Connecting Heusinkveld Ultimates to the Fanatec CSPv3 Controller Board

[Atak_Kat]

So a bit of an update (while I wait for the amplifier boards to arrive from China)

In my earlier ordering frenzy, I completely forgot that I also ordered a couple of these from Leo Bodnar. And they recently arrived.

So, I decided to connect some wires, and really test if knowing 'enough' would be dangerous.
Short version, both the HE throttle and clutch inputs are recognized JUST FINE by the CSPv3 board and Fanatec Control Panel when running through these Leo Bodnar boards. As expected, the min/max range is a bit strange considering the output from the HE load-cells and our earlier testing/suspicions. But the Fanatec Control Panel is able to calibrate the min/max quite easily, and everything seems to work fine after calibration.

I have NOT yet connected to the PS5 to see what happens direct to GT7. The raw output from without calibration is a bit abnormal. Not sure how the PS5/GT7 will auto-calibrate yet. So I think I will wait for the other amps (should arrive soon, I think) which have more ability to fine-tune/adust the output on-board.

All the rest below is a sort of progress/update in case anyone interested.

I have struggled to think how this will look, when finished. I decided to make this without damaging/soldering to the CSPv3 board. There will be some short cables to mange though.
I just put the 3x RJ10 ports (where the HE pedals will plug in) directly above the Fanatec ports. Held there for now, simply with hot-glue. The idea will be to plug the HE pedals there, then the output on the back side wrapped around to be 'input' to the normal CSPv3 inputs. Not so pretty, but it will do the job and be better than too many wires dangling everywhere.

But here is a top view, where you sere the connections to the Leo Bodnar boards just free and not really secured.

So, I made a basic frame/holder in Fusion 360, then 3dprinted on my resin printer. The holes are spaced 40mm to allow eventual mounting to the profile pieces on my rig. More pics to explain shortly

The CSPv3 board will sit in the bottom frame, using the mount/holes already in the CSPv3 board. The bottom of the board is raised approx 7mm, to allow the connection cables to circle under the board to be connected to the CSPv3 inputs.

Here then is the 2nd tier added, that just sits above the CSPv3 board, also using the standard mount points, and some 'pins' coming out of the bottom mount piece. You will see that I'm planning to put those 2x Leo Bodnar boards on this 2nd layer. The slots in the 2nd tier are so that the Leo Bodnar boards can be inserted vertically in those slots.

Here you see the idea a bit better. Note that this pic is now from the opposite view as the others. There are some short slots in the 2nd tier, which allow the Bodnar boards to be set vertical. It's hard to tell in the photo, but the top edge of the Leo Bodnar boards is basically level/flat with the top of the RJ10 ports. So, if I make a cover/case, it should be quite simple to enclose everything without major interferences.
The wires you see ( there are 3x... 2x from the Leo Bodnar throttle/clutch outputs, and 1x from the straight output of the brake port) all have RJ12 connectors on the ends and will be simply routed back 'under' the CSPv3 board (there is a space there, between the board and the base of the 3d print), and then connected into the related CSPv3 pedal inputs. It will not be the cleanest result, but I couldn't think of another way for now.

So that's it for the moment.
I'll update later with another pic how it looks with all the wires connected. It will be a bit busy and I consider this a prototype/proof-of-concept. When everything proves working (also with the other amplifier boards), I will try to make a v2, and hopefully much cleaner.

 

[Atak_Kat]

Today, the small amplifier boards have arrived.
So I guess more testing and wiring in the coming days/weeks

 

[GeekyDeaks]

looks like you are still having fun!

It will not be the cleanest result, but I couldn't think of another way for now.

Honestly, it looks fine to me (I'm guilty of way worse! ) and the way you have done the 3d print supports keeps it pretty clean. I did think you would just make a separate board which had the HE connections going in and the CSP 'ready' connections coming out, but Im guessing the cover for the CSP board was integrated into the pedal assembly?

 

[Atak_Kat]

I'm guessing the cover for the CSP board was integrated into the pedal assembly?

Yes. It's normally mounted underneath the pedal tray of the CSPv3 with a sort of metal shield protecting it, but not really a proper box like there is with the HE Smart Control box.

I did think you would just make a separate board...

For sure that was the first idea. It's still like that now. Only that the separate board is just on-top of the CSPv3 board. In my situation, I will connect the HE pedals to the HE Smart Control box 90% of the time. I don't think there is any reason to use the CSPv3 board with PC games, but maybe some situation will appear in the future. This CSPv3 contraption is really only for PS5 use.....

I even thought to have a sort of simple 'switch' that could be simply flipped to determine if the pedal outputs were going to the HE Smart Control or the CSPv3 inputs (instead of unplugging/plugging the 3x HE pedal outputs from one device into another). I have no idea how to do that, but perhaps another project that would need advice. HINT HINT...!!!

