Connecting Heusinkveld Ultimates to the Fanatec CSPv3 Controller Board

[Atak_Kat]

TLDR Link: Here

---------------------
Hi all,

I'm looking for a person knowledgeable in electronics, that might be able to interact with me as I move through a DIY project.

The project:
- very simply, I want to build some sort of simple interface that can take the direct signal 'output' of my HE Ultimate pedals (all are load cell), and test/convert that to a signal that can be 'input' to the CSPv3 interface/board.

Why?
- ultimately, I want to have the option to use my HE pedals, together with my Podium Racing Wheel F1 base (ie. PS compatible) to have an easier solution to use my existing wheel/pedals for playing also on Playstation.
- I KNOW there is the DriveHub alternative, but I simply do NOT want to do that.
- I already have the Fanatec CSPv3 pedals, but I get annoyed to change them on my rig whenever I want to play on PS. I don't use them otherwise.
- I'm not afraid to modify gear. I have already disassembled my CSPv3 pedals for other reasons, so I have the board/interface available.

My thoughts so far:
- The HE pedals are outputting signals from the load-cells on each (3x). As I see it, it's simple output, where the SmartControl box then does all the interpretation/work.
- The CSPv3 pedals have 2x pedals outputting simple hall-effect sensor output, and the brake is a load-cell output. Again, it seems to me that these outputs are all quite simple, where the CSPv3 control board then does all the interpretation/work.
- I would think, there is some way to build an basic electronic interface so that the direct output of the HE pedals (load cells) can be seen/understood by the CSPv3 control board.

Other:
- although I have ok electronics understanding, I'm probably in the category of 'knowing enough to be dangerous'.
- I have tools/equipment (ie. soldering station with variable power supply, basic testing equipment, etc).
- Not afraid to disassemble things for testing, assuming I can reasonably put it back together. I'm also not so afraid to go down the path of no return.
- I've searched around and cannot find any similar project online. In case anyone knows of a solution already there (other than Drive Hub), please let me know.

Specific ideas:
- how can I make a sort of interface that will take the output of the HE load-cells (2x) and convert that to something that would mirror that to a similar signal that would be coming from the 2x CSPv3 Hall effect sensor outputs?
- how can I test/modify the HE load cell out put so that it would mirror a similar signal to the CSPv3 load cell output.
- If I can do that, I think the rest is just wiring/connectors.

All of this I would be happy to share progress and lessons-learned in a thread here. But I would be grateful if there is someone with electronics knowledge, who may be able/willing to dialogue directly with me as I find my way.

Thanks

 

[stigs2cousin]

Well,

Disclaimer first, I probably know just enough about electronics to be dangerous

Firstly I´d find out if the HE brake loadcell would work with the Fanatec loadcell amp on the board.

If these two play ball you have two choices:

mechanical:
MacGyver hallsensors onto the Heusinkvelds and use the board as is.
HE stainless steel is rather tough, Good Luck drilling holes for mounts.
I´d glue them on.

or electronic:
find adjustable lc amps and put them between the HE loadcells and the Fanatec board.
You´ll have to Google pinouts yourself.
Before Brexit LeoBodnar was a good source for parts and knowledge.

Good Luck Carsten

 

[GeekyDeaks]

Ok, if I was doing this, the first thing I would do is establish what pins do what on both the HE and CSPv3 pedal connectors.

For example. On the hall effect sensors, I'd establish which pins are Vcc and GND and which one is the output signal. For the load cells, I'd want to know the Excitation pins E+/E- and the signal pins S+/S-

Once figuring out the pins, I'd want to the know the voltage ranges expected by the CSPv3 controller. You would be looking at a fairly big swing on the hall sensors (e..g close to GND -> Vcc), for the load cell you would need to measure the difference between S+ and S- and it's going to be really small, like 5-10mV small. For the hall sensor interface, a simple Instrumentation OpAmp should suffice to get the output range you need from the HE loadcell. The loadcell to loadcell may be a bit trickier, if you are super lucky, you might simply be able to use it directly, if not, I suspect you'll need to calculate what additional resistors you will need to make the HE bridge respond similarly.

