Roff,

I've been struggling all day and can't get it working. That capacitor you added in the front there makes a difference. I had to change it from 10n to 100n so that the needle would start responding. The problem is the needle jitters. I think the duty cycle of the output signal is off.

My after market tach has a filter that allows me to connect it directly to the negative terminal on the coil and it works fine. However when I hook it up to the output of my circuit it jitters.

Here's my updated circuit. I also added your mods to it. I ended up adding a second 4093 instead of stealing part of the one already there. Man I wish I had a scope. Can anyone recommend a cheap simple scope I could buy somewhere?

BTW - I looked up the specs on my after market ignition system and it says it outputs a 12 volt square wave at 20% duty cycle.
I assume that means it stays at 12V for only 20% of a cycle and then 80% at 0 volts? Anyhow I noticed the 4017 counter has its outputs at what seems to be 100% duty cycle and one output comes on while the last is still on. I made a 555 circuit going at 1 hz and the used LEDs to see the output of the 4017. One LED comes on and stays on, the next LED comes on and the other goes out what seems to be simultaneously.

Thanks guys
Malcolm

Attached Images

tach_mod.png (13.4 KB, 10 views)

 

Going back and looking at that scope output for a cars ignition it would seem I need a zener diode of around 5v on input (the one you added to my circuit). However all I have is a 15v zener and maybe that's my problem? I will see if I can pick up some 5v zeners on Monday.

Thanks
Malcolm

 

Malcolm, a 5V zener won't work. The circuit I drew (with the 15V zener) should work fine, as should the RC filter. Your logic following the 4017 is the problem. Try the circuit below. This should give you 3 pulses at the output for every 4 input pulses, and the output pulses should be the same width as those from the ignition.
You should probably use coax as the connection from the ignition to this circuit's input, as you need a good ground return path, and the shield may help keep EMI from the high voltage of the ignition secondary from appearing on the input signal.

__________________
Ron (aka Rube)

 

Hey Roff, did you forget to post the circuit? I don't see one?

Thanks
Malcolm

 

Oops! I'll post it in a little while.

__________________
Ron (aka Rube)

 

OK, better late than never, I guess. Here's the tardy schematic. Hopefully the two parallel gates will have enough drive for your tach.

EDIT: Corrected error in schematic and added timing diagram. I drew the duty cycle as 20%, since you mentioned that, but it should work with any duty cycle. Note that, at 33.3Hz, there is a component in the output at 8.3Hz, due to the fact that you are dropping every 4th pulse. You may see jitter in the tach at low RPMs.

Attached Images

tach5 with timing.PNG (34.2 KB, 20 views)

__________________
Ron (aka Rube)

 

Thank you for the diagram. Can you help explain how some of it works for me? I see we now have a 4023 chip in there. A quick search on google says it is a tripple input NAND gate (which I gathered from the symbol on the diagram) but it also says it has a buffered output. I'm not sure what it means if the output is buffered?

Also I'm not sure I understand how that very last 4023 NAND gate will be driven from the 4017 (buffered through the three 4093 gates) as each of the three inputs on that last 4023 will go high one at a time and never all at the same time right?

Thanks again for the help. I will pick up a 4023 and see if I can get this circuit built and tested.

Malcolm

 

See my previous post. I edited the schematic (moved one of the 4023s) and added a timing diagram, so you can (hopefully) see how it works.
A buffered gate just means there are internal inverters following the NAND gate. This allows the output drive strength to be kept relatively high while still keeping the input capacitance low. Also, without the inverters, the output drive strength will vary depending on the logic levels present on the various inputs.

Attached Images

CD4023 sch.PNG (9.0 KB, 20 views)

__________________
Ron (aka Rube)