Problem with tacho dying

[Russ Noble]

Another big milestone tonight! Fired up the motor and boy is she sweet!

There was really only one issue and that was every time we brought the revs up to 4000 the rev counter died. The reading returned to normal when we dropped the revs. If the motor was blipped harder the tacho crapped out at about 3500. Any one got any ideas where to start looking to cure this problem? Ignition is magnetic distributor and MSD 6AL. Tacho is Smiths motorsport 'classic'.

Any pointers most appreciated.

 

[John Martin]

A lot of tachs were designed to work with points. A point sytem will provide the tach with zero volts with the points closed and maybe anywhere from 8 to 14 (depending on wether an igntion coil ballast resistor is used) when open. With an electronic igntion, you only get 0.9 vdc when the unit is conducting (same as points closed, but not zero volts).

I think the MSD unit has a specific output for the tach instead of using the coil (-) terminal. If not, you may have to buy a tach adapter from Summit. Or you can build one.

Here's the diagram...ignore the fact it's for a sterndrive application (you can use a single resistor of 12K, or even higher. It's not that critical. Good diodes to use are P/N 1N4007 and may still be available from Radio Shack ...in the US):

http://www.vintageperformance.com/retrorockets/diodefix.pdf

With the diode adapter, you need 1.4 volts for the diodes to conduct. The 0.9 won't do it, so the tach sees "ground" through the resistor part of the adapter circuit.

 

[Dimi Terleckyj]

Hi Russ

When I was trying to hook up a VDO tach to my 5.4 I could not get it to work with very much the same symptoms as yours.

The problem was that the electronic ignition provides a multispark not just a single firing spark and as the revs increase the tacho cannot read the main spark among all the other spark clutter and so just shuts down.

My tacho was able to read the spark to about 1500 revs but would shut down if the revs went higher.

The solution was to fit a tacho driver module which connects to the crank trigger sensor and all was well after that.

Dimi

 

[Russ Noble]

Thanks for the info there guys.

Looks like I need tach adapter p/n 8920. I hope I can get this locally as a quick and instant fix.

If not I will have to build my own as John has suggested. Thanks for the diagram John. Not a good option right now as time is at a premium and I've no experience at doing that sort of thing. Needs must of course......

 

[Mike Drew]

Another big milestone tonight! Fired up the motor and boy is she sweet!

There was really only one issue and that was every time we brought the revs up to 4000 the rev counter died. The reading returned to normal when we dropped the revs. If the motor was blipped harder the tacho crapped out at about 3500. Any one got any ideas where to start looking to cure this problem? Ignition is magnetic distributor and MSD 6AL. Tacho is Smiths motorsport 'classic'.

FWIW both my Pantera and Cobra have MSD ignitions, and both require the MSD 8920 tach adapter.

Also, if you leave your engine running, and try to kill the power with your remote battery shutoff switch, you'll let all the smoke out of your tach adapter.

Don't ask me know I know this. :goofy:

 

[Ron Earp]

Also, if you leave your engine running, and try to kill the power with your remote battery shutoff switch, you'll let all the smoke out of your tach adapter.

:goofy:

Doh!

That is concerning as I do have a tach adapter for the Lola tach. Dumb Stewart Warner gauges I bought, I should have got Autometer stuff and called it a day.

How is your kill switch wired?

 

[Russ Noble]

If not I will have to build my own as John has suggested. Thanks for the diagram John. Not a good option right now as time is at a premium and I've no experience at doing that sort of thing. Needs must of course......

Thanks for the replies guys. I can't get an 8920 in good ol' Godzone, so need some help please!

I have had a look at the diagram in Johns link and have checked my friendly local electronics store catalogue and have found a 13K, .5 watt resistor which John suggests will replace the two 24.7K wired in parallel. So far so good.

With regard to the diodes it shows two 400V 3A diodes in series. John suggests p/n 1N4007. In my local catalogue, that # refers to a bulk pack of four 1000V 1A diodes. There is a listing for a 400V 3A under p/n 1N5404. I presume I need two of the latter?

The Radio Shack #s from the diagram don't seem to be referenced here.

As you can tell I am a total bunny at this, but I'm sure I will have learnt a lot by the time the smoke clears......

