Volvo 480 Club Europe

Posted: **Fri Mar 11, 2016 3:27 pm**

Hello, I'm new to de 480 word and just bout a nice red 1989 turbo and well... it has been in the shop since

.

The speedometer doesn't work, and it is probably a bad speed sensor, bunt since it is very difficult to find a replacement I have been looking for another solution.
And this is what I found:
"Magnetic Pick up MSP6729 rotate speed sensor Generator parts"
Sorry I cannot put the url since I'm a new user but you can find it on eBay searching "MSP6729"

It is a magnetic sensor, electrically similar to the original one, and should be thin enough to fit the original hole, I do not know yet how to secure it, but i think it is doable, maybe creating a custom bracket.

Still, do you know of anyone else who tried a similar operation? Is there any concern against trying it? What could go wrong?

My only real concern is about the output voltage, that being dependent on the number of turns of the internal coil, could be higher than the original and consequently I could possibly overload the system somehow.

Have anyone measured the output of the original sensor at a given speed? I could test the "Chinese sensor" before connecting it to the speedometer using an oscilloscope, but I have no reference.

Thanks
and have nice Volvo day

Posted: **Fri Mar 11, 2016 3:32 pm**

Hello & welcome

this one? http://www.ebay.co.uk/itm/MSP6729-ROTAT ... SwuAVW1D1r

Posted: **Fri Mar 11, 2016 3:34 pm**

jamesy12345 wrote:Hello & welcome

this one?

Yes that is the model, other sellers also put some specifications in the description:

Specification:
MSP6729
3/8 - 24 UNF Threaded
2.5" Length with 6' Shielded Cable
All are designed and tested to operate within a temperature range of -65°F to 225°F (-55°C to 105°C) .
Magnetic Speed Sensors are with Wire Leads
Item Pickup Length Thread Length Lead Length Thread Size Drill Size
MSP 6729 2.7 2.25 72.00 3/8-24 UNF-2A .332
Introduction:
*The magnetic speed sensor indicates when ring gear teeth, or other ferrous projections, pass the tip of the sensor.
*Electrical impulses are induced within the coil and sent to the speed control unit. The signal from the magnetic speed Sensor, teeth per second(HZ), is directly proportional to engine speed.
Description:
*The magnetic speed sensor is mounted in the ring gear case or flywheel bell housing of the engine.
*The threaded hole for the speed sensor should be perpendicular to the centerline of the crankshaft and centered
over the ring gear teeth.
*A spot face should be present to provide a flat surface on which to anchor the locknut.
*With the engine stopped, screw the speed senor in until it touches a gear tooth, then back it out 3/4 of a turn and secure it with the locknut.
*Any ferrous gear may be used as long as the frequency and amplitude of the resulting signal meet the speed control unit specifications
*The wire leads should be twisted for their entire length from the magnetic speed sensor to the control unit.
*The leads may need shielding if they are longer than 10 ft. (3m), or if external interference is present.
*Do not connect either of the speed sensor leads to anything but the speed control unit used. The shield should not be connected at the speed sensor end.

Posted: **Sat Mar 12, 2016 1:05 pm**

I used a 'speed sensor' BUT bought it as a complete set , ie a sensor and speedo.

Unless you are 100% sure it runs on 12V and correct resisitance etc I'd steer clear . . . . .

The sensor fitted to the 480 has a square wave output, and I'm sure that the one listed does as well, it is just the level of square wave so ideally you want somebody to post the ACTUAL output from a normal unit and then compare on a 'scope.

i bought one of the palm size digi scope for £40 (DSO 201?) and saved me an MoT and lamda sensor etc

Posted: **Sat Mar 12, 2016 5:01 pm**

Alan 480 wrote:I used a 'speed sensor' BUT bought it as a complete set , ie a sensor and speedo.

Unless you are 100% sure it runs on 12V and correct resisitance etc I'd steer clear . . . . .

The sensor fitted to the 480 has a square wave output, and I'm sure that the one listed does as well, it is just the level of square wave so ideally you want somebody to post the ACTUAL output from a normal unit and then compare on a 'scope.

i bought one of the palm size digi scope for £40 (DSO 201?) and saved me an MoT and lamda sensor etc

Hi and thanks for the answer.

Actually the speed sensor on my 480, witch doesn't have abs, is "passive" (I presume) and only has two wire, while active ones usually have three (signal, ground, power).
Likewise the one listed has two wire for signal (plus the shield) and doesn't require power.

This kind of sensors outputs a sine wave directly generated by the movement of the gear teeth interacting with the internals magnets and coil, instead active ones do generate an amplified square wave using more complex integrated electronics.

I found some great images on google explaining the difference between the two systems but again, being a new user, I'm unable to post them.

