lammbn Posted January 10, 2009 Share Posted January 10, 2009 Ok. they made it way too hard to figure out something so simple as exactly what you can and can't do with an analog input signal. My wife really isn't so happy i'm spending so much time on the PC tonight. so what i figure is that you need at least 1700*F (~927*C) on your readout. but that gives you just under 10V output from the AD595 IC. so if you can drop that voltage by half across the scale then you should be able to use it with the brain. then its just a simple matter of using their program to double the voltage value read by the brain and then use their handy dandy included celsius to fahrenheit conversion factor already built in to the program and you should be golden. but because your dividing the voltage by two you are doubling the error factor of the chip to 6*C add that to the 4.88mV resolution of the brain and you get an error of about 10*C or 18*F i would guess. thats pretty high. only problem i see is that just adding a resistor to the output may not have the desired effect of cutting the voltage in half. i have no idea how the circuit inside the IC will affect the overall resistace over the spectrum and that just adds more room for error. but, like previously mentioned, if your only using this to measure the difference between the individual cylinders, not to get the exact temp of each, it should work fine. at least good enough to give you some warning before one of your cylinders goes too lean and burns down. Quote Link to comment Share on other sites More sharing options...
lammbn Posted January 10, 2009 Share Posted January 10, 2009 Oh, also forgot. you might want to check this web site out. i found it about a year ago and it's what got me started on all the EGT work i've done. i installed a similar readout with a custom alarm circuit in my buddy's Jeep for coolant temp. his didn't have a guage just a warning light. a very good read. http://www.autospeed.com.au/cms/article.html?&A=0360 Quote Link to comment Share on other sites More sharing options...
Daeron Posted January 10, 2009 Share Posted January 10, 2009 What the heck... WOW this thread kinda budded and thrived over the course of today, that was unexpected... "divorced" lol. I only posted about the EGT gauges to get sappy and talk about how I liked the dual needle sweep, because I'm used to seeing it in the old fast cars from when I was a little kid... (and I thought I helped to illustrate the low resolution obtainable from one of them, not contradict it) but now all of a sudden... carputer? I don't think I've bumped into that term yet, although the concept has been on my mind for nigh a decade now... must research soon, thank you! Quote Link to comment Share on other sites More sharing options...
Administrators RTz Posted January 10, 2009 Administrators Share Posted January 10, 2009 I only posted about the EGT gauges to get sappy and talk about how I liked the dual needle sweep, because I'm used to seeing it in the old fast cars from when I was a little kid... (and I thought I helped to illustrate the low resolution obtainable from one of them, not contradict it) but now all of a sudden... carputer? I don't think I've bumped into that term yet, although the concept has been on my mind for nigh a decade now... must research soon, thank you! Suddenly, I am overwhelmed by a sense of incredible warmth and extreme fuzziness. Sappy & lurid enough? Quote Link to comment Share on other sites More sharing options...
Derek Posted January 10, 2009 Author 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 Hi Carter Glad to see we finally found a subject that piques your interest. Thanks for the info. I checked Mcmaster for the thermocouples but they were a bit pricey. Probably better than the ones I'm looking at but still pricey! Ok. they made it way too hard to figure out something so simple as exactly what you can and can't do with an analog input signal. My wife really isn't so happy i'm spending so much time on the PC tonight. so what i figure is that you need at least 1700*F (~927*C) on your readout. but that gives you just under 10V output from the AD595 IC. so if you can drop that voltage by half across the scale then you should be able to use it with the brain. then its just a simple matter of using their program to double the voltage value read by the brain and then use their handy dandy included celsius to fahrenheit conversion factor already built in to the program and you should be golden. but because your dividing the voltage by two you are doubling the error factor of the chip to 6*C add that to the 4.88mV resolution of the brain and you get an error of about 10*C or 18*F i would guess. thats pretty high. only problem i see is that just adding a resistor to the output may not have the desired effect of cutting the voltage in half. i have no idea how the circuit inside the IC will affect the overall resistace over the spectrum and that just adds more room for error. but, like previously mentioned, if your only using this to measure the difference between the individual cylinders, not to get the exact temp of each, it should work fine. at least good enough to give you some warning before one of your cylinders goes too lean and burns down. Check out this schematic from the innovate site: And here's the description from Klaus who I think is an engineer there: "Attached is the AD595 circuit for EGT measurement. With a minimum of 13V supply voltage it measures up to 1100 deg C (2012 deg F). Set the LogWorks input to: 0V = 32 deg F (0 deg C) 5V = 2012 deg F (1100 deg C) The AD595 can handle a supply voltage of up to 36 Volts. So it can run on a car's 12V supply without any regulator. Minimum supply voltage required to go up to 1100 deg C is 13V, which should be no problem for most cars. Lower supply voltage just reduces the max temp, not the measurement. The AD595 outputs 10mV/deg C, up to a max of supply voltage - 2V. The divider circuit of R1 and R2 divides the AD595 output by 5/11. R1 and R2 should be 1% resistors. Wattage for the resistors does not matter (1/10th Watt is enough). The Zener diode protects the LM-1 if the AD595 fails. The capacitors of 0.1uF and 10uF filter noise. R3 is required only if the TK used has an isolated junction. On those there is no continuity between the TK wires and the TK body. Type K TKs have color coded wires. Yellow and red. I don't recall which one is which. If you hook them up wrong you won't damage anything, but it will not measure." You can see he's using R1 and R2 to reduce the output. I don't know if this method reduces accuracy or not. Keep in mind they're using this for EGT reading to a 0-5v gauge so it should work for the fusion brain. I found this place (I'm sure there's others) to make the circuit boards. http://www.expresspcb.com/index.htm I was toying with the idea of making them myself but for that kind of pricing it's just not worth it. I'm not sure how many inputs to have on the board. Once I'm done testing the manifold 6 inputs should be plenty for any temp sensing I want to do. One for EGT, one for oil temp, ok that's all that come to mind but I'm sure there will be others. Derek Quote Link to comment Share on other sites More sharing options...