 

[Atak_Kat]

Thanks so much.

mechanical:
MacGyver hallsensors onto the Heusinkvelds and use the board as is.
HE stainless steel is rather tough, Good Luck drilling holes for mounts.

Yep, I sat staring at the pedals for a couple hours yesterday trying to think how to do that. But I couldn't really envision a clean way to achieve it on the HEs. I do have a small hobby lathe and mill (Sherline) as well as a resin 3d printer that might come in handy, but I will leave this as a backup idea, if the first option becomes outside my capability.

if I was doing this, the first thing I would do is establish what pins do what on both the HE and CSPv3 pedal connectors.

I figured this. I'll do some digging and and further disassembly/photos over next several days. I'm going to start mainly with the CSPv3 pieces

Side note, some years ago I was investigating if it was possible to connect the HE handbrake output (I have the old one, not the current one) directly to the Ultimate control box (the old/original one had lots of extra pins to connect other DIY buttons/inputs. I've since upgraded my pedals with the Ultimate+ DIY kit, so now have the new control box) instead of having a separate USB connection just for the handbrake. I checked with Heusinkveld support and they did confirm this WAS possible but the below linked load cell amplifier was needed. I didn't go further with the project, but it might be a helpful hint for the remainder of this project.
Leo Bodnar Load Cell Amplifier


Also, here's a pic of the CSPv3 board, for possible future reference.
- am I correct that the little square chip in the middle (8 pins, 6N11010E) is the load cell amplifier?

 

[GeekyDeaks]

am I correct that the little square chip in the middle (8 pins, 6N11010E) is the load cell amplifier?

I think that is correct - that part is an Instrumentation Amplifier, which is typically the type you use in a load cell circuit to keep noise to a minimum (I used the INA122P for mine)

 

[Atak_Kat]

This is a photo of the load cell hall sensor board on the accelerator pedal, with related wires.
Hard to read it, but the board has written on it (bottom left) "Sensor HMC v3.7"
2nd photo with a closeup of the board with clearer markings.

I'm not sure how I would go about interpreting what each wire is, and then how to test what this is outputting.

 

[GeekyDeaks]

This is a photo of the load cell board on the accelerator pedal, with related wires.
Hard to read it, but the board has written on it (bottom left) "Sensor HMC v3.7"
2nd photo with a closeup of the board with clearer markings.

I think you maybe meant the hall sensor accelerator pedal? That IC is a Magnetic Displacement Sensor: https://www.farnell.com/datasheets/1681861.pdf

Are you able to safely plug it in and measure the voltage on the pins where the cable is soldered onto the board? This will let you identify Vcc and GND. You can then move a magnet near the HMC1501 IC to double check the final signal pin changes value.

 

[GeekyDeaks]

hmmm... I just realised I was assuming it was blindingly obvious how to identify Vcc and GND... if you are not familiar with this kind of thing, let me know and I'll take a look at the sensor datasheet to determine which pin is GND as the starting point as that will let you quickly identify the other two.

 

[Atak_Kat]

I think you maybe meant the hall sensor accelerator pedal?

YES, thank you, I edited the post

Are you able to safely plug it in and measure the voltage on the pins where the cable is soldered onto the board?

I can. It's all in pieces now but I can plug the main board into my PC, and the hall sensor into the board. I have a basic multi-meter to make checks.

I just realised I was assuming it was blindingly obvious how to identify Vcc and GND... if you are not familiar with this kind of thing, let me know...

Right again. This is why I need some hand-holding a bit. I'm not exactly sure how to do that safely and properly....

 

[Atak_Kat]

This is the load cell.

1. This is from the front (how it would be facing you if you were in front of the pedals). The piston of the brake pedal pushes directly on that small silver 'nub' in the middle. This is 3x metal pieces that are pressed/rivetted together. Front plate riveted to the centre (narrow) plate (the rivet at the bottom of the image.) Then that centre (narrow) plate is riveted to the rear plate (the rivet at the top of the image). Only way to get access to the centre (narrow) plate where the wires are connected, would be to grind/mill off the pin (the bottom one in the image) that is connecting the front plate to that middle (narrow) plate..

2. Not the best pic, but that's a side view. Taken from the right hand side. It's impossible to really see any wiring connection because there is a rubbery covering over the entire area where the wiring leads. The only way to remove it and see what's there would be as commented in 1. above.

