PCB potentiometer in a throttle body

[A320mech]

I took apart the electric throttle bodies of my 32 year old BMW and found this built-in potentiometer. It is a circular PCB with resistance traces adjacent to a good old DC brushed motor. The wipers thereof are connected to the butterfly of the throttle and the resistance is proportional to the throttle position and includes a "switch" (or cutoff, seen on the trace as a gap) which goes open at 17 degrees as an independent verification of the throttle position to the ECU. The traces are known to wear out and these units are NLA. I know there are many online services that can print out PCBs but I've yet to find someone who makes this.

Is there any company who can print resistance traces for a potentiometer on a PCB or am I going about it the wrong way here?

[nctnico]

There are services which can print carbon tracks on PCBs for use with silicone buttons but these will likely not have a well defined resistance or surface smoothness.

Random Google find:
https://www.raypcb.com/carbon-ink-pcb/

[EPAIII]

Perhaps a functional equivalent could be made with a number of surface mount resistors and a number of contact pads instead of the carbon directly on the PCB.

As to who can coat a PCB with carbon traces, I would say any company that makes potentiometers. But I doubt that they would be interested in making a one-off. At least at a price that you could afford.

One must wonder if a vehicle from the year 2000 or earlier would stay out of the scrap pile longer than one from 2020. I suspect the answer is yes. I also suspect that a more popular vehicle from any model year would have parts available for longer than one that sold in smaller quantities. Ford, Chevy, GMC, etc. vs. that BMW. That's just the nature of the beast.

[A320mech]

It's a 1992 BMW 850i with a V12 engine and yes it has become somewhat of a rarity. I knew what I was getting myself into when I got it which is why I'm looking for alternatives well before anything fails.

The SMD Idea won't work because the ECU will start fluttering the throttle as it tries to hit a resistance value in-between the SMDs. But thanks for the ideas.

[Nominal Animal]

Would it be possible to 3D print a new enclosure and gearing for an external potentiometer better suited for the job?

Perhaps the mating part could even house a physical potentiometer of suitable dimensions?

Consider, for example, a multi-turn potentiometer that is connected to the current wiper axis with some gears, but only a fraction of the turns is used for the full range (depending on the gearing ratio, of course).

Is it a pure potentiometer, or rather an adjustable voltage divisor, depending on the throttle position?

Apologies if you've already thought of these; me just a hobbyist.

[Someone]

The SMD Idea won't work because the ECU will start fluttering the throttle as it tries to hit a resistance value in-between the SMDs. But thanks for the ideas.

Performance* upgrade! Send us your perfectly normal and working unit to get a specially modified added performance* unit shipped back to you.

* as in show/act

[forrestc]

In the category of probably won't work but interesting to think about:

If I had that in front of me and someone said "hey, you need to replicate the functionality of this using modern technology but make it mechanically compatible", I'd start by learning more about the details of using capacitive sensing technology.  This is a fancy way of saying to use the same technology small microcontrollers use to read pushbuttons and sliders. 

The oversimplified way that these work is by you designing triangles or other similar structures on a PCB that connect to a microcontroller with appropriate capacitive sensing peripherals.  The "slider" can be pretty much any metal which will affect the capacitive sensing.   These can be read by a small microcontroller. Once you have the position you can then either use a DAC or digital pot to output the value.

You can get pretty good resolution doing this but I don't know if this would be fast enough or have enough steps.  Somehow I doubt it, but I thought it was an interesting idea even though it probably won't work and/or would be too complex. 

[nctnico]

It's a 1992 BMW 850i with a V12 engine and yes it has become somewhat of a rarity. I knew what I was getting myself into when I got it which is why I'm looking for alternatives well before anything fails.

I guess you are already in contact with other owners to find alternatives? It could be the same PCB is used in different throttle bodies. You can't be the only one having this problem.

I did find this for example:
https://v12throttle.com/products/e31-bmw-v12-throttle-encoder-plate-replacement-service-pair-850ci-850i-850csi-750il

At the bottom there are links to other models as well. No idea what would fit your car.

[squadchannel]

It is very difficult to replace this with a pot/resistive rotation position sensor or similar.
because, in most cases, the TPS (throttle position sensor) is non-linear.
The linearity is different between about 0-10%, which corresponds to idling, and 10%-100%.