With regard to all this fancy electrical stuff, are all diodes/resistors with the same values created equal? Or are some not suited to vibration or automotive use, or are some better quality than others?

And lastly, querying the diagram, John said use the MSD tach output instead of the coil -ve. Easy. But the other end shown going to a E.S.A. module I presume simply goes to the tach? Not somewhere else? In other words this whole thing just gets inserted into the signal wire to the tach.

With regard to Mike's comment about destroying the unit if the kill switch is employed with the motor running. The scrutineers will check that the kill switch DOES stop a running motor so I would be best to have this unit accessible and easily unplugged for the purposes of the test. At other times just ensure I stop the motor before isolating the power.

I must admit, this project has been/still is a massive learning exercise.....I am also amazed that a unit that can be purchased for about $US60 can be replaced by a handful of gizmos costing less than $NZ2

 

[John Martin]

A few notes:

The tach output from the MSD unit should be used. I am under the impression that only a single pulse per cylinder comes from that output, not multiple sparks as per the coil connection. If you havn't hooked it up this way, try it. You may not need an adapter at all.

Yes, from the diagram, the ESA connection will be the tach connection in your case. The coil (-) connection of the circuit goes to the tach output on your MSD box (I goofed on this in my previous post as I forgot about the "multiple sparks").

None of the components (values, P/Ns etc.) in the diagram are critical. 1N400x series diodes are very common. The higher the last number (where the x is) means the higher reverse voltage the diode can withstand (called PIV, or Peak Inverse Voltage). In theory, there should only be a 12 volt pulse inot the circuit, but from a coil, there is likely much higher voltage spikes along with the pulse. From the MSD output, you probably only get the 12 volt pulse, i.e. no spikes, so a lower PIV rated diode will be fine. But you do need two of them so it takes at least 1.4 volts before they will conduct.

I used the diode/reistor circuit as it was a cheap quick way to try things vs ordering a tach adapter and waiting two weeks or so for it to arrive. Tach adapters are avialable here:

Search Results for tach adapter - SummitRacing.com

There's various kinds, so you would have to check each one out. But the 8920 is typcially the one you need.

There's more information here, and it looks like the 8910 worked for them:

MSD Installation

If you still have trouble, I'd suggest calling the MSD Tech Line @ 915-855-7123 At a minimum, they can probably tell you which adapter best suits your hook-up.

ps: before doing anything, make sure your tach is properly grounded. The tach case should be elecrically connected to ground as well as any ground terminals on the tach (if any).

 

[Howard Jones]

I am using a Stewart Warner tack and driving it from the MSD tack output terminal without anything else. Also if you half the resistance you double the current and therefor also double the power. P= (I)(V) , or I = P/V , So you will need to make that resistor a 1 watt.

http://www.reuk.co.uk/OtherImages/ohms-law.gif

 

[Russ Noble]

A few notes:

The tach output from the MSD unit should be used. I am under the impression that only a single pulse per cylinder comes from that output, not multiple sparks as per the coil connection. If you havn't hooked it up this way, try it. You may not need an adapter at all.

That was the way I had it hooked up.

A few notes:
None of the components (values, P/Ns etc.) in the diagram are critical. 1N400x series diodes are very common. The higher the last number (where the x is) means the higher reverse voltage the diode can withstand (called PIV, or Peak Inverse Voltage). In theory, there should only be a 12 volt pulse inot the circuit, but from a coil, there is likely much higher voltage spikes along with the pulse. From the MSD output, you probably only get the 12 volt pulse, i.e. no spikes, so a lower PIV rated diode will be fine. But you do need two of them so it takes at least 1.4 volts before they will conduct.

Noted

A few notes
I used the diode/reistor circuit as it was a cheap quick way to try things vs ordering a tach adapter and waiting two weeks or so for it to arrive.

Thats my position too!

A few notes:

ps: before doing anything, make sure your tach is properly grounded. The tach case should be elecrically connected to ground as well as any ground terminals on the tach (if any).

Will double check that today.

I am using a Stewart Warner tack and driving it from the MSD tack output terminal without anything else. Also if you half the resistance you double the current and therefor also double the power. P= (I)(V) , or I = P/V , So you will need to make that resistor a 1 watt.

http://www.reuk.co.uk/OtherImages/ohms-law.gif

Howard, wont one 13K, 1/2 watt resistor carry approx the same power as two 24.7K, 1/2 watt resistors in parallel?