3. A fantastic drawing of basically the same view in the picture in 2. above. A bit easier to understand what is there. The wires all connect to the middle (narrow) strip you can see in the 1st photo. That centre strip is the middle of the sandwich. The placement/ordering of the colouring of the wires is written, in case that helps in some way (I doubt)

4. Last is just the mapping of the 4x wires to the connector pins. I didn't test them (I cannot access the place where they connect in the load cell), but it's pretty easy to see them through the connector (not a great pic of the connector, but I can see quite clearly which wires are where)

I'll leave this for tonight. Thanks so much the guidance so far.

 

[GeekyDeaks]

ok, looking at the datasheet https://www.farnell.com/datasheets/1681861.pdf, I see it has the following pinouts:

This means the GND is on two pins (highlighted with yellow arrows)

Make sure your multimeter is measuring DC Volts with a scale in the 10's (it might be 20 on the meter). Pick whichever one is easiest to put the -ve probe of your multimeter on, then put the +ve probe on each of the solder joints of the lead (I'm assuming they are the ones I have highlighted with cyan arrows) and note down the values.

A value of 0v is the GND wire.
5v or 3.3v is probably Vcc (the power supply)
2.5v or 1.6v is probably the Signal

You should be able to verify you have the correct signal wire by moving a magnet close to IC1 and checking if the voltage changes. Usually the hall sensor will change it's output depending on the polarity of the magnetic flux, so you should be able to make the voltage go up and down by swapping the pole you present to the sensor.

I'll have a think about how to tackle the Loadcell next. It might be an idea for you to read up a little bit about how wheatstone bridges work first as then you won't be trying to probe it completely blind.

 

[GeekyDeaks]

for determining the loadcell pins, how comfortable are you with probing the top of socket X6 or the bottom where the pins are soldered?

 

[GeekyDeaks]

Only way to get access to the centre (narrow) plate where the wires are connected, would be to grind/mill off the pin (the bottom one in the image) that is connecting the front plate to that middle (narrow) plate..

I wouldn't worry too much about the internals. I'd recommend reading up a little about how load cells work as they are actually really simple, being just 4 resistors arranged in a wheatstone bridge. The resistors change value a very small amount when under stress and the wheatstone bridge allows accurate measurement of the tiny changes.

For the HE pedals, I'm going to say just e-mail Heusinkveld, tell them what you are up to and ask if they would not mind sharing the pinout details. They are usually very helpful.

 

[Atak_Kat]

Pick whichever one is easiest to put the -ve probe of your multimeter on, then put the +ve probe on each of the solder joints of the lead

Just to be sure, I interpret this meaning to connect the main board to PC/USB, connect the sensor to the main board, and then attache the -ve probe to either of the GND pins directly on the sensor (chip?). And then take the readings from the pins where the wire connects to the sensor board (yes, they are the pins you pointed in CYAN). No problem. Will do and report back.

for determining the loadcell pins, how comfortable are you with probing the top of socket X6 or the bottom where the pins are soldered?

No concern at all. Either way.

For the HE pedals, I'm going to say just e-mail Heusinkveld, tell them what you are up to and ask if they would not mind sharing the pinout details. They are usually very helpful.

Will do that. I will also ask them if they will tell me the specific load cells used so we can check their specs/ranges.

 

[GeekyDeaks]

Just to be sure, I interpret this meaning to connect the main board to PC/USB, connect the sensor to the main board, and then attache the -ve probe to either of the GND pins directly on the sensor (chip?). And then take the readings from the pins where the wire connects to the sensor board (yes, they are the pins you pointed in CYAN). No problem. Will do and report back.

yup!

For measuring the loadcell. Ideally you'll do the same thing and have it all connected, but all the probing will be done on the socket. The approach is to try and figure out which pins are which based on their voltage levels relative to the others. Using this diagram as reference:

Pick a pin and put the -ve probe on it, then measure the voltage of each other pin. Note the one that is the largest -ve. That pin will be Excitation-. If all are +ve, then you have already found Excitation- and have the -ve probe on it.