I was going to make a cover to go over top the entire thing, but I'm pretty sure I will redesign and optimize the whole configuration further. I'd like to end up with something that looks reasonably clean/contained, even if I will have it mounted hidden under my pedal plate. I should really direct solder the RJ10 port wires directly to the terminals on the CSPv3 board, but I'm not that brave (at least not today.....)

Comprehension check on the new boards though:
- 1 of the screw adjustments will be the gain, I guess. I think I understand that, and likely we go close to the max?
- the other I understood is an 'offset' but I am not smart enough to really understand that. I assume it is simply a way to shift the output signal in some way?? Either higher/lower voltage from the raw gain/amplification??

Thanks very much for the comments, encouragement, and guidance/education so far. Very, very appreciated.

 

[GeekyDeaks]

Comprehension check on the new boards though:
- 1 of the screw adjustments will be the gain, I guess. I think I understand that, and likely we go close to the max?
- the other I understood is an 'offset' but I am not smart enough to really understand that. I assume it is simply a way to shift the output signal in some way?? Either higher/lower voltage from the raw gain/amplification??

There is likely to be a bit of a 'dance' between the two. The gain should be 'how much the voltage rises for a given load on the pedal'. The Offset is the voltage when there is no load on the pedal'. As you may have realised, where you start from (the offset) alters how close you are already to the max, which means you need less gain to reach it.

 

[stigs2cousin]

That looks great, you seriously have come a long way

But just for my interest, why change the control box depending on the gaming device?

If your "contraption" works, it works.

I know it´s nice to have options, but than, how often do you really change the device?

I change steering wheels and shifters regurlarly, but that´s about it.

I´ve had second ( and third) wheel set´s for (motor) bikes and mountain bikes and i have four tyre sets for my car. (summer, snow, drift and track) but i do change less often than I would have imagined.

So, if it works to your liking, do you really need the switch?
( Or do you do it because you can and love the challenge?)

MFG Carsten

 

[Atak_Kat]

But just for my interest, why change the control box depending on the gaming device?

So, if it works to your liking, do you really need the switch?
( Or do you do it because you can and love the challenge?)

Honestly I cannot think of a really good answer. I could invent some bullshit about concerns for added latency or signal range (resolution) as a result of my device..... but it seems unlikely that would be anything noticeable or relevant. Maybe some sort of interference will become an issue??I do like the HE Smart Control software, but I also have to be honest that once calibrated, I rarely ever make changes there. In fact, there can be some advantages to just running through the CSPv3 board all the time (like being able to to directly change the brake pressure sensitivity through the wheel menu, which is something I find helpful in-game).

I probably just convinced myself to run it that way (100% HE pedals connected through the CSPv3 board), until a real-world issue comes up.

There is surely a bit of the DIY challenge as well, but the more I think about it, the more I realize there are many other more interesting projects instead of this 'switch' idea.

 

[stigs2cousin]

So of you do it it's "because you can "

 

[GeekyDeaks]

I even thought to have a sort of simple 'switch' that could be simply flipped to determine if the pedal outputs were going to the HE Smart Control or the CSPv3 inputs (instead of unplugging/plugging the 3x HE pedal outputs from one device into another). I have no idea how to do that, but perhaps another project that would need advice. HINT HINT...!!!

I'm not au-fait with doing this kind of thing with an analogue signal. You can probably use a 2:1 multiplexer for the purpose as they are used for audio/video signals, however the main red flag I have is that the HE and CSP run at different voltages, so you need to be 'super' careful with your circuit and ensure the supply voltage to the pedals is properly isolated from the supply voltage to the multiplexer or else you might damage the CSP board by accidentally shoving 5v across it.

 

[GeekyDeaks]

So of you do it it's "because you can "

I think that pretty much sums up the motivation behind most of my projects!

 

[Enzo Fazzi]

Have you considered any issues with linearity? Hall effect throttle pedals often have non-linear output for travel (input), simply because magnetic fields aren't linear in strength. So potentially, the CSPv3 board might apply some correction for this, but then ends up making our loadcell output non-linear instead. (Our throttle/clutch pedals are linear)

 

[GeekyDeaks]

Have you considered any issues with linearity? Hall effect throttle pedals often have non-linear output for travel (input), simply because magnetic fields aren't linear in strength.

This is a good point worth checking. If they are not simply rotating the sensor in the field then it's probably not going to be linear.

 

[Atak_Kat]

so you need to be 'super' careful with your circuit and ensure the supply voltage to the pedals is properly isolated from the supply voltage to the multiplexer or else you might damage the CSP board by accidentally shoving 5v across it.

Thanks for the ongoing education. Yep. We will put this 'switch' idea into the bin. Doesn't really make sense on several different levels.

 

[Atak_Kat]

Have you considered any issues with linearity? Hall effect throttle pedals often have non-linear output for travel (input), simply because magnetic fields aren't linear in strength. So potentially, the CSPv3 board might apply some correction for this, but then ends up making our loadcell output non-linear instead. (Our throttle/clutch pedals are linear)

Thanks so much for your comment!

