Problem with Raildriver throttle and reverser

Hey everyone, I just received the Raildriver I purchased, but after putting some time into it, I'm finding that there's two problems:

1. The throttle seems to have a HUGE deadzone when throttling up or down. I basically have the exact same issue described here: http://forums.auran.com/trainz/showthread.php?25768
2. Frequently, when I change the reverser from forward to neutral (while my train is stationary with no power applied), the reverser will refuse to switch, and sometimes will even jump to reverse!

I was wondering if anyone else is running into the same issue with their Raildriver, and if so, what the fix is? Please let me know, thanks!

ExpertNewbie said:

Hey everyone, I just received the Raildriver I purchased, but after putting some time into it, I'm finding that there's two problems:

1. The throttle seems to have a HUGE deadzone when throttling up or down. I basically have the exact same issue described here: http://forums.auran.com/trainz/showthread.php?25768
2. Frequently, when I change the reverser from forward to neutral (while my train is stationary with no power applied), the reverser will refuse to switch, and sometimes will even jump to reverse!

I was wondering if anyone else is running into the same issue with their Raildriver, and if so, what the fix is? Please let me know, thanks!

Click to expand...

No such problems with mine. You are using it in Cab mode, I hope. I also would make sure you run though the calibration again to make sure it is correct.

No problem here. I use mine every time I run TS12 and it works perfect.

Try recalibrating it. I have found from time to time the calibration gets a little off, or not to my liking. Be sure to have the levers in the exact spot you want them when calibrating.

Also, make sure your not running Macroworks. That program is for some other sim, but causes strange things in TRS.

After going through the entire calibration process, many people (myself included until I learned better) zip right past the last and most important step - park the controls (reverser to neutral, throttle closed, train brake to CS, independent brake to release, headlights & wipers off) before exiting, that ensures when the calibration is saved it's reading the default positions correctly.

Sorry, I forgot to mention, I calibrated the controller about a dozen times (and remembering to reset all the controls at the end) to no avail. I also tried it with 3 sims (Trainz 12, Train Simulator 2013, and Run 8) and all 3 exhibit the same issue. It seems to be an inherent issue with the controller itself.

ExpertNewbie said:

Sorry, I forgot to mention, I calibrated the controller about a dozen times (and remembering to reset all the controls at the end) to no avail. I also tried it with 3 sims (Trainz 12, Train Simulator 2013, and Run 8) and all 3 exhibit the same issue. It seems to be an inherent issue with the controller itself.

Click to expand...

Raildriver has really good tech support. I would contact them http://www.raildriver.com/

I verified with PI Engineering's SDK that the throttle stick is indeed not sending any updates for the entire travel through the 'dead zone', so it looks like a hardware issue unfortunately. The reverser stick also exhibits a similar problem when moving from forward to neutral. The readings from the SDK tester app shows that the reverser updates properly while moving the stick down until it snaps into the neutral position, but then the reading stays on what it was right -before- it snaps into neutral. This doesn't occur when moving it from reverse to neutral, it gives the proper reading for neutral.

Both brake sticks perform well though. The train brake has a bit of play before it starts sending over updates, but the indy brake is spot on; moving it a tiny bit sends an update as expected and there's no play that I can see.

I've also noticed that the LED, the right number's lower left segment is dimmer than the rest of the LEDs and looks like it's pulsing.

I'll post some pics up later today in case anyone else runs into similar issues.

I've e-mailed PI Engineering and I'm awaiting a response now. Unfortunately, all signs point to a defective unit. Hopefully I can get a replacement soon. I really like the idea of the Raildriver and I'm anxious to try out a perfectly working unit.

I don't have one yet, but if you've verified with PIE, then they'll probably setup an RMA to get your Rail Driver replaced. Make sure you saved all your packaging so you can return it to them.

John

I have heard infrequent reports of defective ones, but it's pretty rare - there's a reason for the high price tag, the quality is unbelievable. I bought two back in March 2003 when they first came out, both still work perfectly despite almost 10 years of use and abuse.

Couple of things I would check;
(1) Win7 or Vista? Checked that it's not a UAC issue?
(2) Tried unplugging and plugging into a different USB port?

I'm on Win 8 and I've been using the same USB port. I don't think it's either of those things since the controller works fine (it's just 2 out of the 4 sticks that don't behave as expected), but I'll go home and try another USB port if only just to eliminate it as a possible cause.

Not familiar with anything after Win XP, but if Win8 has the same kind of UAC as 7 and vista I would double check that it's not something stupid like UAC saving the calibration data to a virtual folder and leaving the original file the way it was. After running the calibration check the date and time on the ModernCalibration.rdm file in the TS12 folder.

ExpertNewbie said:

I'm on Win 8 and I've been using the same USB port. I don't think it's either of those things since the controller works fine (it's just 2 out of the 4 sticks that don't behave as expected), but I'll go home and try another USB port if only just to eliminate it as a possible cause.

Click to expand...

Do you have a windows 7 machine you can try it on?

I do, but I don't have the game loaded. It would take me several hours to set it up.

I could, however, compile the tester app from the SDK and then try it out that way. I'll see what I can do when I get home.

