millsn Posted July 16, 2004 Share Posted July 16, 2004 I've an MBE ECU sending to my rev counter from a 94 ish car. There are 3 connections to the rev counter at the back of it. One is from the ECU, the other appear to be +ve and -ve. Question is, how doews the rev counter move, does it use the 12V from the car according to the output from the ECU? How should the ECU signal vary? Is it changing mA, Volts, mV or does the rev counter look for some other thing? Reason I ask is that mine winds round to 4500 before ceasing to register at all. Back off the revs and the rev counter springs bakc into life. As I don't understand what makes them work, Ive no idea why it's doing this. Anyone able to help? ☹️ Nigel Mills - 2.0 Zetec TB's. Up and running, still tweaking but it's great Link to comment Share on other sites More sharing options...
Chris W Posted July 16, 2004 Share Posted July 16, 2004 Nigel The rev counter is actually counting pulses from the ECU. So many pulses equals so many revs. Most rev counters contain some tiny switches which may be set such that the tacho can be used on 4, 6 or 8 cylinder cars and also on "wasted spark" systems (like the K series) where there are actually twice as many pulses as a "normal" ignition car. It is possible that your rev counter is playing up because somehow one of these switches has become set incorrectly and the tacho thinks the engine is running twice as fast as it is in reality and therefore going off the scale. Or there may be a component failure. The three connections are indeed +ve, -ve and input. The +ve is normally green, the -ve normally black and the input normally white/black. Chris 2003 1.8K SV 140hp see it here Link to comment Share on other sites More sharing options...
david nelson Posted July 17, 2004 Share Posted July 17, 2004 Chris why does my shift lights flash at 7000rmp on the rev counter when they it is set at 6500rpm. Can rev counters be inaccurate? thanks David Link to comment Share on other sites More sharing options...
Chris W Posted July 17, 2004 Share Posted July 17, 2004 David I have designed, built and installed a series of shift lights which I have fitted to many 7's. I usually find the rev counter is inaccurate (as measured against an oscilloscope) by as much as 10% either way. So it could be your rev counter is out by 7% (1 - 6.5/7.0) or the shift light is actually not set at 6500rpm as CC believe. Chris 2003 1.8K SV 140hp see it here Link to comment Share on other sites More sharing options...
Chelspeed Posted July 18, 2004 Share Posted July 18, 2004 Most tachos work by applying a voltage to a set of coils that drag a piece of magnetic material into the middle of the coil against the pull of a spring. The more revs you have the more volts are applied to the coils so the harder they pull against the spring and so the the needle moves further round the gauge. So if they are 90% accurate I think that's a miracle in itself! Being a mechanical man the mechanism for the pulsed frequency signal from the ecu being converted into a voltage is just a bit of magic. Tachos like the stack one use a stepper motor to drive the needle to the required place on the dial. They actually look at the pulsed input and say thats 5350rpm so tell the stepper motor to move the needle to the 5350 point on the dial, no springs or magnets. No doubt this is more expensive to build but I would think is inherently more accurate. Again the bit that reads the frequency signal and then drives the needle is electronic magic. Link to comment Share on other sites More sharing options...
millsn Posted July 18, 2004 Author Share Posted July 18, 2004 Essentially I've changed the ECU so where previously I used a Dunnell one which drove it without trouble I'm now on MBE. The counter now conks out at anything above 4500 rpm. I can't believe the ECU is at fault - why should EFI adversely affect the rev counter? Nigel Mills - 2.0 Zetec TB's. Got some cams coming soon... Link to comment Share on other sites More sharing options...
V7 SLR Posted July 18, 2004 Share Posted July 18, 2004 My rev counter worked perfectly until we were in consistently hot conditions in Italy last year, where it refused to work at all in the mornings (when it was cold) until ambient temperature had warmed up a bit. Basically, it stuck. Cheap rev counter. Mechanical problem. Both Chris and Graham are right about the operation of rev counters. It is unlikely that yours is set wrong. It's more likely to have a mechanical problem. They are pretty cheap ones when you look at alternatives. Link to comment Share on other sites More sharing options...
NickW Posted July 18, 2004 Share Posted July 18, 2004 Mmmmmm, that's interesting, I've got the same problem. Just changed from a dizzy with Lumintion ingition to a waisted spark setup. I'm driving the tacho thru a diode circuit from a coil pack to provide the kick-back voltage to the tacho. Tacho gets to about 2,700 and then drops to zero, if I drop the revs then the tacho starts again. Looks like a duff tacho then 🙆🏻 🙆🏻 Nick. Link to comment Share on other sites More sharing options...
millsn Posted July 18, 2004 Author Share Posted July 18, 2004 Can anyone advise where can you take your tacho to be tested to see if it's faulty? Nosense replacing it if the problem's somewhere else. Nigel Mills - 2.0 Zetec TB's. Got my cams, bugger it's quick Link to comment Share on other sites More sharing options...
