2006 850K Joystick Issues

[Bush_S2K]

All,

New to the forum. Helping a friend with some electronics problems on this 850K Series II. While I have experience with some machinery, I am not an everyday mechanic, BUT I am an electronics engineer with a lot of background in hydraulics, CAN, controls systems, etc. On to the problem...

Brought me the 850K travel joystick saying that forward and reverse movement was intermittent. Got to the point where it was a bear to operate because you had to hold the stick in the perfect spot to drive in any direction. We proceeded to disassemble to joystick to see what was inside. Two potentiometers and two pairs of microswitches. I checked all components with an initial outlook that all were "ok" with decent resistance readings while rotating the stick in all directions. Potentiometers can be finicky carrying current, depending on the type. Wrote down the numbers and began my research. Made by Contelec Swiss, 0-5K, basic stuff other than shafts are metal and 6mm in diameter. Couldn't find EXACT matches, but did find a VISHAY model that was similar with 1/4" shaft and exact type and overall dimensions. At $30 each, was worth a shot considering the cost of a new joystick. Received the new pots, disassembled originals out of stick to determine rotation and angular range. Originals were 225-240 degrees and infinite turns, new ones are 340 degrees and 1 turn. Not a huge difference, let's throw them in and if anything, it'll be slow. Sure enough, that was all true and original intermittent problem was FIXED, EXCEPT we had to tune each one to match the original calibration at neutral position. Did that with a few iterations between Forward and Reverse, Left and Right. Loaded onto the trailer to bring home to the shop so we could fine-tune the speed with some resistors added to match the original rotation angle.

Knowing my electronics, I added a 10K resistor across each pot to bring the total resistance down to 3.3K ohms. Resistors in parallel (5k * 10k) / (5k + 10k) = 3.3K. I determined the 10K resistor from an equivalence of X over 225 degrees to 5k over 340 degrees. I actually did this backwards first, but it made the machine run slower. It would not counter-rotate.

After adding the 10k resistor, it did NOT bring the machine back up to factory speeds. Now, I'm stumped. We had spoken to the local dealer before any modifications to the new pots, because I knew there had to be a calibration routine available. However, they are very reluctant to perform it, even if paid. They say it won't fix anything and we need to buy a new joystick.

Has anyone ever tried a repair like this?

Is there a calibration procedure we can run on the machine to reset range on the new potentiometers?

Thanks all ahead of time. We are hoping for a resolution, in any way.

 

[Dickjr.]

Pretty interesting approach on this repair. I have dabbled with electronics very little but finally learned that most parts are generic just need to know where to order them. I wonder if there is anyway to find them oem potentiometers or exact replacements or where the made an odd size just for CASE. I'm sorry I have no info to offer but I think its pretty cool that you so called cracked the code on this. I would think there will be a big market in repaired controllers in the near future. I do know that some manufacturers have things made to they're specs to keep these kinds of repairs from happening , back to dealer. Which I'm sure your more than aware of.

 

[Bush_S2K]

I searched long and hard to find the exact manufacturer and part number of those potentiometers. They are not available anywhere I know of, or was able to find, in the USA. They were obviously Swiss-made and may no longer be available. From talking to the dealer, they "used" to be able to repair the sticks with new pots, but they can't get them anymore, so that supports what I discovered. These items are very simple, so my replacements and method of integration should work, one way or another. If we can get this to work, great. If not, I would be inclined to hear an explanation on why not from CASE or DEERE.

 

[Dickjr.]

I hope you get this figured out. I have a Komatsu which uses basically the same stick set up. I doubt you will get an answer from a dealer as to why or how. CAT would definitely be a booger to try to repair needing CAT ET to calibrate it hence needing a CAT tech. This maybe simple to you but it pretty complicated to me but I have an understanding of how it works. It would be great if someone opened a shop to rebuild these type components to help save owner operators money. I think the steering control joystick is about 2200$ which at times is a hard one to swallow.

 

[DMiller]

Bush S2K, question and no I am not trying to steal the thread. If you wish PM me on this. You spoke of engineering on CAN Systems, is that related to Harley Davidson CAN-BUS style electronic engine controls? Insights?

 

[fpgm04]

Bush_S2K, Welcome to HEF !

As some of the real experts on the forum have not yet chimed in, let me offer the following which will also bump your thread.

There is not a dealer or owner joystick calibration on the K series that I am aware of. There is a throttle and pedal calibration for those associated sensors that is done through the instrument cluster, but the joystick as far as I know is precalibrated from the component manufacturer. Therefore, the dealer could not help you with his electronic tools for stick “calibration” in the joystick controller.

