First of all, I'll be honest. Most everything here is WELL beyond my skill set. I'm a welder/fabricator. So I'm looking for some help from you guys.

I'm working on an old Honda that has a new(er) speedometer transplant. When we first did the speedometer swap, we weren't concerned with odometer mileage accuracy. Now we've decided we would like to make the odometer reading accurate to the original. This means we need to run up approximately 40k miles.

Dissecting the speedometer/odometer to roll the miles forward is NOT an option due to the way it's mounted and the irreplaceability of the gauge housing itself. We don't want to risk it whatsoever.

Me being mechanically inclined decided we'll do it with a drill or a small motor hooked to the speed sensor. This works, but isn't ideal, and we can't safely spin the VSS for extended periods of time without constant supervision.

I would like to run up 40k miles electronically by bypassing the VSS all together and tricking the speedometer into thinking it's getting a signal and it's running 100-130mph. I need a way to supply a signal to the speedo for a few weeks straight.

Some info on the VSS based on my testing and research:

-output signal from VSS to speedometer is square wave (no clue what this means! But I'm sure you do)
-output VOLTAGE from VSS to speedometer ranges from 0-5v. 5v when the speed is zero, and somewhere closer to 0volts as speed increases. (Yes, 5v at zero and decreases with speed)
-Honda uses a 4k pulse per mile

So knowing the above, I figured I need to supply somewhere between a 111hz and 144hz signal to the speedometer (1/4,000*111*3,600 seconds per hour)

111hz = 100 mph
133hz = 120mph
144hz = 130mph

I don't want to spin the speedo faster than 130mph due to mechanical limitation. 130mph will get me to my target mileage in approximately 12-13 days.

I will be supplying the speedometer unit its own 12v power source, so no problem there. I just have to figure out how to emulate the VSS signal.

I know there's a way to do this, but I'm not exactly sure how. I need your help.

Will this do what I need? Safely - for up to 2 weeks straight? Looks like it'll generate in the frequency range I need, but I'm not sure about the voltage. How does the Honda VSS's output voltage of 0-5v impact the situation? Or does it even matter as long as the frequency is correct?

http://www.amazon.com/RioRand-Generator-Adjustable-Frequency-Module/dp/B00LHBZCCC


Thanks for any insight you guys can provide. Keep in mind I can't make circuits and things like you guys can, that stuff is way way way over my head, so I would need to be able to buy something or two to accomplish this! And preferably cheap!

I really appreciate you taking the time, and I look forward to your advice.

 

The simplest way to get a 120 HZ pulse train would be to take most any common older type of power pack that has a transformer in it and simply feed its rectified signal to the speedo input via a potentiometer so that the output voltage can be varied.

Without any smoothing capacitor the output will be a steady 120 hz and the potentiometer will allow you to fine tune the signal voltage to where the VSS system is happy with it.

 

Can you post the speedometer docs you have? Don't see why voltage is used to indicate speed. Frequency should be sufficient; 4k pulses in an hour = 1MPH.

If a higher frequency is tolerated, you could advance the miles faster.

I Google for info and it seems that the ECU also stores mileage and having different values could cause legal issues if speedo reads too low.

The pulse generator you referenced operates from 5-12V so it won't work if you really need pulses near zero volts.

 

I suspect the 'near zero volts' is an average resulting from the sluggish response of the meter used. Could the VSS be putting out fixed width negative pulses (e.g. from an open-collector output stage) at a speed-proportional frequency?

 

My experience with vehicle speed sensors is most are reluctance based which without the correct type of meter will show incorrect output voltages being as the frequency goes up most will start to show a bit of DC biasing which screws up most meters accuracy on doing AC readings.

The only way I consider to be a reasonable way to check there signalis with a O-scope and visually look at the signal.

Odds are with the system you have most any signal with a good deal of voltage ripple to it would work just fine.