Chris W Posted July 18, 2004 Share Posted July 18, 2004 Nick It's not necessarily a tacho fault. One would need to look at the pulses you are providing to the tacho on an oscilloscope. It could be that the pulses are getting so narrow above a certain rpm level that the tacho can no longer "see" them and therefore drops to zero. The fact that the tacho comes back to life below this figure is a strong indication to me that this is happening. Is this "diode circuit" something that you have recently added? Chris 2003 1.8K SV 140hp see it here Link to comment Share on other sites More sharing options...
NickW Posted July 18, 2004 Share Posted July 18, 2004 Chris, I've got a 1700 SuperSprint Xflow and I'm half way thru converting the ignition to a 3D mapped system using a Ford EDIS ignition system controlled by a bastardised MagaSquirt ECU. The diode circuit is used to take the two -ve signals form the two halves of the coil pack to the tacho. I've used a couple of 1N4004's and a 1N5253B 15v Zener. This circuit has been used before on other cars, I've taked to a guy who has it on a Spitfire. So the signal I am now feeding the tacho could be very different from what it's expecting, hence the behaviour?? Either way it looks like I need a new tacho, question is, do I get a more modern mechanical tacho that maybe can cope with the signal from the coils or do I get an electronic tacho and feed it from the EDIS module??? Or, do you think I could change the characteristic of the signal by changeing the diodes I'm using?? Now there's an electronics question for you..... Thanks, Nick. Edited by - NickW on 19 Jul 2004 00:08:48 Link to comment Share on other sites More sharing options...
Chris W Posted July 19, 2004 Share Posted July 19, 2004 Hi Nick Without a scope of course you're working blind. One can't see how good or bad the pulses are nor their shape etc. The tacho may need -ve pulses or it may need +ve pulses. It may need to be ac coupled (ie: connected through a capacitor) or it may be OK to dc couple it. One also needs to see if their width is changing with frequency too. It would be good electronic practice (and a fairly normal thing to do) to take the raw pulses from the coils and put them into a wave-shaping circuit which would remove all the "crap" on the signal and straighten out the edges and the tops (one such circuit for doing this is called a "Schmitt trigger"). Otherwise you are relying on the tacho's doing this and it may not have such an input circuit. With the shift light system I have developed, and installed on many 7's, I use +ve pulses clipped to around 5v using a zener and capacitively coupled into the input circuitry. If you are able to get over to my neck of the woods I would happily take a look at it for you. By connecting the tacho to a pulse generator, one would immediately know if the fault is the tacho or the circuit. Chris 2003 1.8K SV 140hp see it here Edited by - Chris W on 19 Jul 2004 10:21:05 Link to comment Share on other sites More sharing options...
NickW Posted July 19, 2004 Share Posted July 19, 2004 Chris, Thanks for the kind offer, but I don't think I would be able to get down to you in the forseeable future. I used to have a scope once but can't remeber what happened to it, must have given it back to were I pinched it from?? The tacho is a bog standard Caterham item c1990 and was taking the raw -ve from the original coil and working fine. I'm now doing effectivly the same, but with the Ford coil pack. Would you happen to have any info on a suitable Schmitt Trigger circuit that I could build to try and clean up the tacho pulses? Thanks, Nick. Link to comment Share on other sites More sharing options...
Chris W Posted July 19, 2004 Share Posted July 19, 2004 Nick Re the Schmitt trigger. It's difficult to say what is "suitable" because I don't know what the input spec of the tacho is. You will find many generic circuits for Schmitts on the web. It's basically a two transistor amplifier with the emitters joined together down to ground through a common resistor. It has the properties of switching at a very precise level on the rise and fall edge of the pulse and thus squares off any raggedness. Amplitude can be handled by a zener as you are doing now. Have you tried a smaller value Zener? Also contact the tacho manufacturer (if you can ascertain who this is) and email them for the input characteristics of the tacho. Chris 2003 1.8K SV 140hp see it here Link to comment Share on other sites More sharing options...
Chelspeed Posted July 19, 2004 Share Posted July 19, 2004 Said it was magic. Not understood a single word of the last ten posts. Link to comment Share on other sites More sharing options...
NickW Posted July 21, 2004 Share Posted July 21, 2004 Chris, A bit more info on the problem. I tried a smaller value Zener, but no change. If you take a look at the photos here you'll see the images of the signal from the original coil and the Ford coil pack I'm using. The images are very different and hence the problem. (Please exuse the website, I know it's sh1te, I just knocked it up in the last 5 mins) Interstingly, the signal doesn't really change when sampled infront of the Zener?? I beleive the operating range of the zeners I've used might be upto 200v, the peak of the kick-back signal is about 300v. Might I be overloading the zener so that it is not operating correctly? So, do you think the schmitt trigger will produce a signal similar to the original coil? Thanks, Nick. Link to comment Share on other sites More sharing options...