Posted: **Sat Mar 12, 2016 5:55 pm**

Gmac34 wrote:
Alan 480 wrote:I used a 'speed sensor' BUT bought it as a complete set , ie a sensor and speedo.

Unless you are 100% sure it runs on 12V and correct resisitance etc I'd steer clear . . . . .

The sensor fitted to the 480 has a square

Actually the speed sensor on my 480, witch doesn't have abs, is "passive" (I presume) and only has two wire, while active ones usually have three (signal, ground, power).
Likewise the one listed has two wire for signal (plus the shield) and doesn't require power.

This kind of sensors outputs a sine wave directly generated by the movement of the gear teeth interacting with the internals magnets and coil, instead active ones do generate an amplified square wave using more complex integrated electronics.

I found some great images on google explaining the difference between the two systems but again, being a new user, I'm unable to post them.

Fair point about square vs sine ; but I'd still be canny, the speedo input is different input from the engine sensor and pretty sure they are BOTH just two wires

have asearch on here as I rememeber 'doing it sideways' (i think) looking at various options, and maybe he found the 'alternative' fitting

http://www.volvo-480-europe.org/forum/v ... or#p207477

Posted: **Sat Mar 12, 2016 6:13 pm**

Alan 480 wrote:
Gmac34 wrote:
Alan 480 wrote:I used a 'speed sensor' BUT bought it as a complete set , ie a sensor and speedo.

Unless you are 100% sure it runs on 12V and correct resisitance etc I'd steer clear . . . . .

The sensor fitted to the 480 has a square

Actually the speed sensor on my 480, witch doesn't have abs, is "passive" (I presume) and only has two wire, while active ones usually have three (signal, ground, power).
Likewise the one listed has two wire for signal (plus the shield) and doesn't require power.

This kind of sensors outputs a sine wave directly generated by the movement of the gear teeth interacting with the internals magnets and coil, instead active ones do generate an amplified square wave using more complex integrated electronics.

I found some great images on google explaining the difference between the two systems but again, being a new user, I'm unable to post them.

Fair point about square vs sine ; but I'd still be canny, the speedo input is different input from the engine sensor and pretty sure they are BOTH just two wires

Have you measured the engine rpm sensor? If it has two wires it could be passive as well, unless the metal mounting bracket is used to connect ground (unlikely).

I was thinking about connecting to the car (were the sensor would be connected) a signal generator (formerly my iPhone) to send sine wave to the speedometer and simulate the sensor, I would be able then to increase step by step the voltage and the frequency until I see some sign of life (if any).
This would give me a rough idea of the operating voltage of the system.

I suspect I'll end up frying something.....

Posted: **Mon Mar 14, 2016 9:41 am**

Alan 480 wrote:I used a 'speed sensor' BUT bought it as a complete set , ie a sensor and speedo.

Unless you are 100% sure it runs on 12V and correct resisitance etc I'd steer clear . . . . .

The sensor fitted to the 480 has a square wave output, and I'm sure that the one listed does as well, it is just the level of square wave so ideally you want somebody to post the ACTUAL output from a normal unit and then compare on a 'scope.

i bought one of the palm size digi scope for £40 (DSO 201?) and saved me an MoT and lamda sensor etc

Funny that I have testing speedos with a sine wave, because the switch that selects square wave on my prehistorical valve-powered frequency generator is broken

Usually the sensor works, or not at all. It is basically a coil that is interrupted and the interruption causes a spike in voltage: u(L) = L*di(L)/dt. The electronics in the speedo don't care much about how the signal looks like, it probably amplifies the input up to the point it is clipping anyway. Almost any problem in the speedo circuit stems from bad connectors, either near the speedo sensor, or near the speedo itself. Cleaning all connectors and proper re-soldering of the joints on the reverse side of the speedo circuit board does the trick in most cases.

Sometimes people think that a new sensor solved the issue, because the disconnecting and reconnecting of the sensor more or less cleaned the contacts and made the system work again.

You can check the ohmic resistance of the sensor, it should be around 3 ohms, but cheap digital multimeters are notoriously inaccurate in this range, as they cannot zero out the measuring probe resistance properly. However, if you measure significantly higher resistance, there is a problem in the sensor (if you measure directly on it) or the wiring (if you measure on the speedo connector). Also, check if there is proper earth. The speedo cable itself is shielded for interference protection, there are many points in which impedance loading problems may occur along the transmission line, but the frequencies involved are so low (around 400 Hz if you are doing top speed, superimposed on a 220 kHz carrier) that I am not expecting trouble with that.

Posted: **Mon Mar 14, 2016 10:10 am**

Gmac34 wrote:Have you measured the engine rpm sensor? If it has two wires it could be passive as well, unless the metal mounting bracket is used to connect ground (unlikely).

I was thinking about connecting to the car (were the sensor would be connected) a signal generator (formerly my iPhone) to send sine wave to the speedometer and simulate the sensor, I would be able then to increase step by step the voltage and the frequency until I see some sign of life (if any).
This would give me a rough idea of the operating voltage of the system.