 

I suspect the 'near zero volts' is an average resulting from the sluggish response of the meter used. Could the VSS be putting out fixed width negative pulses (e.g. from an open-collector output stage) at a speed-proportional frequency?

I think the same way... a scope reading would be in order, perhaps you could ask an electronics-savvy friend to do it for you

 

Thanks for all the feedback guys.

Most of what you're saying is over my head. I have no electronic savvy friends! That's why I'm relying on you folks.

The factory service manual for the car the vehicle speed sensor came from states that to test the VSS, you check DC output voltage with a DMM and should see between 0-5volts while spinning the front wheels by hand. I assume this is so most average mechanics can easily check it with average tools.

At rest, the DC output of the VSS is around 5volts measuring with my Fluke DMM. At 80mph (the fastest I could get the VSS to spin with a drill, output was around 2.5volts).

I was thinking too stupidly and used these findings as an aha moment. I was thinking I could just supply somewhere between 0-5 volts to the speedo to make it run. Boy did I feel stupid when it didn't work!

You guys correct me if I'm wrong, but I suspect it's not the "voltage" that's important. The voltage could be anywhere in the 12v DC operating range. The voltage is just a byproduct of the pulse/frequency and is just an easy way for a novice mechanic to test that the VSS is doing something, anything.

In other words, I'm not sure the speedo needs to necessarily see any "voltage" persay. It just so happens that the VSS will output 0-5volts when rudimentarily testing it - but more precisely, the speedo is looking for a certain pulse/frequency. I suspect the factory service manual has owners test for voltage rather than pulse/frequency because the latter is much more difficult for the average mechanic. In my case, I would need to see somewhere between 100 and 150hz to get my desired speed.

Do any of you see any reason why the item in my link won't work, assuming the DC operating voltage range is incorrect? I could supply somehow in the 1-5volt range to the device, which I assume would keep the output voltage the same but have the "correct" frequency.

How does voltage relate to frequency? Or do they operate irrespective of eachother?

If I'm too ignorant to even be making sense I apologize! I'm really quite out of my league with this stuff.

 

Thanks for all the feedback guys.

Most of what you're saying is over my head. I have no electronic savvy friends! That's why I'm relying on you folks.

The factory service manual for the car the vehicle speed sensor came from states that to test the VSS, you check DC output voltage with a DMM and should see between 0-5volts while spinning the front wheels by hand. I assume this is so most average mechanics can easily check it with average tools.

At rest, the DC output of the VSS is around 5volts measuring with my Fluke DMM. At 80mph (the fastest I could get the VSS to spin with a drill, output was around 2.5volts).

I was thinking too stupidly and used these findings as an aha moment. I was thinking I could just supply somewhere between 0-5 volts to the speedo to make it run. Boy did I feel stupid when it didn't work!

You guys correct me if I'm wrong, but I suspect it's not the "voltage" that's important. The voltage could be anywhere in the 12v DC operating range. The voltage is just a byproduct of the pulse/frequency and is just an easy way for a novice mechanic to test that the VSS is doing something, anything.

In other words, I'm not sure the speedo needs to necessarily see any "voltage" persay. It just so happens that the VSS will output 0-5volts when rudimentarily testing it - but more precisely, the speedo is looking for a certain pulse/frequency. I suspect the factory service manual has owners test for voltage rather than pulse/frequency because the latter is much more difficult for the average mechanic. In my case, I would need to see somewhere between 100 and 150hz to get my desired speed.

Do any of you see any reason why the item in my link won't work, assuming the DC operating voltage range is incorrect? I could supply somehow in the 1-5volt range to the device, which I assume would keep the output voltage the same but have the "correct" frequency.

How does voltage relate to frequency? Or do they operate irrespective of eachother?

If I'm too ignorant to even be making sense I apologize! I'm really quite out of my league with this stuff.

Don't worry, we'll try to help you one step at a time, if wecan.
Can you get your hands on a Fluke multimeter with a digital display capable of showing a waveform? This could be used in place of an oscilloscope

 

So basically I am just talking to myself here.