Chris W Posted July 21, 2004 Share Posted July 21, 2004 Nick Indeed the two inputs to the tacho are very different. Was there a diode circuit with the original coil? Can you post a diagram of your diode circuit and where you are connecting to the coils in the original case and the present case. At what rpm value were these photos taken? Chris 2003 1.8K SV 140hp see it here Edited by - Chris W on 21 Jul 2004 21:02:46 Link to comment Share on other sites More sharing options...
NickW Posted July 22, 2004 Share Posted July 22, 2004 Chris, The diode circuit is at the bottom of the page here The original coil just has the normal connections, +ve to the ignition switch and -ve to the tacho and the dizzy. With the new coil pack I have basically disconnected the two wires that went to the old coil, (+ve and -ve) and connected the +ve to the centre tap of the coil pack to provide it with 12v and the -ve is connected to the output of the diode circuit above. I've diconnected the plug to the Luminition ignition module. The photos were taken at about 800-900rpm. The only change in signal as the rpm increases is the pulses get closer together, I don't think the period of the pulse decreases, but can check again. I'll try and have a closer look later at the signals before and after the zener to see how different they are, I don't think the zener is doing much!! Thanks, Nick. Link to comment Share on other sites More sharing options...
Chris W Posted July 22, 2004 Share Posted July 22, 2004 Nick Thanks for that. I'll try to find some time today to study it in detail. The zener is blocking any pulses below its value (15v). It could be that the tacho is looking for that longish 6ms, 10v step in which case the zener would block this. It is not present in the new set-up scope shot so that might be the zener's doing it or it might just not be there in the new set-up. The pulse period in the original set-up is around 8ms (taking the time from the beginning of the 25v pulse step to the end of the 10v pulse step). At 900rpm this corresponds to a pulse repetition rate (prr) of 67ms. If the pulse width doesn't change (as you believe) this gives a prr of say 8 - 9ms max which this would put an upper rpm limit of around 6700rpm. Does this stack up with reality? Chris 2003 1.8K SV 140hp see it here Link to comment Share on other sites More sharing options...
Chris W Posted July 22, 2004 Share Posted July 22, 2004 Nick An afterthought. If you connect to the Megajolt Tacho output connection do you get any activity from your tacho? have you looked at the scope traces from this output? Chris 2003 1.8K SV 140hp see it here Link to comment Share on other sites More sharing options...
NickW Posted July 22, 2004 Share Posted July 22, 2004 Chris, Your calulations look about right, the max rpm of the tacho is 7000, close enough. I've just checked the signal from the new setup and the period of the pulse reduces very slightly with increased rpm. The signal before the zener does not have the extended signal at 10v as does the original coil output, so all the zener is doing is altering the voltage of the signal. I removed the zener from the circuit and the tacho does not work at all from the output of just the two 1N4004 diodes, so I'll try a 10v zener to see if that makes a difference. I don't have the MegaJolt ECU at the minute, shoudl have in a few days. There is a tacho output from the EDIS module, but there is currently no connection on the harness I have, so would need to bodge something together to test this. Thanks, Nick. Link to comment Share on other sites More sharing options...
Chris W Posted July 22, 2004 Share Posted July 22, 2004 Nick If you're using a wasted spark system, you will be feeding the tacho twice as many pulses per rpm as it is normally expecting. Have you tried changing the dip switches on the tacho to compensate for this? This may be the issue. If it's a standard CC instrument made by Caerbont try here Chris 2003 1.8K SV 140hp see it here Link to comment Share on other sites More sharing options...
NickW Posted July 22, 2004 Share Posted July 22, 2004 Chris, The tacho is a VDO unit, can't see any switches or access panel for switches. I can't see how I'm giving the tacho any more pulses than it had before. The coil pack has two halves, one side drives plugs 1&4 the other side drives plugs 2&3, so I should just be getting a -ve kickback each time the coils fire for each plug which is what the tacho needs. The tacho is accurate upto 2,700rpm where is dies. Thanks, Nick. Link to comment Share on other sites More sharing options...
Chris W Posted July 22, 2004 Share Posted July 22, 2004 Nick If you've changed to a wasted spark system (and you didn't have one before) you will have twice as many pulses now because a wasted spark system fires both coils at once (eg: 1 & 4) rather than a conventional system which would fire 1 then 4. So one plug fires on the power stroke and its partner fires on the exhaust stroke at the same time and so is "wasted". The reason for doing this is that it makes the ignition system a whole lot simpler if one only has to time for two spark discharges (ie: 1&4 and 2&3) rather than four as in a conventional system. Chris 2003 1.8K SV 140hp see it here Link to comment Share on other sites More sharing options...
millsn Posted July 22, 2004 Author Share Posted July 22, 2004 Warning -thread hijack. My rev counter has no reason to have played up. I've onl;y changed the ECU, not the ignition system yet it conks out at 4500. Why would it read correctky up until then and only then fail to count where it previously ran fine. I dont understand. Second question: Where do I get an oscilloscope from to check the ECU output? Nigel Mills - 2.0 Zetec TB's. Got my cams, bugger it's quick Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now