With the latest build plunger behavior is back to normal, thanks Mike ![]()
++
Posted 26 January 2017 - 04:22 PM
Little help needed...
I installed the KL25Z as per the directions and that seemed to be successful with all firmware up to date. I then followed the directions to turn on the nudge in VP, again following the directions in the Pinscape setup guide. With the ball in the outlane, I do not get any nudge unless I physically pick up the KL25Z and turn it sideways - and then just a little bit of nudge can be seen on the screen. I'm sure this is just a settings issue, can someone please advise on how to fix? Much appreciated.
EDIT: After calibrating the device in Windows it seems to work a bit better. I have noticed that for some reason if I nudge it several times in a row, the table incline changes as if the table were on its end and the ball will shoot down to the bottom of the table.
EDIT 2: (Really sucky with wiring) I'm wondering how the wires connect to the KL25Z from the potentiometer? Do these need to be soldered (ugh) or do they connect with some sort of clip?
Edited by Michael Goldsmith, 26 January 2017 - 06:48 PM.
Posted 26 January 2017 - 06:54 PM
Little help needed...
I installed the KL25Z as per the directions and that seemed to be successful with all firmware up to date. I then followed the directions to turn on the nudge in VP, again following the directions in the Pinscape setup guide. With the ball in the outlane, I do not get any nudge unless I physically pick up the KL25Z and turn it sideways - and then just a little bit of nudge can be seen on the screen. I'm sure this is just a settings issue, can someone please advise on how to fix? Much appreciated.
EDIT: After calibrating the device in Windows it seems to work a bit better. I have noticed that for some reason if I nudge it several times in a row, the table incline changes as if the table were on its end and the ball will shoot down to the bottom of the table.
First, you should go back and delete the Windows calibration data. That interferes with the self-zeroing calibration on the device. To delete the Windows calibration, return to the "Set up USB Game Controllers" control panel where you did the calibration in the first place, double-click the device in the list to bring up properties, click the Settings tab, and click "Reset to default".
What you want to adjust instead is the sensitivity in VP. Open the VP editor and select Preferences > Keys from the menu. The key settings are "X-Gain" and "Y-Gain" over on the right side of the dialog about halfway down. These are basically scaling factors that determine how much acceleration is applied to the ball for a given reading from the accelerometer. Higher values mean more acceleration. If you're not getting enough acceleration, increase the values. If you're getting too much, decrease the values. VP 9 and VP 10 have different scales for these, and you have to experiment a bit to find the right values for your system, because the results can vary a little bit according to the particulars of your system's performance. Typical values in VP 9 are around 1000, and in VP 10 they're around 100.
Posted 26 January 2017 - 07:37 PM
mjr - thanks for the info. I'll do as suggested. I think i edited my post after you answered the other q's, but i'll ask again. what is the connector type (some sort of pin connector) to connect the wires to the kl25z controller? Sorry if this is a noobie question...
Edited by Michael Goldsmith, 26 January 2017 - 07:37 PM.
Posted 26 January 2017 - 08:19 PM
mjr - thanks for the info. I'll do as suggested. I think i edited my post after you answered the other q's, but i'll ask again. what is the connector type (some sort of pin connector) to connect the wires to the kl25z controller? Sorry if this is a noobie question...
Edit: I misread your post; thought you were talking about the optical sensor. You just need ordinary 0.1" pitch pin headers for the KL25Z. The parts list (http://mjrnet.org/pi...p?sid=partslist) has options.
That's actually a really tough question! As far as I can tell, there is no standard connector for it. They must have intended OEMs using the device to create their own custom cable for it, which leaves us DIYers to improvise something on our own.
Fortunately, the pads are fairly large and have little holes in them, so it's not too hard to solder ordinary wires to them. I used 24 gauge on mine. That's just a hair too thick to fit through the holes, so I soldered the wires to the surface of the pads. If you went down a couple of notches, maybe to 26 or 28 gauge, you might be able to fit the wire through the little holes, which would make soldering easier.