 

I didn't understand the first part of your first response (over my head). I'm a dummy in a smart mans game!

I see at the end probably any signal will work - do you think the device in my link will provide adequate signal?

It's cheap enough I may try it.

I cannot get my hands on a Fluke with the capabilities the other poster inquired about . I actually have the fanciest DMM in our tech group, which is sad.

 

What I am talking about is simply taking common 60 cycle alternating current and running it through a full wave bridge rectifier to get a DC output that has a high level of 120 HZ ripple to it.

For the most part to the VSS systems input that will look close enough to a typical speed sensors output to make it work.

 

Thanks for the replies everyone. I would like to update this thread for future searchability. I hate when I stumble on a relevant older thread with no conclusion.

I ended up ordering two more of the 555 pulse generators. I soldered some wiring extensions to the leads to make it less cumbersome for hookup and lower the chance of my accidentally touching something wrong which is what blew up my first NE555 pulse generator. I have it running off a plain old 12v car battery but I have the voltage reduced to 3.5v as the speedometer per the FSM only should be seeing 0-5v so I would rather be safe than sorry. It probably would have been okay to not reduce the voltage but I already had the voltage reducer so I figured why not.

I have a car battery feeding an LM2596 voltage reducer ($5-6 on Amazon) that I had leftover from another project to reduce voltage to 3.5 which then powers the NE555 pulse generator which then feeds to proper frequency to the speedometer signal input.

So far this seems to have done the trick.I thought it had a needle limitation on the speedometer but on closer review I was actually able to get it almost clock all
The way around before touching the 0mph resting needle (don't know the exact term). Doing some math tells me that I've got the speedometer turning at 192mph. It's been running unattended for a few days now and has already racked up over 10k miles.

I should be at my target mileage within a week. I'll update when it's done.

 

Hello. I am having the same issue with my vehicle. I will like to run up the miles by 4k. I am hoping you can help me out? Your solution seems to be exactly what I need to do. I would like a bit more info.

MOD NOTE: Phone number removed.

Please don't post personal contact information; it attracts spammers both to you and to AAC.

 

Well, I might not have an answer to your question, but here's how I came to know what I do know about speedometers: My Toyota Tacoma, when traveling at an indicated speed of 65 MPH my actual speed was 57 MPH. At first I brought the truck back to the dealership but they said there was no problem because wheel speed as indicated by anti-lock brakes was consistent with the speedometer reading. I knew something was wrong, so I set cruise control and drove slower than most traffic for 10 miles (per mile markers) and timed the trip; then did the math. Got clear indication I was going much slower than what the speedo was telling me. Armed with that info the dealership still insisted there was no problem. I know they allow 2% error anywhere from accurate to 2% OVER on the speedo, meaning you could be going slower than indicated or as indicated. They NEVER set up a car to indicate it's going slow when it's going faster. They don't want liability.

So I considered tire size, rear end gear ratio and engine RPM at 60 MPH. That also gave me a clear indication I was not going as fast as indicated. So I started asking about the speed sensor mounted in the transmission output yolk. The dealership agreed to change the one I had. I put a tool mark on the original just to see if they'd try and snow me. They actually did replace the sensor. But the results were the same. So I grabbed my oscilloscope and set it up on the hood and connected it to the speedometer input lead in the dash. Got a nice 5 volt square wave, between 0 and 5 volts. No, not a "Range" of voltage between zero and five, it was either zero OR five volts.

The sensor simply senses the presence of a magnet (on my vehicle). As the magnet moves past a hall sensor it switches the sensor ON and OFF. The faster it switches on and off is translated as greater speed. The tire size on my truck accounted for 3% of the inaccuracy, but not all of the inaccuracy. Also, I'd drive 100 miles and my odometer would read out 109 miles. So just guessing here, but that sounds like a 9% error. In the end my solution was simply to change the gear in the sensor from 29 teeth to 30 teeth. That one extra tooth corrected that 9% beautifully. But the speedo was still not accurate. I eventually had to set my speedo to -2 MPH when parked. At 25 indicated MPH my actual speed was 26. At 50 it was spot on, and at 75 indicated MPH my actual speed was 74. I could live with that.

