280Z28 Posted April 1, 2006 Share Posted April 1, 2006 My controller hardware uses open collector TTL digital outputs that I'm trying to hook up to the various parts of my car that need controlling. Up first: connecting a digital output to an ignition coil. Two options: 1) Integrated circuit like the ones on this page 2) Discreet such as the SCR controller shown on this page Up next: stepper motor (IAC) Should I just use an L293D IC (double H on chip) Quote Link to comment Share on other sites More sharing options...
mobythevan Posted April 2, 2006 Share Posted April 2, 2006 That is where the GM module is pretty nice because it gives you the power transistor and built in dwell control on the 4 pin module for ~$12. Quote Link to comment Share on other sites More sharing options...
280Z28 Posted April 2, 2006 Author Share Posted April 2, 2006 Not sure what GM module you're talking about, but let's talk about the auto dwell control. It seems like it could cause some major problems in a fully electronic ignition setup. Say I want the spark at 12* advance. I calculate 5ms of dwell means I pull the coil transistor low at 21* advance, then cut it off at 12* to create the spark. This would be manual dwell control but I'm sure of where the cutoff/spark occurs. If I have an auto-dwell circuit in there, where do I trigger the coil charge to assure accurate spark location? Also, I have a coil-per-cylinder. $12 x 8 adds up! The ST Micro circuits would run me less than half of that, and take less space! Quote Link to comment Share on other sites More sharing options...
mobythevan Posted April 2, 2006 Share Posted April 2, 2006 That is fine, I'm not saying your a dummy or anything for wanting to set up your own power transistors. Since you didn't mention 8 coils, one $12 GM module would have worked nice. If you are running coil on plug are you going to use 4 triggers from MS and fire two transistors at a time, since 4 ignition triggers seems to be the limit? I think we are going to use four EDIS 4 post coils on our 2004 HEMI because it has 16 plugs. There are plenty of circuits on the net for the L293D stepper controller, so that should work fine. Quote Link to comment Share on other sites More sharing options...
280Z28 Posted April 2, 2006 Author Share Posted April 2, 2006 Oh I wasn't being defensive, I'm just trying to figure out how people make use of auto-dwell control ICs in electronic ignitions. My setup is nothing like the MS, except for the fact that I made it myself. The controller program is written in C# and the interfaces are all USB2. Somewhat pricey overall, but my goals are different and have you ever seen a gauge panel done in OpenGL? Quote Link to comment Share on other sites More sharing options...
mobythevan Posted April 3, 2006 Share Posted April 3, 2006 Oh I wasn't being defensive, I'm just trying to figure out how people make use of auto-dwell control ICs in electronic ignitions. No problem, to tell you the truth I am not sure how the auto dwell of the 4 pin module works, but I'm sure James on the MSnS site could explain it. I am actually running with dwell control in the MSnS unit. I just know that the 4 pin GM module does provide an auto dwell (which must be tailored to V8 rpm range 0-4500 and GM HEI coil) and the 7/8 pin GM modules do not have any dwell control. Good luck on your setup, sounds like fun. Quote Link to comment Share on other sites More sharing options...
z-ya Posted April 3, 2006 Share Posted April 3, 2006 You can drive either one of the ST (VB325, VB326) devices with a 5V TTL output. The high input threshold (VINH) is 4v, so as long as the input on this device goes higher than 4V, it should work. I've used these Nology ignition modules with good luck. These drive the MSD GM replacement coils just fine. They also take a 5V input signal (TTL). http://nology.com/module.html Quote Link to comment Share on other sites More sharing options...
280Z28 Posted April 3, 2006 Author Share Posted April 3, 2006 What's a recommended circuit for the following: I need accurate triggering of a counter input to get the engine speed. It needs to support the ignition reference from the GM HEI, and needs to support one of these (triggering once per tooth after de-bouncing): http://www.emi.cc/triggerkits/220-72707.html Quote Link to comment Share on other sites More sharing options...