I also created a little custom PC board that you can use for the connection. It's especially good if you're using the expansion boards, because it's set up for a ribbon cable with a layout matching the expansion board plunger header, so it makes for a very tidy setup. If you're not using the expansion boards, it'll still work, but you'd want to just run individual wires to the KL25Z pins instead of using the ribbon cable, since the KL25Z pins aren't grouped as nicely as on the expansion board. The other nice thing about the connector board is that it has all of the pin-to-pin connections for the sensor on board.
Here's a link to the board on OSH Park:
https://oshpark.com/...ojects/eHHhonYk
You can order it from them for $3.20, which will include shipping and 3 copies of the board. You can also grab the EAGLE files and have it made elsewhere, but it's hard to imagine doing better than $3.
If you do go with the board, you can either try to solder the pads on the board directly to the pads on the sensor, or you can run some fine-gauge wire through the holes on each and connect them that way. I think the latter is the easier approach.
Edited by mjr, 27 January 2017 - 01:31 AM.
Posted 26 January 2017 - 08:27 PM
I was hoping something like this (http://www.globalspe...c7510690f59.png) would work...
Posted 26 January 2017 - 08:39 PM
I was hoping something like this (http://www.globalspe...c7510690f59.png) would work...
Edit: I thought you were talking about the optical sensor. Those should work fine for the KL25Z headers.
If you look at the mechanical specs for that one, I'm pretty sure you'll find that those leads are way too big for the holes on the board. Those probably aren't ideal anyway since you'd need a good deal of vertical clearance for them, and space around the plunger area will probably be pretty tight when you get everything together.
If you do find something that works, though, definitely let me know.
Edited by mjr, 27 January 2017 - 01:31 AM.
Posted 27 January 2017 - 01:30 AM
I was hoping something like this (http://www.globalspe...c7510690f59.png) would work...
You know, I misread your original post. I thought you were talking about the optical sensor (TSL1410R). The KL25Z is easy. The pins you found will work, as will any standard 0.1" header pins. The parts list (http://mjrnet.org/pi...p?sid=partslist) includes options.
Posted 03 February 2017 - 10:17 PM
I've just posted updates to the various pieces of Pinscape software with some nice new features. As always, you can find everything on the Pinscape site (http://www.mjrnet.org/pinscape).
Setup tool: There are now built-in testers for your feedback device outputs, button inputs, and TV relay. These should make it a little easier to set up a new system and troubleshoot any problems.
Firmware: This has some little additions needed to support some of the new testing features in the config tool as well as the new LEDWIZ.DLL virtual unit capability (which we'll come to shortly). It also has a few bug fixes and performance improvements.
One of the big bug fixes is that the flash writer (for settings updates) should be more reliable, hopefully 100% reliable. The KL25Z's flash writer is notoriously prone to random freezing and crashing, and no one seems to have a good handle on exactly why; if you've ever had the unit freeze up during a config settings update, or come back after an update without the changes sticking, you've seen this in action. I think I've finally gotten to the bottom of this, in the sense of figuring out what actually causes the problem and how to avoid it. But it's hard to be absolutely certain with any amount of testing because there's always been a big random element to it, so there might be some other reason it's happening that I haven't tracked down yet. If you do see any settings update problems, let me know - and please take note of what color the LED is showing if the unit freezes up on you, since that indicates which step in the process is responsible.
LEDWIZ.DLL: The replacement LedWiz DLL now lets you access ALL of your Pinscape ports from legacy LedWiz-only software, even if you have more than 32 ports. It does this by creating "virtual" LedWiz units to represent additional ports beyond the first 32. Say you have a Pinscape unit with a power board and a chime board, for a total of 73 ports, and say you've assigned it as LedWiz Unit #8. The DLL will give you access to ports 1-32 on LedWiz #8 as usual. But it will also create a "virtual" LedWiz #9. The 32 ports on this fake unit #9 will represent Pinscape ports 33-64. The DLL also creates unit #10 to represent ports 65-73. So older programs like Future Pinball will think you have THREE LedWiz units in your system, #8, #9, and #10, so they'll be able to access all of your ports.
