Is a +5V Square Wave VSS Safe for an EEC-IV?

[Past_Miner]

I'm working through the problems associated with connecting an electric Autometer speedometer to my stock VSS with my EFI. I successfully proved that the signal is too weak if you split the same VSS to both and was finally able to nest two Ford VSS in series on my D20 after hacking the second one apart. It works but I am looking for a cleaner solution.

I found this old thread that indicates the +5V square wave signal the speedometer puts out is compatible with the EEC-IV:

efi vss & autometer Hall Sender

Back in those days Ryan McCormick indicated that it was ok to route that signal to the EEC. I normally take his EFI advice as gospel but its a bit dated in this case and I'm wanting to know if anyone has any feedback about doing this before I try it and risk frying the idle circuit in my EEC.

Any guidance would be appreciated.

 

[Digger556]

Disclaimer: I'm not an EEC-IV expert, but I do know about VR and Hall effect sensors. (Also I stayed at a Holiday Inn Express last night)

The problem I see is that VSS is a VR sensor that produces a sine wave with amplitude and frequency proportional to vehicle speed. VR sensors require a chip to translate the sine wave signal into a square wave which is useable by a microprocessor. (typically 0-5VDC)

The chip that does the signal conversion typically looks for the zero-crossing points, which is the point where the signal crosses 0VDC, going negative or positive. With a square wave, the voltage drops to 0VDC or slightly above 0VDC, but never crosses below that threshold. This can cause erratic performance. One solution would be to add a slight DC-bias to the signal so the output toggled between say -0.5V and 4.5V.

One other solution may be to use a 1:1 OP-AMP between the VSS and the Autometer speedo so it's input impedance doesn't drag down the signal voltage going to the computer.

 

[Past_Miner]

I read my original post and I should be a little more clear about the current situation. The Ford VSS (sine wave) will drive the Autometer speedometer if it is not connected to the EEC. When it is connected to the computer and the speedometer at the same time, the computer absorbs enough of the signal that the speedometer can't see a pulse until you are running about 60 MPH. Then the speedometer will suddenly jump to mid-range. The computer never skips a beat when wired this way. The signal is always strong enough for the EEC but it isn't strong enough for the speedometer.

One solution, which I have made work, is to piggy back two Ford VSS and send separate signals to the computer and speedometer. The speedometer sees pulses at all speeds. The problem with this on a Bronco with a D20 is that the second VSS has to be cut up quite a lot to clear the shift rail caps on the transfer case.

The Autometer speedometer has the circuitry described above built in. It can take a sine wave input and it creates a secondary square wave output that can be used for aftermarket cruise contracl and "some efi computers", according to the instructions.

I am thinking that the most elegant solution is to run the sine wave signal to the speedometer and then run the square wave to the EEC for the idle control circuit (which my truck has always needed to prevent stalls at stop signs).

All of the research I could do indicated that the pulses per revolution were not too critical for my EEC but I can't find a definitive answer about the +5V signal into the EEC. I'm really trying to understand if the +5V is too much for the computer.

 

[Digger556]

I read my original post and I should be a little more clear about the current situation. The Ford VSS (sine wave) will drive the Autometer speedometer if it is not connected to the EEC. When it is connected to the computer and the speedometer at the same time, the computer absorbs enough of the signal that the speedometer can't see a pulse until you are running about 60 MPH. Then the speedometer will suddenly jump to mid-range. The computer never skips a beat when wired this way. The signal is always strong enough for the EEC but it isn't strong enough for the speedometer.

Not surprising. It sounds like the input impedance mismatch of the two devices you are feeding is causing one to see good signal and the other to be below its operating threshold.

One solution, which I have made work, is to piggy back two Ford VSS and send separate signals to the computer and speedometer. The speedometer sees pulses at all speeds. The problem with this on a Bronco with a D20 is that the second VSS has to be cut up quite a lot to clear the shift rail caps on the transfer case.

This is why I suggested what is called a "voltage follower" to allow the EEC to operate normally and feed adequate signal to the speedo
http://reviseomatic.org/help/s-op-amp-advanced/Op%20Amp%20Voltage%20Follower.php
This requires a little electronics knowledge, but OP-AMPs are fairly cheap and simple devices.

The Autometer speedometer has the circuitry described above built in. It can take a sine wave input and it creates a secondary square wave output that can be used for aftermarket cruise contracl and "some efi computers", according to the instructions.