Thanks John and Howard for your replies.

 

[John Martin]

wont one 13K, 1/2 watt resistor carry approx the same power as two 24.7K, 1/2 watt resistors in parallel?

Yes, the total power dissapated would be the same, but in the twin resistor scenario, each resistor would dissipate 1/2 the total making it "easier" on each component. But I wouldn't worry too much about the resistor dissipation.

A steady 13.8 VDC across a 13K resistor would dissipate 14.7 mW of heat. As it is a pulse, and most likely no more than a 50% duty cycle, you would be looking at only about 7 mW of heat being generated.

Anyway, it's worth a try. It didn't for me with my S-W tach, but further testing revealed the tach had become defective. Stupid of me to try it really....if it used to work, and then didn't, adding a tach adapter wouldn't fix the problem. Hope you have better luck.

 

[Jac Mac]

Since the MSD6AL (around 20 discharges ) becomes '''' one single discharge----- or 1SD''' @ or above 3000rpm & this is where the problem starts to occur, perhaps by winding extra loop's into the sensor or pickup coil of the 'Smiths' revcounter/tacho it might be enough to trigger the unit with the lower voltage.
IIRC we had to reduce the number of loops/coils back in the good old days where we had things like points/condensers/''Sports'' coils etc & resulting strong signals.
Incidently I was just reading on the MSD tech forum that the tach adaptor sends a 300v 'spike' signal.

 

[John Martin]

Russ, I may have jumped the gun here. It didn't occur to me you may have a "current sensing" tach until I read Jac Mac's post above. The tach adapter circuit we've been discussing is for voltage sensing tachs (which most are).

Before you go any further, you may have to determine which type of tach your Smiths unit is. If it's a voltage type, then keep going. If it's a current type, you can play around with a coil of wire and a resistor, or a small signal tgransformer, or a bunch of other things that may require a lot of trial and error, or just call MSD and find out what they recommend for a tach adapter.

Sorry about that...I should have asked which type of Smiths tach you have first.

EDIT: Check out page 16 (of 20) to see Crane's take (and fix) on the older current sensing Smiths tachs:

http://69.20.53.62/pdf/90000700.pdf

 

[Howard Jones]

Brain math fart. Maybe I can redeem myself. Have a look at this. There are two possible hook ups for a 6al and a tach adapter. Current and voltage triggered. See second page. http://www.msdignition.com/uploadedFiles/MSDIgnitioncom/Products/Accessories/8920_instructions.pdf

 

[Malcolm]

This was a standard fault on some GTDs for a while. I seem to remember it was cured by switching from a volt driven tacho to a current driven tacho or vice versa, I can't remember which way round it was. But it is curable. Looks like you are on the way to sorting though from the posts above.

 

[Russ Noble]

Thanks for all the help guys.

I think I may have solved the problem. My Caerbont Smiths tach has two signal input wires. One is for 'standard' old school points and coil only system. The other is for ECU or MSD.

I have just checked the wiring and the wrong input wire has been used. I guess I should have checked earlier, but all the dash wiring was done by a mate who was helping me out. He made a beautiful job of it and I thought he had it all sussed. My bad for not going through it and double checking before now.

Next start up is planned again in two nights time, I am confident there will be no problem. Provided the wrong hook up hasn't caused an internal meltdown....... I will report back.

One bonus of this minor saga is that I have been forced to study things like resistors and diodes and they are no longer quite the mystery that they were.....Special thanks to John for his explanations of how things work.

Cheers,

Russ

 

[Ross Nicol]

You can name me Russ it's ok. Never assume an expert in electronics is incapable of making a mistake. Most of the time I'm able to fix my stuff ups because I can test, at the job completion. Of course I was unable to test your set up. Seems I would have been better off to leave the trigger wire disconnected rather than having an educated guess. I can understand how such a matter seems insurmountable to yourself with limited experience in electrics/electronics but to me it's an everyday possibility. My philosophy is -
if it doesn't work try something else. Believe it or not doctors seem to use the same diagnosis strategy.

Ross:thumbsup:

 

[Maurice Butler]

how is that engine rebuild of yors going ross?

I hope russ doesn't mind me hijacking this thread, its too late i've done it