280Z28 Posted April 3, 2006 Author Share Posted April 3, 2006 You can drive either one of the ST (VB325' date=' VB326) devices with a 5V TTL output. The high input threshold (VINH) is 4v, so as long as the input on this device goes higher than 4V, it should work. I've used these Nology ignition modules with good luck. These drive the MSD GM replacement coils just fine. They also take a 5V input signal (TTL). http://nology.com/module.html[/quote'] Maybe I'm crazy but are they seriously charging $75 for a TTL driven transistor for a single coil?? I'm looking for an embedded application anyway. DIP modules are best for my protoboard (the ST chips are surface mount). Are there any other ignition ICs to consider past the VB235/236? Quote Link to comment Share on other sites More sharing options...
mobythevan Posted April 4, 2006 Share Posted April 4, 2006 What's a recommended circuit for the following: I need accurate triggering of a counter input to get the engine speed. It needs to support the ignition reference from the GM HEI, and needs to support one of these (triggering once per tooth after debouncing): If I was setting that up I would use an opamp circuit to prepare the signal before running a ttl counter or processor counter, not sure if you are using discrete ttl parts or some kind of micro-controller for the counter. I love op amps and since they say it is a magnetic pickup (not optical which could be made into a ttl signal to start with) that is what I would suggest. I didn't dig on there site to see if they have an electrical specs on the sensor output, so just run an opamp as a threshold detector with adjustable hysteresis at the crossover point and run it on 0v and +5v so its output will be ttl compatible. You may have to hunt around for a stable threshold for that sensor, but then it would be set. Then the opamp can serve the purpose of debouncing the trigger and converting to +5volts. If you like this idea but haven't worked with opamps much let me know and I can try to put something together for a schematic and part numbers, these are very cheap, like an LM311 and supporting parts. Its just my initial thought. Opamps will be the base for all kinds of neat car circuits to condition input signals before passing them to an ECU, like knock sensor, speed sensor, vehicle weight sensors (load cells), accelerometers, thermocouples, etc, etc. I have thought a few times about having load cells on the car to have accurate weight at any time , my dad drove a semi that was set up that way, really neat when it was being loaded. Now I sound like Tony D, Tony where are you, ha ha Quote Link to comment Share on other sites More sharing options...
280Z28 Posted April 4, 2006 Author Share Posted April 4, 2006 I've worked with opamps some in the past. I've got my BS in EE, and I've been accepted into the PhD program which I'm starting in the Fall. However, all of my tech areas have been and are in DSP, communications, networking, control systems, and software... so this hardware stuff is hurting my progress on my computer like you wouldn't believe. If I can get a signal to these National Instruments devices that I know won't break them, I could have a running car on my own software in a day or two, and then I could focus on the new SOFTWARE features I've been researching like crazy. I would LOVE some help with the hardware interface design. It's what I desperately need before I can move this hardware to my car and start driving it again. Quote Link to comment Share on other sites More sharing options...
mobythevan Posted April 4, 2006 Share Posted April 4, 2006 Here is a circuit to get you started, but you need to determine 3 things now, 1. Amplitude of the input sine wave from the magnetic sensor to ensure that a gain of 5 gives you a large enough amplitude for the second stage level detector so it is easy to size your resistors. If you need a different gain let me know and I'll put up the equation. 2. You then need to select Vref, R, and nR to pick a mid crossover point in your sine wave and however much hysteresis you want to accomplish the debouncing. 3. Depending on the rate of triggers from that wheel at your maximum rpm, make sure you do not exceed the slew rate of the 741 opamp, otherwise pick a faster slew rate opamp. This one will output a 0-5v signal at a 10us edge rate. I would think that would be fast enough. I would start with this and measure the magnetic sensor sine wave from the sensor directly, then from the output of the first stage and make sure it is amplified by 5, then pick values for the second stage and measure the final output to make sure it is a useable TTL type signal. It won't hurt my feelings if you do this a different way, but you will never regret becoming more familiar with opamps. you can view a larger image in my photo album. Quote Link to comment Share on other sites More sharing options...
