Convert 12v a switched ground (-) sensor signal to switched +12v signal. (Automotive)

[DHR]

Hi everyone,

I am dangerously under-qualified, so please be understanding if I'm way off base...

The Story:

I'm doing an automotive project where, among other things, I'm changing the transmission in a Toyota 4x4 from automatic to manual. Its part of an engine swap, so I'm building a wiring loom, and adding a circuit will be very easy.

The manual transmission I have used a rotating cable to tell the instrument cluster the speed of the vehicle, but the vehicle I'm putting it into uses an electronic signal generated from a Vehicle Speed
Sensor (VSS) mounted on the transmission. I found a VSS from another vehicle that physically fits that will convert the rotating mechanical output to an electrical signal. The VSS I have outputs a switched ground signal, but the vehicles instrument cluster needs a switched +12v signal. I want to design a circuit that will go between the cluster and sensor to convert the signal polarity. So to clarify, VSS I have is powered, (It has +12v and ground (-) wire going to it, as well as a signal output... so 3 wires.) The output it generates alternates between open and ground 4 times per rotation. I need to convert that signal to alternate between open and +12v 4 times per rotation instead.

The Circuit:

The circuit I have come up with uses an IFR9Z30 and a 10k resistor; I have drawn a very tiny circuit diagram and attached for a visual reference with the questionable portions in gray.

+12v goes to the source pin. (This can climb as high as 14.5v when the alternator is charging, for those who are not familiar with Automotive.) The output of the VSS goes to the gate. The 10k resistor goes between the source and gate to keep the MOSFET off when the gate signal is open. The drain pin goes to the instrument cluster as the new signal.

The Questions:

Does my circuit make sense, or is it totally wrong or missing something?

Is my component selection ok? I'm especially concerned about the 10k resistor, but would appreciate confirmation on my MOSFET selection as well.

I'm not sure how the cluster works, but there's a good chance that it's inductive; should I add a schottky diode between drain and +12v to be safe? (in gray on the diagram)

I also see some circuit diagrams have a resistor from drain to ground; do I need that? If so, what value resistor would be appropriate? (I have lots of 10k resistors... will that work?) (also in gray on the diagram)

Thank you for taking the time to read my long post, and I appreciate any feedback or suggestions anyone can provide.

 

[Alec_t]

Welcome to EP!
Do you have a datasheet for the IFR9Z30? Is it a P-channel MOSFET?

 

[Harald Kapp]

IRF9Z30 (?) datasheet. PMOS, 50 V, 11 A (18 A @ 25 °C).

 

[Bluejets]

The mosfet you link to already has a protection diode built in.........
Cannot follow the rest of your circuit...can you provide complete diagram of what you intend to do.?

 

[DHR]

Welcome to EP!
Do you have a datasheet for the

Welcome to EP!
Do you have a datasheet for the IFR9Z30? Is it a P-channel MOSFET?

? Is it a P-channel MOSFET?

Oh Jeez, how silly of me. I really should have provided this in the first place... If I had, I would have realized that I had the description wrong. Its is not IFR9Z30, it is IRF9Z30. Please find the datasheet attached. Yes, it is P channel, as far as I can tell. I couldn't determine if it is enhancement or depletion, but in this case it actually doesn't matter, because the signal would have the same result either way.

Thanks for your reply, and sorry for the confusion.

 

[DHR]

IRF9Z30 (?) datasheet. PMOS, 50 V, 11 A (18 A @ 25 °C).

Yes, absolutely correct. Thank you, and sorry for the confusion.

 

[DHR]

The mosfet you link to already has a protection diode built in.........
Cannot follow the rest of your circuit...can you provide complete diagram of what you intend to do.?

Oh, well, that's good to know. I can eliminate that portion of the circuit then. Thank you!

The mosfet you link to already has a protection diode built in.........
Cannot follow the rest of your circuit...can you provide complete diagram of what you intend to do.?

I have updated my diagram to include a clear power source, chassis ground, the VSS, and mosfet symbol, I also removed the diode from the circuit and enlarged the image... I also corrected the most embarrassing spelling error; I hope this makes it clearer than before. Please let me know if it still needs work, and thank you for your help.

 

[Alec_t]

Looks workable now.

 

[Bluejets]

Nit picking but battery upside down.

 

[DHR]

Looks workable now.

That's excellent. Thank you.

So just to clarify, both resistors are required, and their values are ok? Forgive me for being uninformed, but I I dont understand what the resistor in gray is doing, and I don't know how to calculate what resistor value to use.

 

[DHR]

Nit picking but battery upside down.