You can refer to the image search or the resistance values listed in the TPS test method in the service manual.
It is easier to find compatible ones from there.
Are there any used ones out there? It is quite possible that a similar throttle body is used in a similar year or other model.

Custom made? just one? It's going to cost you hundreds of dollars, but you want it?

[tom66]

Can you not install a more modern conventional throttle position sensor?  Potentiometers are still used today in cars but the common solution is that there are two potentiometers in opposition to each other, which allows the ECU to sanity check the reading.  The problem is finding a way to replace that '17 degree sensor' - a switch on a bracket might work but as this is the only thing providing you with a throttle shutdown in the event of a fault it is critical to get it right.  You could find a way to add an 'e-stop' switch to the dashboard, or maybe you could use the existing throttle position sensor's switch but disconnect the potentiometer component..

[Whales]

Stencil it with paper and paint it with conductive paint?  I have a sample of conductive acrylic based paint from Intelliparticle but I don't know how hard it is; perhaps there are conductive epoxies or DIY it with off-the-shelf epoxy and powdered graphite?

[JohanH]

Stencil it with paper and paint it with conductive paint?  I have a sample of conductive acrylic based paint from Intelliparticle but I don't know how hard it is; perhaps there are conductive epoxies or DIY it with off-the-shelf epoxy and powdered graphite?

Perhaps this is similar?

https://www.2spi.com/item/z05006/ (Carbon Conductive Paint)

It doesn't say anything about resistance (that I could find). At least the Intelliparticle says something.

[Haenk]

https://www.extra-classics.de/Drosselklappe-Instandsetzung-EML-M70-S70-8er-BMW-E31-7er-E32-750i-850i-CSI-13541736856-Reparatur/EC53081

Maybe contact these guys. They claim to have a solution for worn potentiometers. But they don't like working on previously opened ones...

[A320mech]

Thanks all for your replies, I find them quite informative. I know I can buy these remade and I've no problem shelling out the $400 for a pair of new boards, if anything to support the guys who go the extra mile to produce parts for our old cars as we were abandoned by BMW and their OEMs. The problem is the shipping and the German customs make that price quite steep when I can buy a complete pair of used throttle bodies for around 300 Euros. And I'm not going to learn anything new if I just let anyone else overhaul them for me. As long as I'm not stuck with an inop engine, I prefer to find a way to do it myself.

3D printing is probably not such a great idea given the heat these things experience, hobby-grade plastics will probably melt or at least warp.

Removing the board itself and replacing it with an appropriate Potentiometer and a switch on a cam is probably the most realistic way of going about it, if printing resistance traces on a PCB proves difficult. Having said that, the Potentiometer needs to have a good resolution and be resistant to constant wear under varying temperature extremes. BTW I've seen a circuit somewhere and to my untrained eyes, it looks like an adjustable voltage divider.

Replacing the entire circuit with something more modern is waaay more interesting and a subtle way to upgrade my beloved oldtimer without actually looking like it's been modified. I'm all for it, let me hear your ideas! In aviation, we use RVDT (Rotating transducers) to send position feedback of flight controls for example. I gave this thought but given the original ECU was made to operate a DC  motor and read off the potentiometer directly, I do not know how I could easily turn the DC power into an AC, drive the RVDT and then convert that back to DC voltages.

[mendip_discovery]

Dont forget they make throttle position sensors for motorcycles that just fit on the carb. I have a 3 pin one for my Yamaha TRX850 that might be easier to use.

[Miti]

There are still enough carbon traces there, they’re just not under the wipers. Can you change the wipers with some soft brush type wipers?

[langwadt]

I seem to remember an episode of wheeler dealers where they fixed a similar problem on a Maserati. So guy had a business fixing them by replacing the internals with a circuitry with something like an AS5600 magnetic angle sensor   

[A320mech]

Slightly bending the wipers 0.5mm inwards or outwards would make them rub on relatively untouched resistance on the boards and would probably be the quickest and easiest way to restore the function as if it were new. I am however scared of even looking the wrong way at those tiny whiskers as they are so thin I'm scared they'll break.