I am thinking that the most elegant solution is to run the sine wave signal to the speedometer and then run the square wave to the EEC for the idle control circuit (which my truck has always needed to prevent stalls at stop signs).

All of the research I could do indicated that the pulses per revolution were not too critical for my EEC but I can't find a definitive answer about the +5V signal into the EEC. I'm really trying to understand if the +5V is too much for the computer.

I highly doubt 5V is too much for the computer. IIRC, most VR chips can convert signals in excess of 25-30V. Easiest way to tell would be to set your DMM to AC and measure the VSS signal at high speed. ( A good job for a passenger ;D) I bet you will see much higher than 5V

Again, I would guess the zero-crossing problem would be the biggest hangup. Some chips can bias the input signal so they can read square wave or sine, but I don't know EEC-IV that well.

Feeding the VSS into the speedo and then the square wave into the EEC is a straight forward solution. My issue with that is the EEC is probably much more reliable than the speedo. Like it or not, the OEMs are much more concerned with longevity of their parts than the aftermarket. If your speedo craps out, you will lose signal to the engine as well.

Either way, good luck!

 

[Viperwolf1]

Is the VSS signal to the computer even necessary? A lot of people don't use or need it.

 

[Past_Miner]

Is the VSS signal to the computer even necessary? A lot of people don't use or need it.

I hear this from a lot of people but both of my EFI Broncos have pretty significant stall problems without it. It's irritating enough that I don't like driving them without the VSS hooked up. The one with the ZF is managable with some throttle feathering but the automatic just plain drives terrible without the signal.

 

[Past_Miner]

I highly doubt 5V is too much for the computer. IIRC, most VR chips can convert signals in excess of 25-30V. Easiest way to tell would be to set your DMM to AC and measure the VSS signal at high speed. ( A good job for a passenger ;D) I bet you will see much higher than 5V

Thanks for the responses. I think I will try to measure the VSS output voltage this weekend. As long as I see something in the 5V or more range I'll just hook it up and try it. The worst that could happen would be that I would have to put up with the stalling issue until I got home and reversed the wiring.

 

[ransil]

I have looked at a vss on a scope before is not a perfect "sqaure wave" more like a bunch of pusles, I would guess the computer is looking for edges vs trying to find a "pulse with zero crossing".
An Op amp (aka buffer) could work to maintain signal amplitude , but i think this add to much complexity.

I would try using the signal only disconnect from the EFI computer, I have used them with & with out VSS and not sure I see a difference.

 

[DirtDonk]

Past_Miner's experience with engine stalling out is fairly common to these conversions. Luckily not everyone has the same experience, but enough have that I would think he'd need to keep the signal to the ECU to avoid it. Be nice if it was that simple though.

Have you tried this particular setup without the signal to the ECU PM? Same result as your other? Were they both manuals, or any of them auto transmissions? Does it happen when slowing down to a stop?
I think the manual trans has more susceptibility, but again, not all experience the issue.

Paul

 

[Past_Miner]

Past_Miner's experience with engine stalling out is fairly common to these conversions. Luckily not everyone has the same experience, but enough have that I would think he'd need to keep the signal to the ECU to avoid it. Be nice if it was that simple though.

Have you tried this particular setup without the signal to the ECU PM? Same result as your other? Were they both manuals, or any of them auto transmissions? Does it happen when slowing down to a stop?
I think the manual trans has more susceptibility, but again, not all experience the issue.

Paul

I have run both of these without the VSS signal and with. The stumble is actually worse for the auto, which is an E4OD. The manual has issues coming to stops without the signal but otherwise runs well. I suppose I could compensate by simply adjusting the idle up manually but I don't know why I would when the VSS is so easy to plug in. The E4OD has some stumble problems coming to stops but also has lousy shifting without the speed signal. This one is a SD conversion from a FSB. With the signal it runs just like it came off of the showroom floor.

The piggybacked sensors will be painless on the E4OD bacause it also has a NP 205 transfer case and there really aren't any clearance problems for two sensors. This Bronco is realy the one that would be the best to test the square wave signal on because it has obvious drivavbility problems without the signal. This one also happens to be completely disassembled for paint right now so I will do the testing on the manual and then decide if I just keep installing two sensors of if I use the square wave signal.

I will probably get around to this over the weekend. I will give an update when (or if) I have anything intelligent to report.