However as part of vehicle diagnostics, virtually every dealer electronic tool has ability to monitor the joystick’s controller reported percentage of stick movement for both the x and y axis. Therefore, it could confirm if you are not getting full range -100%..0%..100% positioning as reported/interpreted by the joystick controller. If not getting full range movement, perhaps you could adjust your resistors to produce values the controller will interpret as 100%. Ideally if you could get your dealer to hook up his tool and confirm stick position as reported from the controller viewpoint, it would be at least a very good piece of diagnostic information and possibly allow for fine tuning resolution/calibration via resistor adjustment.

That being said, I would recommend you confirm a few vehicle items to insure the slow speed is truly being caused directly by the joystick surgery as opposed to some secondary induced condition due to debugging etc.
1) Is the display menu item “MAX_SPxxx” at 100% ? (I seem to recall this can go as low as 70% and a factory reset of the controller would put it at 90%)
2) Is the operator speed selection at 100% on the instrument cluster using the joystick’s rabbit/speed-up button? (recall at each key-on, the default is reset to 60% speed)
3) If using reverse, confirm speed is not being limited by the reverse ratio rotary switch at 80%. Ie switch needs to be at 100% or higher position or reverse speed will be slower.
4) Is the brake pedal fully upright? Ie if not fully returned to full upright position speed would be limited. A simple pull on the brake pedal is a quick check to insure the pedal is proper positioned.
5) Is the joystick steering x-axis working correctly? Ie only y-axis not producing full speed?
6) Do you have the Owners/Operator’s manual to walk you through the instrument cluster menus?
7) Are you getting any fault conditions displayed on the instrument cluster?
8) Fundamentally, how “slow” is the slow speed condition?

I hope some of the items might help get you un-stumped.
Please keep us apprised of progress.

 

[Bush_S2K]

Thanks fpgm!

Essentially, I have matched the original rotation with resistors already. I don't have a good way to confirm that, but in theory, we should be back to factory range.

I appreciate all of your suggestions. I will print this off and see if he is willing to check those items.

I have strictly been relying on his knowledge of the machine's operation. We don't have any way to know track speed to match before and after repair. I do know that it is slow in both axes, forward/reverse and left/right. His measure of speed is rate of counter-rotation. After the first iteration without resistance adjustments, it would hardly counter-rotate. He says that it is still "slow" in that regard. We do have the manual with the cluster menus, and we have been through them, but i didn't know enough about the settings there to say that max_speed was 100%, etc. I do not know how to tell if there are fault conditions present, either.

 

[Bush_S2K]

After some conversation and more thoughts on paper, I decided that my first configuration of added resistors would have been correct. I need to match the original voltage range with angular rotation. The only way to do that is to extrapolate resistance to 340 degrees, then add that amount overall so that at 240 degrees range, we still have 5k ohms.

We also checked the brake pedal position, fault conditions, and reverse ratio switch. All check out fine. Right now, speed is down to 25% or so, which makes sense after realizing total range has dropped to 3.3K ohms.

The plan at this point is to remove the 10K resistors and do some trial and error on placement of additional resistors. We may have to re-zero pots manually, but that's not too big of a deal. A few iterations through forward and reverse should be good. I'm not sure that buying 5k ohm pots was a good idea. Maybe a 10K at 340 degrees would be fine, then divided down to match originals. Hopefully we can find a solution soon. For now, he has a job to finish with it, then we can get back to troubleshooting.

Thanks for any and ALL help.

Chad

 

[Dickjr.]

Keep us posted as to what works out.

 

[Bush_S2K]

Haven't had a chance to get back to this problem yet. I have an RFQ out to price new pots to be made specifically for this joystick with a component manufacturer who not only makes the pots, but also custom joysticks for other applications. Even if they're $100 each, it would be cheaper. I realized that electrically, there is no way to achieve the original range without re-gearing the new 340 degree pots, but i don't have time with the stick to make measurements and find gears that will fit the application. I'm sure they're out there, just a matter of downtime with the machine and reverse engineering the mechanics. Then, any joystick could be rebuilt with locally-available aftermarket parts.

Was hoping to have some of the greater minds chime in on this post with experience on this stick. If anyone has ever tried this before....

 

[Dickjr.]

You are the pioneer on this as far as I know. I would bet that any factory tech has been gagged about any of these repairs due to the potential income from repairs.