From the initial testing I didn't notice anything abnormal across the travel range (after calibrating the min/max in app). Did I somehow make a proper test? No.
The Fanatec Control panel doesn't have any way to control linearity, as far as I know. That's also a pretty good reason to keep using the HE Smart Control box/app as default, where linearity customization is embedded. Also means that any existing muscle-memory (especially braking) is not confused/upset.

At the moment, the only major hurdle I see is the next testing with PS5/GT7 to see if/how the output will be seen, and hopefully calibrated (I am not aware of any calibration feature in GT7, so we need to see how the 'resting' output will be calibrated (or not) in GT7.
Those new/other amplifier boards (from China) might help with that, but we will find out in the coming days/weeks.

 

[Atak_Kat]

Some quick testing with PS5/GT7 today. Here's what I found
- Just to note, all below was simply with the device as I last reported its progress. So still with the Leo Bodnar load cell amp boards installed. I have not yet done/tried anything with the other boards that have some adjustment capabilities.
- It does work. GT7 re-calibrated the min/max input of the throttle pedal seemingly ok.
- However, there is definitely a non-linearity with the throttle.
- To be fair, I never really checked this before with the straight CSPv3 inputs and how their linearity behaved in GT7. And, since I've basically disassembled those pedals, I don't really want to go back to do that.
- I tried to measure the HE throttle linearity (physically using the shaft/rod that extends out the back of the spring assembly when depressed) vs the game graphic showing the throttle input:

- the first 25% (approx) of throttle input was essentially deadzone. It didn't register on the game graphic.​

- then from about 25% to 75% throttle input, was seen in game as about 0% to 50% throttle.​

- then from about 75% to 100% throttle input, was seen in game as about 50% to 100% throttle.​

So there is a very clear non-linearity resulting here. I'm a bit curious why there is approx 25% deadzone though.... That seems like a lot to me, and I never really took note before whether that was the same when just having the CSPv3s connected.

Later I will recheck it in a similar way through the PC and the Fanatec Control Panel to see how different it is there, vs. GT7.

​

 

[Atak_Kat]

Out of curiosity, I checked the throttle linearity when connected to the PC, using the Fanatec Control Panel App (it shows also the % reading throughout the range)
- As before, I measured the HE Throttle pedal travel (the distance the centre rod extends out the back of the pedal base), at various intervals, and then reading the %Throttle showing in the Fanatec Control Panel App.
- Pedal was calibrated in-app for min/max without any manual dead-zones.

Looks quite linear to me. Measurement method was not so perfect, so I'm gonna assume any variability from the straight line is measurement error.

Which to me implies that it is GT7 that is applying some sort of input curve.
It would be really great if ANYONE might be able to make a sort of simple test with their hardware (I think any pedals) and GT7, just to see if in fact they find a similar input curve in from GT7. I only checked deadzone (amount of travel before the first input registered in the GT7 graph), amount of travel when 50% registered in the GT7 graph, and amount of travel when 100% registered in GT7 graph.

This below would be the approx curve being applied in GT7.
Note that I only took 4x readings when connected to GT7, and the method was quite rough. In any case there clearly is a curve being applied, even if not exactly as below.

 

[Atak_Kat]

More testing with the PS5 and GT7 today.

Lo and behold, I discovered that there *is* a pedal calibration utility in GT7. And it's actually not so bad. You can set the starting deadzone (if you want, there is a slider) and also you can configure a top-end deadzone by simply holding the throttle a bit short of the maximum travel. It also nicely shows the % input being received.

I found that the throttle input was registering completely linear. 25% throttle was 25% of the pedal travel. 50% was 50%, etc.

So I have no idea why in game it was showing different on the throttle bar/graph.

I did play about 2 hours today in GT7 (mostly license tests since I'm not through that yet, and a few quick races just to test a bit longer game play. For me it was fine and I'm gonna call it all a success, at least for my purpose and what I was trying to accomplish.

At this point, the only thing left is to substitute the other load cell amp boards (with the adjustability). At this point, there's not really much need, since everything seems to work just fine now with the Leo Bodnar load cell amps. It will depend how bored I am and looking for a 'project' one day.

In the next couple days I'll try to draw up a more complete 'final' diagram of the wiring done, for a sort of final reference for anyone else that may be interested to try this.

A HUGE thanks to @GeekyDeaks for all the patience and education along the way.
(Also, I wouldn't mind changing the title of this thread to something that better describes what it's about, in case anyone knows how to do that???)

 

[GeekyDeaks]

kudos for getting it all working in less than a month chap! @Neilski should be able to help you change the title to something appropriate for the project...

 

[Neilski]

Also, I wouldn't mind changing the title of this thread to something that better describes what it's about, in case anyone knows how to do that???

What title would you like?

 

[Atak_Kat]

What title would you like?

How about something like>
Connecting Heusinkveld Ultimates to the Fanatec CSPv3 Controller Board
(or something like that, for anyone that may search for something similar in the future).