With the -ve probe on Excitation-, find the pin with the highest voltage. This will be Excitation+

The remaining two pins are the Output- and Output+. You can determine which is which by putting the -ve probe on one and measuring to the other (you need to put the meter scale to mV). If the voltage is +ve, then Output- is the one you have the -ve probe on, otherwise it's the other way around. The remaining one is Output+

You might want to consider recording the minimum and maximum mV between the Ouput- and Output+ pins are as that will give you a reference to aim for. I'd probably put the cell back in the pedal and get someone to press it whilst you are probing.

If you want to see if you are lucky (it's not going to break anything to try this), once you have the E-/E+ pins and the Output-/Output+ pins from HE you can try wiring it up and see if it works as is.

 

[Atak_Kat]

I'd probably put the cell back in the pedal and get someone to press it whilst you are probing

I was thinking to attach a C-clamp across the front/back plates of the load cell, and apply pressure that way? Or with the back laying flat on a piece of strong metal. I thought it would be easier and more controllable that way. Also easier and more stable to take the related mV readings since I expect that will be quite delicate re the positioning and stability of the probes?

Is that a stupid idea?

 

[GeekyDeaks]

I was thinking to attach a C-clamp across the front/back plates of the load cell, and apply pressure that way? Or with the back laying flat on a piece of strong metal. I thought it would be easier and more controllable that way. Also easier and more stable to take the related mV readings since I expect that will be quite delicate re the positioning and stability of the probes?

Is that a stupid idea?

It's not stupid at all, but just make sure you are applying the load on exactly the same points as when it's mounted in the pedal. I have had load cells give false readings when not doing this.

The other thing to consider is that the pedal may not be using the entire range of the load cell. To be honest though, first thing I'd do is just wire it up and see if it works....

 

[Atak_Kat]

Here's the result of the testing with the hall sensor board.
Notes:
- sensor board plugged to CSPv3 controller 'accel' input.
- CSPv3 connected to PC THROUGH the DD1 input (direct to the back of the DD1, NOT to a USB port on the PC). I did that because that's the most likely way I will use it. If it's better to do it connected direct to USB, let me know and I can redo.
- I had the Fanatec Control Panel app opened on the PC, and could see there the pedals were recognized as connected. Note that the default reading on in the app, for the accelerator, was about 5% 50% when the sensor was connected, but no magnet near the sensor. (both the brake and clutch read 0%, and neither of those were connected to the CSPv3 control board.)

RED wire (left pin in the pic) is clearly GND
GREEN wire (middle pin in the pic) has variable v result depending how you position/move a magnet near the sensor. (EDIT: I used the pivot pin with the magnet embedded in the end of it, from the pedal itself.)
- No magnet present, was about 1.24v. So approx 50% accelerator as shown in the Control Panel.
- Highest reading when moving the magnet was about 2.0v and the Control Panel showed that was 100% accelerator. It could get it to read a bit higher than 2.0v (maybe 2.1), but needed a very steady hand to move it past 2.0v.
- Lowest reading I could get was close to 0.0v dependent how I positioned the magnet. But anything between about 0.0v and 0.5v resulted in the Control Panel showing 0% accelerator. The accelerator graph in the app only started to increase when the readings moved above 0.5v (approx)

Then, this is how those translate to the CSPv3 board when connected to the accel. port.

 

[Atak_Kat]

Just a little additional info which may (or not) be interesting.

- I played around with my multimeter, using the continuity test (the buzzer that indicates a path exists between two points). No power or anything connected. Just probing the pins of the pedal input ports, with no connections.
- I found that for all 3x of the pedal input ports, that the outermost pins, were all connected.
Example. the WHT noted left-most pin in the diagram below, registered as connected to the same left-most pin on the middle port (brake), as well as the left-most pin on the right port (clutch).
In same way, the RED noted right-most pin in the diagram below, registered as connected to the same right-most pin on the middle port (brake), as well as the right-most pin on the right port (clutch).

 

[GeekyDeaks]

Just a little additional info which may (or not) be interesting.

That's actually a great observation!

I expected it across the Throttle and Clutch, but confirming it for the Brake means you have probably already identified the Excitation- (Black) and Excitation+ (Red) pins (assuming I have the plug you photographed orientated correctly?)