 

[Bush_S2K]

Received a quote for custom potentiometers to match the factory originals. Minimum purchase is 5 pieces and additional engineering fees. I'd like to know how many models use the same stick and if this is a useful repair to make. What would be a reasonable cost? The real question is, is this the only failure component? How do i separate this problem from others? I'm still working on a mechanical solution for standard pots. Gearing is the key to make this easy.

 

[GaryHoff]

Very interesting thread. I am always interested in repairing vs replacing.
The following models use the same joystick : 650k, 750k, 850k, series 1, 2 and 3. 650L, 750L, 850L, 750M, 850M. The larger dozers (1150k, 1650k, 1850K) were different.

Case just uses a Rexroth controller, so if you had Rexroth Bodas, in theory you could do a calibration. I have that program, but haven't made a cable to test it. There is even a factory diagnostic plug for the Rexroth controller in the harness.

 

[Bush_S2K]

Thanks GaryHoff. That's a lot of great information. I looked into the BODAS tool a bit. They offer a demo. I wonder if it has enough functionality to work with that joystick? We've had the stick off and there is only a 4-pin weatherpack connector on it. Is that the same diagnostic plug you're talking about? Do you know what the BODAS software costs?

 

[GaryHoff]

I will make a connector this week to try it. I am not a 100% sure that it will work. There is a 6 pin Deutsch connector, on the left side of the machine. It is referenced in the schematics also. The bodas software and cable was about $1500 Canadian.

 

[fpgm04]

Bush_S2K,
For clarity as Gary brings up the Bodas tool. Bodas works with Rexroth controllers, it does not work with other controllers such as the dozer’s joystick that I have ever seen.

That being said, Gary has a lot of experience, so perhaps he has some knowledge of Bodas that my mind has long forgotten.

However if you do have access to Bodas, it could potentially provide you the Rexroth controller’s interpretation of joystick position. Therefore, you might use it to confirm if the potentiometer surgery has resulted full range positioning like I described previously. (or conversely how far off the potentiometer is)

My recollection of Bodas is that it is painfully slow updating real-time data from this vintage of controller to the computer’s display, so be prepared to hold the joystick steady for many seconds to get an accurate reading.

 

[GaryHoff]

Well, as it turns out connecting using the Bodas software was not hard at all. In fact, I already had a connector. The machine I had access to was the case 650L, but for our purposes, this should be sufficient.

The case dozers use a complete Rexroth hydraulic drivetrain package. The pumps and motors are Rexroth, as well as the drive controller and joystick.

Lets start with the connector. All the Rexroth RC controllers use rs232 communication, so that means pins 50 and 51 on the controller are the communication wires. These already go to the 9pin deutsch connector on the dash. Perfect.

So make a wire with the following

At the rs232 connector (rs232 is also called a serial connector)

Pin 2 (rs232 Rxd)---------------> Pin G (9pin deutsch) ------------->Wire 250 pink (machine harness side)
Pin 3 rs232 TxD)---------------> Pin F (9 pin deutsch) ------------->Wire 251 green (machine harness side)
Pin 5 (ground) -----------------> Pin A (9pin deutsch) ------------->Wire 0 black (machine harness side)

Connect this, and you should be in business. You will also have to get and install a rs232 to usb adaptor. Make sure that you change your adaptor settings to COM 1, or this will not work. If you are like me and have a computer with a serial port still, then you will not have any issues.

If you are really lazy, you can buy the cable from Case part number "380000087" if it is still available.

Also, not to mislead anyone, but the pictures below are of Bodem/25 software. This is only available on windows xp and older. It was replaced with Bodas software. Bodem had a free mode, where you could monitor parameters. I don't believe that Bodas has this feature. This software is available for download at the Rexroth website.

 

[GaryHoff]

 

[GaryHoff]

As FPGM04 has mentioned, this is a older software, and is slow due to the speed limitations of RS232. Its, not too bad, just suffers from a 1 second lag or so on some functions.

This is not directly linked to the joystick, as the joystick is a CAN joystick. The drive train controller will be able to tell you all of the joystick functions. You can not calibrate the joystick. This is due to a controller software limitation put in by Rexroth. You can adjust the deadband of the joystick. It is set at 5%. It can go up to 10%. This may be all you need to make your joystick work.

I have seen other Rexroth RC controllers where you can calibrate the joystick. There is a few software versions, as the older case dozers used a different version, so I am not sure if the calibration feature is enabled or disabled in the older machines.

 

[atc007]

Hi guys. Has anyone come up with a remanned joystick yet,here ,aftermarket etc? What are my options for replacing one as chaep as possible? Thanx as always for the wealth of info in here.