280Z28 Posted April 4, 2006 Author Share Posted April 4, 2006 My car won't see over 7k rpm, but I want the trigger system reliable to 10k rpm. The 60-2 tooth trigger wheel gives a tooth rate of 10KHz. The slew rate you gave me indicates a nyquist frequency of 50KHz... not a problem. Random question: what's the difference between a high side driver and a low side driver? I need a TTL output compatible driver for my fuel pump & fan relays. I'm actually hearing stuff about solid state drivers replacing the mechanical relays for reliability and load feedback to test for error states (broken load connection, useful warning for a fan for example). Quote Link to comment Share on other sites More sharing options...
z-ya Posted April 4, 2006 Share Posted April 4, 2006 The Nology unit is more than just transistor with a TLL input level threshhold. There are at least three coil drivers in there. I use them because they are all sealed and ruggedized. My customer pays for it. For me to build one would cost a heck of a lot more than $75 at my hourly rate. What's a recommended circuit for the following: I need accurate triggering of a counter input to get the engine speed. It needs to support the ignition reference from the GM HEI, and needs to support one of these (triggering once per tooth after de-bouncing): You can also just use a non inverting Schmitt trigger device. A ST has a high and low going threshhold voltage. Once the threshold is crossed, it will switch. Iit will trigger on a high going pulse, but will not trigger again until a low going pulse occurs with the proper threshold. This garantees a clean output pulse. Moby, your op amp circuit will work, but requires a lot more external components. Hear is your basi 74LS14 hex SC inverter: http://focus.ti.com/lit/ds/symlink/sn74ls14.pdf Connect two inverters to make a non inverting SC (I know I did't have to tell you this) Random question: what's the difference between a high side driver and a low side driver? I need a TTL output compatible driver for my fuel pump & fan relays. I'm actually hearing stuff about solid state drivers replacing the mechanical relays for reliability and load feedback to test for error states (broken load connection, useful warning for a fan for example). A high side drive provides a positive voltage to the coil, where a low side driver provides the ground to the coil (drives it low). For driving relay coils, use an open collector (OC) TTL driver. Connect one end of the relay coil to +12, and the other to the output of you OC driver. The OC driver will drive the relay coil - low. This is the prefered way (and output safe) to drive coils You can use an 74LS05 hex inverter with OC outputs. Just verify that the 74LS05 can source enough current to energize the relay coil, and use two to make a non inverting gate. http://focus.ti.com/lit/ds/symlink/sn74ls05.pdf Quote Link to comment Share on other sites More sharing options...
mobythevan Posted April 4, 2006 Share Posted April 4, 2006 Moby, your op amp circuit will work, but requires a lot more external components. yeah, no problem, I just always think in opamps. Quote Link to comment Share on other sites More sharing options...
280Z28 Posted April 4, 2006 Author Share Posted April 4, 2006 The Schmitt triggers are cheap too. I like that! The outputs of my NI devices are OC outputs. I have one of each of these: http://sine.ni.com/nips/cds/view/p/lang/en/nid/14605 http://sine.ni.com/nips/cds/view/p/lang/en/nid/14604 http://sine.ni.com/nips/cds/view/p/lang/en/nid/201737 Quote Link to comment Share on other sites More sharing options...
280Z28 Posted April 4, 2006 Author Share Posted April 4, 2006 Wait... pin R on the GM HEI is a square pulse that goes low at 10* BTDC on each cylinder?! Can I directly connect that to a TTL input and poll it to know the engine location? Quote Link to comment Share on other sites More sharing options...
z-ya Posted April 4, 2006 Share Posted April 4, 2006 When you drive the relay coil with an OC output, just be sure not to exceed the current sinking capability of the OC output. Use a current limiting resistor to protect the OC output. Pete Quote Link to comment Share on other sites More sharing options...
280Z28 Posted April 6, 2006 Author Share Posted April 6, 2006 Does anyone know the pulse width of the GM HEI reference pulse? Edit: I guess dwell would be the correct term. Quote Link to comment Share on other sites More sharing options...
280Z28 Posted April 8, 2006 Author Share Posted April 8, 2006 Well, I have working gauges now. They're ugly, but they work smoothly. I've tested it with the throttle body (which has been living on my desk for a while now) and it's dead on. I'll work on making things prettier (and 3D) once everything is running smoothly. I also ordered a Fluke 88V/A kit today. 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.