I've had a CHT gauge on #3 cylinder of the my VW Microbus since 1982 and while it's nice to look at the resolution SUCKS! I'd much rather have a stepper motor controlled Defigauge with a 270 degree sweep. http://www.aircraftspruce.com/catalog/inpages/ei-engineanalyzer3.php http://www.aircraftspruce.com/catalog/inpages/westcombos6.php I used one of these initially, but got sick of the Westech display/resolution quality, so I substituted a Digital EGT Gauge and used it with their thermocouples and rotary switch and was MUCH happier. For $280 it's not bad, but you can buy the individual components and use a much better digital K Type 'universal' thermocouple reader (IF YOU HAVE A DIGITAL MULTIMETER YOU CAN USE THAT!!!) and you will get much better empirical results from the readouts. The 100ms scan rate on my Fluke updates WAY faster than the analog gauge can (especially after switching the rotary dial!), with FAR better resolution! Down to .1 degree at the lower temps. You can use this setup to then check radiator in and out temperatures. Turbo inlet and discharge air temperatures. Oil cooler bypass temperatures. Once you have the rotary switch, and a good digital readout, you can buy thin thermocouples to stick just about ANYWHERE you want to quantify heat transfer efficiency. You can ready CHT on each spark plug with it with the right adapter rings for the plugs... For the cost it can be a VERY versatile tool. To be honest I stole mine from Uncle Sugar while doing a 500 hour on a Lycoming...just comdemned it and took it home. I'm not porud. For $647.43 a month, I figured I was probably saving someone's life removing it and putting a fresh one into the airframe. Yeah, that's it! Just as an aside here....which gauge do you think is going to get second billing for drivers eyepath in the Turbo Bonneville Car? Quote Link to comment Share on other sites More sharing options...
Careless Posted January 9, 2009 Share Posted January 9, 2009 a friend of mine just bought a ViPEC with QuickTune. he got the v44 which is sequential to four ouputs, but you can take a look at the v88. http://www.vi-pec.com/page_files/V88.html The QuickTune is really cool and seems to be quite effective. Quote Link to comment Share on other sites More sharing options...
Derek Posted January 9, 2009 Author Share Posted January 9, 2009 Here's what I'm thinking 6 of these Running into one of these http://www.fusioncontrolcentre.com/FusionStore//catalog/product_info.php?products_id=62 hooked up to this My current carputer With this skin But with 6 inputs labeled cylinders 1-6 I'm pretty sure this is doable. I just need to calculate whether the resolution of the fusion brain will give me an accurate enough reading. it's 10 bit over a 0-5v input if any of you geniuses out there know what the hell that means. I don't! The software has data logging over all the inputs. $59.00 for the brain 7.00 each for the thermocouples carputer already installed Cool factor........ Priceless Derek Quote Link to comment Share on other sites More sharing options...
garvice Posted January 9, 2009 Share Posted January 9, 2009 That means that it has 2^10 = 1024 different values to output. If you put in a 0-5V input it means that you can read a precision of 5/1024V = 4.9mV. So you could input 0V and get one signal, then put in 4.9mV and get a different readout etc, etc. EDIT: P.S. Impressed with the Carputer, well done. Quote Link to comment Share on other sites More sharing options...
lammbn Posted January 10, 2009 Share Posted January 10, 2009 Ok, i've been quitetly following this project of yours for quite a while now and I've very impressed with the project! but now i have to ask a question. i just spent the last hour and a half looking into this fusion brain thing. its very interesting and i think it will solve my problem of how to get additional features into my Mini ITX computer I'm going to run in my car without going through the very limited options megasquirt offers. mainly the amount of sensors i want to feed into it has been the problem. also the lack of quality megatune offers. anyways, how are you planning on feeding the K-type thermocouples into the brain? i've been searching their fourm section like mad on MP3car.com but can't find much. i have some very elaborate schematics i drew and tested for an standalone, digital EGT sensor readout with built in alarms but i have to use the AD595 IC to process the data from the thermocouple. are you planning on building a complete second board to process the data from the thermocouples and then feed it into the brain in a 0-5v format? not hard to do once you figure it out but it's a little bit of work. sorry to take the thread off course. Quote Link to comment Share on other sites More sharing options...
Derek Posted January 10, 2009 Author Share Posted January 10, 2009 Ok, i've been quitetly following this project of yours for quite a while now and I've very impressed with the project! but now i have to ask a question. i just spent the last hour and a half looking into this fusion brain thing. its very interesting and i think it will solve my problem of how to get additional features into my Mini ITX computer I'm going to run in my car without going through the very limited options megasquirt offers. mainly the amount of sensors i want to feed into it has been the problem. also the lack of quality megatune offers. anyways, how are you planning on feeding the K-type thermocouples into the brain? i've been searching their fourm section like mad on MP3car.com but can't find much. i have some very elaborate schematics i drew and tested for an standalone, digital EGT sensor readout with built in alarms but i have to use the AD595 IC to process the data from the thermocouple. are you planning on building a complete second board to process the data from the thermocouples and then feed it into the brain in a 0-5v format? not hard to do once you figure it out but it's a little bit of work. sorry to take the thread off course. Well about 5 hours ago I had no Idea what I was going to do! At this point I'm probably going with 6 of the AD595 on a separate board. Digikey has them for $11.00 each. I found this post on the innovate forum that has a lot of good info: http://www.innovatemotorsports.com/forums/showthread.php?t=371&highlight=ad595 You probably have this but I'll post it for others PDF for the AD595 http://www.analog.com/static/imported-files/data_sheets/AD594_595.pdf Am I correct in thinking that you can vary the range that it reads by changing the value of a resistor or am I confused. I was planning on getting the Fusion Brain anyway and I have a Fluke pyrometer so I can always read the thermocouples individually. We may want to spin this off to another thread if there's interest. Derek Quote Link to comment Share on other sites More sharing options...
clshore1 Posted January 10, 2009 Share Posted January 10, 2009 Hello Derek, long time lurker, first time poster here. Great project. If you need to do multi-channel data capture, and you have a laptop, it's hard to beat this for $25: http://www.dataq.com/194.htm The Chart Recorder data playback looks like an oscilloscope, but you have full control. And it will capture as much data as you have disk space to hold. Carter Shore Quote Link to comment Share on other sites More sharing options...
clshore1 Posted January 10, 2009 Share Posted January 10, 2009 Oops, almost forgot, if you use the AD595, remember the output is 10mV/degree C, not F. Also, look at McMaster-Carr for thermocouples. Carter 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.