From this long long winded story You need to understand that your speedometer is seeing a 5 volt square wave who's frequency varies by speed. This can be simulated electronically, with no need for any mechanical work whatsoever. I'd think you can simply take a coil, probably out of a relay or something else you can scrap out, remove the speed sensor and then just place the coil (distance would have to be experimented with) and have an electronic package such as a 555 timer / transistor circuit to drive the relay coil at a given frequency. If you know that you get two pulses per revolution (that's what I think I recall my Tacoma was) then you can determine (mathematically) the frequency and time required to advance your speedometer to near your desired circuit.

Since I went through all this, another thing I went through was learning that there are shops that will flash your odometer for you. Just bring them both gauge clusters and tell them which is the accurate and which needs to be flashed. It's a fairly easy process, but it's also a closely guarded secret. There's a memory chip they remove, reflash and then reinstall. It won't be free or cheap, so you might want to call around and find some places that do speedo work.

Long, but I hope helpful.

 

I should be at my target mileage within a week. I'll update when it's done.

This is an example of why I should read ALL posts before posting my own comments. Sorry, I got a little impatient and thought I could throw my two cents into the pot not knowing that the hand was already over.

Glad to hear your solution is working. Sounds like what I would have done to a large degree. Only, I would have gone with 5 volts. But it appears that 3.5 volts is working for you, so - - - . ")

 

I appreciate your response. I have contacted several people who claim that they can correct the mileage on my odometer. They would have to remove the tiny 8 pin chip. Reprogram and re-solder back in place. Although this is a option, it is pricey but more importantly risky. I fear that the chip pins can end up broken on removal. The chip looks VERY sensitive. I bumped into this thread via google. It has exactly the same scenario as what I am going through and what I want to accomplish. I simply need to run my miles up by 4,000 miles. I have read that my vehicle uses a 5v square wave as well. I purchased a Ne555 yesterday from amazon. It will arrive tomorrow. Description mentions that it is adjustable 5v-15v or so. Here is what I plan to do and please tell me if something doesnt seem correct. As I wouldn't want to damage my vehicles electronics in the process. So i plan to give the Ne555 12v power and ground via a battery. The signal out on the Ne555, i will connect to my vss signal wire. I will use the required frequency to get near max MPH output reading on the odometer. I am guessing that i place the jumper on the frequency pins that are near my required frequency and then i can raise or lower the frequency via the adjustment knob on the ne555? Can adjutsment be done while everything is fully connected and visible mileage increase are present on the odometer? Can the voltage be adjusted on the Ne555 via adjutable knob? I am guessing so but I am new to all this. Any help will be greatly appreciated.

 

I fear you might have a misconception or two about how the 555 works. Yes, you CAN input the pulses directly into the VSS (Variable Speed Sensor) but as I suggested in my first post, use an electro-magnet to trick the sensor into thinking the vehicle is traveling at a given rate.

Yes, you CAN change the frequency, but you need to understand exactly how to go about adjusting the frequency. Rather than hijack this thread, why not post your own NEW thread? That way you'll get a lot more exposure.

 

As for the fragility of the leads, having worked in electronics for over 30 years, yeah, they can be a bit fragile. But a lot of people know how to remove and replace those 8 pin chips. And yes, I knew it was a small chip. The good news is that if you pay someone to re-flash your odometer and they damage something - THEY are responsible for repairs or replacement. But yes, it does cost a bit more than building your own circuit.

 

My intentions was NOT to "hijack" anyones post. I commented in hopes of help from the op. Since he went through the same thing. So who better to help me than the op? The op didnt mention having to use a elctro magnet. I know its your recommendation but is it really needed?

 

This is the ne555 that i purchased. I am hoping it can get the job done