Keep in mind that LedWiz ports are always numbered 1-32, so Future Pinball won't know anything about "port 65". So you'll have to do a little arithmetic to make all of the port assignments. Ports 1-32 on the second LedWIz unit represent Pinscape ports 33-64, and ports 1-32 on the third unit represent Pinscape ports 65-96. By the same token, if you don't have enough ports to fill out all 32 LedWiz ports on the last virtual unit, it will still look to everyone like that unit has 32 ports, since the LedWiz protocol is hard-wired for 32 ports on every device. That that's okay; the ports that don't physically exist will simply ignore any commands sent to them.
The virtual units will always be numbered consecutively after the "real" first unit number you assign in the config tool. So if you assign it as LedWiz #1, you'll get virtual units #2, #3, etc. If you assign it as LedWiz #8, you'll get virtual units #9, #10... The highest possible unit number is #16, so if you push past this the virtual units won't be there. Also, if you have a real LedWiz at unit #N, the DLL gives the real unit precedence, so it won't create a virtual unit at all in that slot.
NewLedTester: The LedWiz DLL download includes a spiffy little tester program called NewLedTester. I wrote this mostly to test the new virtual LedWiz scheme, but it's pretty handy for general LedWiz testing as well. It does basically the same thing as SimpleLedTest, but the UI is a little nicer: e.g., you can control multiple ports at once, test the various LedWiz flash modes, and leave ports on while you test other ports. It's also 100% open source, of course, so you can customize it or extend it if you want.
Edited by mjr, 03 February 2017 - 10:20 PM.
Posted 04 February 2017 - 07:53 AM
Nice Mike, I will test these new features ASAP ![]()
Edit : I've just updated Config Tool and firmware, seems great ![]()
The issue I have right now is with NewLedTester.exe : when I launch it nothing happens ![]()
No process in the background, no window opening, nothing. I'm running Windows 10 x64 up to date.
Should I move all the files in the archive somewhere else ?
Thanks for your great work Mike ![]()
++
Edited by shadowshd, 04 February 2017 - 06:41 PM.
Posted 04 February 2017 - 08:02 PM
Mike,
I'm in the process of building my first Pinscape setup, and have been playing with the optical plunger sensor. Everything seemed to be working well in the config tool, but I'm seeing odd plunger behavior in VP. The virtual plunger responds perfectly if I quickly pull & release the physical one. However if I pull back and hold, or slowly move the plunger rod, the virtual one will 'release' on a regular interval - about every 3 seconds. This can be reproduced on both SS & EM tables in my cab, and just to rule out something specific to that build, I put a fresh install of VP on my laptop, and reproduced in there. So it seems like the behavior is triggered by something in the Pinscape board.
By the way, the only thing currently interfaced with the board is the optical plunger.
Here are some Dropbox links to videos that demonstrate the behavior - one shows the plunger in a virtual table, and another that shows the sensor reading in the config tool. Can you make any sense of this behavior?
Table : https://www.dropbox....G_9894.MOV?dl=0
Config Tool : https://www.dropbox....G_9895.MOV?dl=0
Thanks for any help you can provide - and for your amazing contributions here!
Posted 04 February 2017 - 09:47 PM
The issue I have right now is with NewLedTester.exe : when I launch it nothing happens
No process in the background, no window opening, nothing. I'm running Windows 10 x64 up to date.
I have no idea what's wrong - sounds like Windows is killing the process immediately when you launch it, but without any sort of error message it's hard to guess why. Try looking at properties for the .exe and the new LedWiz.DLL and see if they're marked "blocked" or "internet download" or anything like that. The program depends on having a current .Net installed, but I thought that was a given on Win 10.
However if I pull back and hold, or slowly move the plunger rod, the virtual one will 'release' on a regular interval - about every 3 seconds.
You're probably getting some instability in the sensor reading at the optical level that's making it jump around. Things look pretty stable in the setup tool video, so it's not obvious there, but if VP thinks you released the plunger, it's because it saw a momentary reading on the joystick input that looked like it had moved rapidly forward, which is probably a momentary mis-read from the sensor.