Derek Posted January 10, 2009 Author Share Posted January 10, 2009 What the heck... WOW this thread kinda budded and thrived over the course of today, that was unexpected... "divorced" lol. I only posted about the EGT gauges to get sappy and talk about how I liked the dual needle sweep, because I'm used to seeing it in the old fast cars from when I was a little kid... (and I thought I helped to illustrate the low resolution obtainable from one of them, not contradict it) but now all of a sudden... carputer? I don't think I've bumped into that term yet, although the concept has been on my mind for nigh a decade now... must research soon, thank you! Yes it's my first divorce! So Sad..... Everything you ever need to know about carputers http://www.mp3car.com/vbulletin/ Right now mine has my music, GPS navigation, and I integrated Megatune into it. I'm working on a better solution than Megatune as the gauges are tiny tiney incy wincy little things! Derek Quote Link to comment Share on other sites More sharing options...
510six Posted January 10, 2009 Share Posted January 10, 2009 My setup is currently using a AEM EMS 1040 box with a modified wiring harrness to a Datsun L28et (now 3.0 liter) with a T67.There was a question because of the stock intake tends to make #1 , #2 cyl run run rich and due to running a methanol injection system using 2 160lb injectors in the intake track, running lean in cylinders 1and 2 became a real concern and since the wideband takes all the cyls A/F ratio into account a way was needed to check the difference between cylinder 1 and 6. The AEM and Autometer themocouples are pretty bulky so after a bit of searching a 1/8 npt thermocouple was found and a really slick and small 0-5v adaptor was found to allow the EGT`s to be dattalogged.Thanks to the versatility of the AEM 5% extra fuel was added to cyl #1 and 3% more to 2 and 3 after 4000rpm.The setup worked great and was able to monitor 12 different parameters in our dattalogs (which i will PM upon request).The EGT between 1-6 was within 15* at full boost, so the EGT became a usefull tuning parameter. The current setup will use E85 with six injectors on the intake manifold and an additional two in the intake tract. The two thermocouples in cyl 1 and 5 will be used to tune the engine. http://redlinegauges.com/product_info.php?cPath=22&products_id=20&dfwtekid=40842a4019067b873f980fa4a447c67b http://webstore.o2simulator.com/index.asp?PageAction=VIEWCATS&Category=6 http://www.cardomain.com/ride/2071134 Quote Link to comment Share on other sites More sharing options...
Tony D Posted January 11, 2009 Share Posted January 11, 2009 The AEM and Autometer themocouples are pretty bulky so after a bit of searching a 1/8 npt thermocouple was found and a really slick and small 0-5v adaptor was found to allow the EGT`s to be dattalogged. "Really Slick and Small 0-5V Adaptor?" Puke info, dude! I didn't see it in the three links, and that sounds like something I can add to my Megasquirt!!! Daeron, good photo, it does show exactly what I was saying about 1280 vs 1320, I wasn't saying you were contradictory. At least this time! But back on that 0-5 vdc adaptor.... That in conjunction with the datalog package for the old spare laptop would make some nice data acquisition for some projects that are coming up! Quote Link to comment Share on other sites More sharing options...
510six Posted January 12, 2009 Share Posted January 12, 2009 http://webstore.o2simulator.com/inde...ATS&Category=6 You can contact o2 simulator and ask them if they still make the EGT adaptors. I paid $ 40 apiece for mine. Quote Link to comment Share on other sites More sharing options...
lammbn Posted January 12, 2009 Share Posted January 12, 2009 cool. I was stuck at drill for the national guard all weekend and had no access the the outside world. But it looks like everybody got it figured out already, I'm happy about that. I'm not very good with this electronics stuff, just proficient enought to fake my way by on stuff like this and that. and thats why i'm a mechanical engineering major not an electrical! Quote Link to comment Share on other sites More sharing options...
Pharaohabq Posted January 12, 2009 Share Posted January 12, 2009 Wow. N I thought I was geeky... hehehe Well, it all looks like it is a doable situation, though from what I understand MS doesn't do sequential injection so even if you find that your cylinders aren't even, how are you to fix them without changing to some more expensive EFI computer or piggy back? I'm still a fan of GM EFI, but without a piggyback, you can't change the individual cylinders either. Seems this project has become much more expensive. Your Carputer looks awesome! I'd thought many times about building one. Is yours touch screen? it certainly mates your Car requirements: Look cool.← Look really cool ← Actually run ← yup passed phar Quote Link to comment Share on other sites More sharing options...
Derek Posted January 12, 2009 Author Share Posted January 12, 2009 Wow. N I thought I was geeky... hehehe Well, it all looks like it is a doable situation, though from what I understand MS doesn't do sequential injection so even if you find that your cylinders aren't even, how are you to fix them without changing to some more expensive EFI computer or piggy back? I'm still a fan of GM EFI, but without a piggyback, you can't change the individual cylinders either. Seems this project has become much more expensive. Your Carputer looks awesome! I'd thought many times about building one. Is yours touch screen? it certainly mates your Car requirements: Look cool.← Look really cool ← Actually run ← yup passed phar Thanks Phar The individual EGT sensors are merely a way for me to analyze the performance of the individual runners. Assuming the injectors are flow matched (supposedly they are) and Megasquirt delivers consistant amounts of fuel (it should) then any anomilies in temperatures can (sorta maybe) be atributed to intake design or assembly. The carputer has a 7" touchscreen. Derek 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.