LOL, not at all; that's ok. It helps me learn learn? Thank you.

 

[Bluejets]

That's excellent. Thank you.

So just to clarify, both resistors are required, and their values are ok? Forgive me for being uninformed, but I I dont understand what the resistor in gray is doing, and I don't know how to calculate what resistor value to use.

Gate resistor ensures mosfet stays off and the gate is not left floating.
Grey resistor is signal out pull down when mosfet is off.
Probably easier to see if it was drawn in a conventional manner.
Addition of 150R is also common current limit in case of failure, not essential to operation but good practice.

 

[DHR]

Oh, I see. That makes a lot of sense.

The 150R sounds like a good idea. If I'm understanding this correctly, it's a 150 ohm resistor that limits the current in the case of the MOSFET failing and allowing current to flow from the source to the gate?

 

[Bluejets]

No, it limits any gate fault current from damaging the input device.
Many instances this would be a microcontroller output pin which would normally be perhaps 5v or 3.3v.

 

[DHR]

Oh, I think I get it. I will add the 150 ohm resistor as you suggested. The VSS only costs about $20 but it doesn't make sense to not protect a $20 part to save $0.05 on a resistor... I have a feeling I might need to buy some 150 ohm resistors anyway; I just made a new thread about another circuit I'm working on that this might also relate to, as it uses two 5v microcontroller outputs.

Thanks again for all of your help; I really appreciate it; you've been a great help.

 

[Kiwi]

What about calibration of the speedo with the new gearbox?

 

[DHR]

What about calibration of the speedo with the new gearbox?

That's a really good question.

The honest answer is that I hope I don't have to. The the manual transmission did come from the same year make and model that it's going into, and VSS versions do exist. The FSM VSS test procedure says that the output should switch 4 times per rotation, which the VSS I have does... Except that it switches ground, not power.

If I do have to calibrate the speedo, it can be done by taking the cluster apart and adjusting a rheostat on the gauge mechanism... A pain but simple enough.

If the signal is too out of range, it is possible for me to change the output signal frequency by modifying the number of magnets inside the VSS. Right now there are 4, but I could change that to 2, 6 or 8 with minimal effort.

Funny storey about that. When I got the VSS, I wasn't sure if it would work, so I tested it based on the Toyota procedure to see if it gave me the signal I need. No matter what I did, I couldn't get a signal out of it and I couldn't find a test procedure for the vehicle the VSS came from. I noticed that the input rotor didn't feel right, so in the interest of "science", I decided to take it apart to see if I could confirm if it was defective, or just wasn't going to work for my application. Inside, it appeared that one of the magnets had come loose from the rotor and was interfering with the movement of the rotor. I tested it again, this time introducing a magnet to the pickup, and boom, I had a signal. I glued 4 small neodymium magnets to the rotor in alternating polarity and stuck the two halves back together using epoxy... it works perfectly.

 

[Bluejets]

Seems strange why the magnets would be alternatively placed given the output is digital.

 

[DHR]

Would that be considered a digital signal? I guess so... it is off or on... 0 or 1... but it's just a switching ground and there is no digital processing involved at either end of the signal, and it's a direct representation of the movement of the vehicle. More akin to a sound being played on a record, than from a CD, I would think... You probably know more about it than I do, so you could be right.

With less powerful magnets, two magnets might well produce the same result. Honestly, I didn't think much about it. I introduced one pole and the circuit was open, I introduced the other pole and the circuit was closed as long as the magnet was there. The magnets I have are quite powerful and would close the circuit from a solid half an inch away, which is further than the magnets can get from the pickup in the sensor. I think, in my case at least, the alternating poles is necessary to create separation to allow for the off state to occur. I did this all by trial and error; I only know that it works... Again, you probably know about this than me, but the sensor now outputs the signal I want in to, and the input rotates feely, so I'm happy with the result.

Now that I think about it, the arrangement of the magnets can change the duration of the output pulse... Right now the signal turns on for pretty much exactly 1/4 of a rotation and off for 1/4 of a rotation. If the cluster needs a signal thats on for say, 1/8 of a rotation and off for 3/8 of a rotation, or some other distribution, the reading might be affected. I don't think anything will be damage by this because it's possible for an always on signal to be sent to the cluster when the vehicle is not in motion, so probably the worst case scenario is that, I won't be able to adjust for accurate speed read out, and have to shell out for the more expensive off the shelf alternative. Based on the fact that indefinitely on or off signals still register as null movement on the speedo, I suspect that the frequency of the switching is more important than the duration of the pulses, so I still think it may work... Maybe you have some insight into that...