I suspect I'll end up frying something.....

The sensors are not containing any active electronics. It's just a pickup coil. So they don't generate anything, they're just part of a circuit. I have read somewhere that the speedo itself sends out a 220 kHz signal. That will be interrupted by the coil, because the teeth of the gearbox move past and thus change the inductance of the coil. The 220 kHz signal is fed to a low pass filter which will leave the actual speedo information (0 - 400 Hz).

So it is not a question of sine or square wave, or voltage (well, you shouldn't plug it into the mains

), it's a question whether the combination of pickup coil and gearbox work with the speedo electronics. You cannot break any part electronically, only mechanically, i.e. if you screw in the sensor too far, it won't survive and probably will damage the gearbox as well.

(using an old smartphone for a signal generator: nice!

I should try and revive my old Samsung Galaxy S-II and install this: http://www.keuwl.com/FunctionGenerator/ )

Posted: **Mon Mar 14, 2016 1:07 pm**

brinkie wrote:
Gmac34 wrote:Have you measured the engine rpm sensor? If it has two wires it could be passive as well, unless the metal mounting bracket is used to connect ground (unlikely).

I was thinking about connecting to the car (were the sensor would be connected) a signal generator (formerly my iPhone) to send sine wave to the speedometer and simulate the sensor, I would be able then to increase step by step the voltage and the frequency until I see some sign of life (if any).
This would give me a rough idea of the operating voltage of the system.

I suspect I'll end up frying something.....
The sensors are not containing any active electronics. It's just a pickup coil. So they don't generate anything, they're just part of a circuit. I have read somewhere that the speedo itself sends out a 220 kHz signal. That will be interrupted by the coil, because the teeth of the gearbox move past and thus change the inductance of the coil. The 220 kHz signal is fed to a low pass filter which will leave the actual speedo information (0 - 400 Hz).

So it is not a question of sine or square wave, or voltage (well, you shouldn't plug it into the mains ), it's a question whether the combination of pickup coil and gearbox work with the speedo electronics. You cannot break any part electronically, only mechanically, i.e. if you screw in the sensor too far, it won't survive and probably will damage the gearbox as well.

(using an old smartphone for a signal generator: nice! I should try and revive my old Samsung Galaxy S-II and install this: )

Hello, and thanks for the comprehensive answer.
still a few elements are unclear:

a magnetic sensor (as opposed to an hall-effect sensors) is indeed a small ac generator so it does in itself generate a variable ac current, variable in both amplitude and frequency, both relative to the speed at which the related gear is rotating, therefor I should be able not only to measure the resistance of the coil but also to see the wave with an oscilloscope connected directly to the sensor.
I'm sill unable to post URLs so please search" Magnetic speed sensor instructables" to see a diy example connected to an oscilloscope in the last page.

I do not really understand how the 220 kHz signal is used, if you can find the source about that I would really like to understand more, it seems like It could be the frequency at which the electronic counter is working but I don't understand why it would be passed thought the sensor.

I had to postpone my session of tests since I'm now in a different city than the car, so I'll have time to gather more information before doing anything.
Thanks again
Giovanni

Posted: **Mon Mar 14, 2016 2:26 pm**

What I think - but lacking a speedo sensor (one was promised to me but went to the scrapyard along with the car where it was sitting in

) I have never been able to confirm this theory - that together with the pickup coil, a resonant circuit, i.e. an oscillator is formed. The oscillating frequency depends on the coil's self inductance. Every time a tooth of the gearbox passes, the oscillator stops. But that is information I got from someone who claimed to have experimented with this and measured a 220 kHz carrier.

The speedometer of the Volvo 480 involves an ITT SAF1091/SAF1092 chip set. My assumptions are based on this diagram:

The sensor coil (down left corner of the schematic) is connected in parallel with a 10nF capacitor. Hence, I am pretty sure this circuit doesn't pick up the bare pulses generated by the sensor, but instead functions as an oscillator which will be interfered by the passing teeth of the gearbox. Inside the SAF1091 (which is a bipolar IC, the SAF1092 is a CMOS digital divider) there is probably a low pass filter which irons out the 220kHz pulse and leaves the actual signal.
I could try and connect a 47µH (closest E12-value of the calculated 52.3µH according to f=1/sqrt(2*pi*L*C) where f=220kHz and C=10nF) coil across the input, and measure with an oscilloscope what happens, to prove my theory.

You can safely assume very little current will flow through the sensor.

This is probably the link you were referring to: http://www.instructables.com/id/Magnetic-speed-sensor/

When a peice [sic!] of ferrous metal is moved towards the end of the sensor it changes the shape of the magnetic field in the coil, this changing magnetic field then induces current to flow in the windings of the coil resulting in a small amount of electricity being generated.