Try going to the setup tool and running the joystick viewer window. That shows the plunger position as read through the joystick interface. Pull the plunger back and hold, just like you're doing in VP, and see if you see any jumpy behavior there. You can also try the "capture" button to capture some raw data samples, which might show a jumpy reading that's not visually apparent on the viewer.
If you do see any jumpiness in the joystick viewer or capture data, try tweaking the optical setup a little to see if you can get a sharper shadow line on the image. It looks like you're using two LEDs for the light source; you might be better off with just one, since it will give you a smaller point source for the light and less fuzziness in the shadow edge. You might also try shading the blinking light from the KL25Z and anything else in the cab that's blinking, since that light will be picked up by the sensor as well, and blinking especially will cause momentary changes in the lighting that could be read as motion.
Posted 04 February 2017 - 11:32 PM
Thanks for the quick feedback. I made some adjustments, and while I was able to get a sharper shadow edge, the repetitive release kept happening. After a while, I stumbled upon the root cause - the trigger was coming from the Config Tool, which was still open in the background. I had seen that VP would go haywire while the plunger sensor screen was open, but didn't consider that there might be additional interference from the main screen of the Config Tool too. As soon as I shut that down, the 3-second release stopped occurring.
So, probably not a normal use case. But you may want to consider some sort of heads-up in the Config Tool telling people to shut it down before attempting to test a table.
Thanks again - looking forward to finishing my Pinscape build!
Posted 04 February 2017 - 11:51 PM
Thanks for the quick feedback. I made some adjustments, and while I was able to get a sharper shadow edge, the repetitive release kept happening. After a while, I stumbled upon the root cause - the trigger was coming from the Config Tool, which was still open in the background. I had seen that VP would go haywire while the plunger sensor screen was open, but didn't consider that there might be additional interference from the main screen of the Config Tool too. As soon as I shut that down, the 3-second release stopped occurring.
I should have thought of that! The config tool has that effect because it piggybacks its own special data reports from the device on top of the joystick interface. Programs like VP that think they're reading ordinary joystick input see those occasional special data packets and think they're ordinary joystick data with weird values.
Glad you tracked it down. I'll try to find a suitable place for a warning about it.
Posted 05 February 2017 - 09:12 AM
The issue I have right now is with NewLedTester.exe : when I launch it nothing happens
No process in the background, no window opening, nothing. I'm running Windows 10 x64 up to date.
I have no idea what's wrong - sounds like Windows is killing the process immediately when you launch it, but without any sort of error message it's hard to guess why. Try looking at properties for the .exe and the new LedWiz.DLL and see if they're marked "blocked" or "internet download" or anything like that. The program depends on having a current .Net installed, but I thought that was a given on Win 10.
Mike,
It seems the new Ledwiz.dll doesn't totally work on my computer.
- DOF : OK
- NewLedTester : KO
- PinballX : KO
- SimpleLEDTest (just to be sure) : [02/05/2017 10:07:30] Unable to load LEDWiz library: [C:\Users\PinCab\Desktop\Pinball_work\LEDBlinky\LEDBlinky\ledwiz.dll].
If I put back the previous 13 Kb Ledwiz.dll, everything works.
The DLL and the exe aren't blocked.
Thank you ![]()
++
Emulation Station →
Virtual Pinball Cabinets →
4 Channel Power MOSFET Output ModuleStarted by rickh , 20 Nov 2025 |
|
||
Visual Pinball →
Visual Pinball →
Linux - Pinscape support in VPX ?Started by Tesla , 03 Nov 2025 |
|
||
Emulation Station →
Virtual Pinball Cabinets →
DOF not recognising more than 32 toysStarted by fichtenelch , 02 Sep 2024 |
|
||
Emulation Station →
Virtual Pinball Cabinets →
Pinscape RGB LED Wiring Help!Started by Vegasdodger , 05 Jun 2024 |
|
||
Emulation Station →
Virtual Pinball Cabinets →
'Little Bang' A New Open-Source Virtual Pinball InterfaceStarted by rickh , 28 Apr 2024 |
|