Although I wasn't able to get to a Windows 7 computer today, I did modify the test app that was included in the PI Engineering SDK to graph the raw values coming from the Raildriver and this is what I got:



This is for the throttle. I began by moving the throttle back and forth quickly without stopping from 0-50%, then from 0%-100%, then from 50%-100%. From 0-50%, there is hardly any response at all from the Raildriver. From 0-100%, it should look like a sine wave, but you can see that it flattens out when it gets to the top and bottom (when I changed directions). This was the dead zone I was talking about; when you reverse directions, you have to move it a certain distance before the Raildriver starts outputting updated values. From 50-100%, you can see that the throttle is more responsive, but still not half the sine wave compared to the 0-100% values (looks more like 10-20% to me).



And this is the graph for the reverser. I moved it from Neutral to Forward, to Neutral, then Reverse, and repeated this, pausing for a moment at each position. It's very obvious that when I move the reverser from Forward to Neutral, it rests at a higher value than when I move the reverser from Reverse to Neutral. This causes the game to think that I'm still going forward when I move the reverser from Forward to Neutral.

Oh, and I also tried out the Raildriver on several different USB ports. Same issue as stated in my original post. I've not heard back from PI Engineering yet. If they don't get back to me by tomorrow, I will give them a call and see what they can do for me.

I would also really appreciate it if someone with a Raildriver could run the app I modified and see what kind of behavior they get out of the throttle and reverser. The app I wrote looks like this:



It's a modified version of the app that comes with the SDK. The two progress bars show the position of the throttle and reverser, and the graphs below show the positions over time. The list on the left shows the position of all the levers and switches. It's these values that I'm graphing out.

Here's a link to the app:

http://www.filedropper.com/raildrivertester

It requires .NET 3.5, but most people should have that installed. I know it's a tall order to trust a complete stranger with an .exe file, but if any good Samaritans with a Raildriver is willing to try it out, here's what you need to do in order to get it running:

1. Plug in your Raildriver if it isn't already.
2. Run the app. You may get a message about Telerik controls being only a trial (this is what I'm using for the graphs). Click ok on the dialog.
3. Click on the Enumerate button. Your Raildriver should show up in the dropdown list.
4. Click on Setup for Callback. There will be no confirmation or message.
5. Start moving the sticks around. You should see the list begin to fill up with data.
6. Move the throttle and reverser to its extremes to calibrate the progress bars.
7. Click on the "Speaker On" button at the bottom to start graphing. I changed the behavior of the button but not the button text.
8. Move the sticks around like how I did to create my graphs in the previous post.
9. Click on the "Speaker Off" button to stop graphing.

If you do run the test, please let me know what you get! I'm looking for confirmation from another user that I do indeed have a defective unit. I think it will help the RMA process if I can show solid proof that my unit is indeed behaving erratically. Thank you very much!

I don't have a raildriver, as I said before, so I can can't test your program.

My gut feelings is there's something wrong with the hardware. I wouldn't fuss with the device anymore and would go through the RMA process and be done with it. The worst thing you can do is break something else and be stuck with a device you can't return.

John

ExpertNewbie said:

Although I wasn't able to get to a Windows 7 computer today, I did modify the test app that was included in the PI Engineering SDK to graph the raw values coming from the Raildriver and this is what I got:

This is for the throttle. I began by moving the throttle back and forth quickly without stopping from 0-50%, then from 0%-100%, then from 50%-100%. From 0-50%, there is hardly any response at all from the Raildriver. From 0-100%, it should look like a sine wave, but you can see that it flattens out when it gets to the top and bottom (when I changed directions). This was the dead zone I was talking about; when you reverse directions, you have to move it a certain distance before the Raildriver starts outputting updated values. From 50-100%, you can see that the throttle is more responsive, but still not half the sine wave compared to the 0-100% values (looks more like 10-20% to me).

And this is the graph for the reverser. I moved it from Neutral to Forward, to Neutral, then Reverse, and repeated this, pausing for a moment at each position. It's very obvious that when I move the reverser from Forward to Neutral, it rests at a higher value than when I move the reverser from Reverse to Neutral. This causes the game to think that I'm still going forward when I move the reverser from Forward to Neutral.

Oh, and I also tried out the Raildriver on several different USB ports. Same issue as stated in my original post. I've not heard back from PI Engineering yet. If they don't get back to me by tomorrow, I will give them a call and see what they can do for me.

Click to expand...

USB is a data connection, so obviously, there is an A to D converter in the Raildriver that is not working properly. Could be the circuitry around the pots themselves is faulty, gnd, + voltage, or the A to D converter itself is defective. Anyway, it's broken apparently, and they should probably replace your RD, since you just bought it.

By the way. I taught component level electronic techs for most of my working career, and you did a nice job of troubleshooting that problem. I would have downloaded that sdk app, but the site on that link wants credit card, etc, even for the free version and I hate to do that.

I hit the reload button on that web page and I get the download button replacing the sign up button. Downloaded twice, got a 5.8 meg executable, got .NET Framework 3.5 installed, but whatever I do I get an error trying to run the utility.

Thanks Bob. Sorry about the link, I wasn't aware it wanted your CC details. It's been getting harder and harder to find a file sharing site that doesn't want your firstborn child just so you can share a file. I'll contact PI Engineering and try to get that RMA set up. I still haven't received an e-mail reply from them yet, so I think I may just go ahead and call them instead.

In the meantime, I re-uploaded the tester app here if anybody wants to play around with it (hopefully this host won't ask for your signature in blood):

https://anonfiles.com/file/1dd9627bc3b1252a4f2ec58ff04cc713