Still, this doesn't involve active components. There is no external power supply, the sensor is energized by the passing teeth of the gearbox (i.e. slowing down the gearbox a wee tiny bit)

Posted: **Mon Mar 14, 2016 2:59 pm**

Interesting, it works quite differently than I thought.

this actually raises further questions:

- Is the capacitor part of the sensor or is it on the speedometer board? Looking at the schematics it seems to be on the board but who knows what volvo actually did, otherwise I would have to add the capacitor to the Chinese sensor.

-does the original sensor include a magnet? I guess I can check this by using a ferrous nail and feel if there is any attraction between the two.
If the originals sensor doesn't have a magnet in it, it is'n generating a voltage and connecting the new one could potentially mess the ic.

Also, when you said that you tested speedometers using a signal generator, how did you connect it? At this point I do not see how a signal generator could replace the coil.

P.s. I found right away the data sheet of this part and it describes the inner working exactly as you did, so there is no need for a test

Posted: **Mon Mar 14, 2016 3:11 pm**

Gmac34 wrote:- Is the capacitor part of the sensor or is it on the speedometer board? Looking at the schematics it seems to be on the board but who knows what volvo actually did, otherwise I would have to add the capacitor to the Chinese sensor.

-does the original sensor include a magnet? I guess I can check this by using a ferrous nail and feel if there is any attraction between the two.
If the originals sensor doesn't have a magnet in it, it is'n generating a voltage and connecting the new one could potentially mess the ic.

Also, when you said that you tested speedometers using a signal generator, how did you connect it? At this point I do not see how a signal generator could replace the coil.

The capacitor is on the board. Actually the values on the board are pretty much the same as in the ITT application note. I don't know how the internal components of the sensor exactly look like, only that they are low impedant and it is some sort of coil, of which I assume it has a value around 50µH.
About the testing, I boldly connected a signal generator to the input terminal of the sensor on the speedometer and it worked.

The carrier frequency of 220kHz is completely lacking, but since that signal is filtered anyway, it isn't important.

Posted: **Mon Mar 14, 2016 3:31 pm**

Gmac34 wrote:Hello, I'm new to de 480 word and just bout a nice red 1989 turbo and well... it has been in the shop since .

The speedometer doesn't work, and it is probably a bad speed sensor, bunt since it is very difficult to find a replacement I have been looking for another solution.

Before trying anything else, I strongly advice to properly solder the joints of the green and grey connectors again and do a proper cleaning of the contacts of all connectors involving the speedometer. I don't know where you are located; I have a spare MY 1990 unit that is known to work. The sensor itself is rarely at fault, and is easy to measure (high resistance or even open circuit).

Posted: **Tue Mar 15, 2016 2:13 pm**

Ok, I'll do that,
I still think of testing with the signal generator first, as it is easy and fast to connect it.

I don't know when I'll be able to experiment, probably in a couple weeks, I'll keep you posted.

Thanks for all the infos.
Giovanni

Posted: **Wed Mar 16, 2016 1:39 pm**

I tried an Android mobile phone powered frequency generator yesterday, but it didn't work. Either the phone didn't generate enough power or the headphone connector I used was broken. I might try an iPad today, they have a more powerful audio amplifier.

Posted: **Thu Apr 07, 2016 8:40 pm**

Hello, it's been a while, but I finally got time to do the experiment.
First I tried to use my iPad as a signal generator hooked directly to the sensor socket and the needle moved (I used an headphone amplifier in order to have more voltage, I don't know if the iPad alone would have been sufficient).
So confident that the problem was probably a bad sensor I tried to hook up the Chinese sensor and it worked straight away.
Whe had to machine an adaptor that screws on the Chinese sensor in order to properly mount it and seal the original hole.
I'll have to do a proper test on the road and check if the speed is precise, now I can't since there are no front seats

Otherwise, could you write me a couple of gear number/ rpm/ speed combinations so I can check it without leaving the shop, I'm rellay curious.
Thanks
Giovanni

Posted: **Thu Apr 07, 2016 9:08 pm**

Maybe will help, assuming your car is a manual

Gear ratios (from here http://www.volvo-480-europe.org/volvo48 ... engine9=on)
1. 3.091:1
2. 1.842:1
3. 1.320:1
4. 0.967:1
5. 0.758:1
reverse. 3.545:1
final drive. 4.067:1

Speed calculator (need tyre sizes)
http://www.tremec.com/calculadora.php

Posted: **Thu Apr 07, 2016 10:08 pm**

3000rpm on 5th is quite accurately 120km/h on my turbo

Posted: **Fri Apr 08, 2016 8:19 am**

Giovanni you have a pm

Volvo 480 Club Europe

Chinese Speed sensor, any advice?

Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?

Re: Chinese Speed sensor, any advice?