975 replies to this topic

[MikePinball]

What am I missing?

 

Night mode configuration for the button? See screenshot below:

 

[Polyphemus]

 

What am I missing?

 

Night mode configuration for the button? See screenshot below:

 

 

 

Sadly, no. It has #25 and the button shows the 'moon' icon next to it...

[JLPicard001]

Hey guys I just recently thought I'd add MJR's PinVol to the pinscape mix.  I have a KL25Z in my cabinet running updated firmware and it's been configured for awhile now to interface with my cabinet buttons just fine.  I made the switch over to PinballY awhile back and am very happy with the features, love the easy media add methods.  

PinVol however, although it seems to work just fine with my keyboard I can't seem to get my KL25Z configured to include a SHIFT button.  IT APPEARS as if I do, I have

 

PTB11 (Joystick 11) configured as my SHIFT showing the upwards green arrow in Pinscape

PTB9 (Joystick 13) configured as my up arrow in PInscape

PTB8 (Joystick 14 configured as my down arrow in PInscape

11 is my normal Exit, 13 is my Ex. Ball, 14 is my Coin button(s)

 

When I use the button tester in Pinscape, the held down 11 followed by either of the 13 or 14 shows as a shifted up arrow, but won't react that way to change the volume on a table, I have to reach for my keyboard.  It's the Pinscape designated SHIFT that isn't working as I can hold down SHIFT on my keyboard and use the shift designated Joystick 13 or 14 (extraball and Coin) and change the table volume.

[mjr]

Hey guys I just recently thought I'd add MJR's PinVol to the pinscape mix.  I have a KL25Z in my cabinet running updated firmware and it's been configured for awhile now to interface with my cabinet buttons just fine.  I made the switch over to PinballY awhile back and am very happy with the features, love the easy media add methods.  

PinVol however, although it seems to work just fine with my keyboard I can't seem to get my KL25Z configured to include a SHIFT button.  IT APPEARS as if I do, I have

 

PTB11 (Joystick 11) configured as my SHIFT showing the upwards green arrow in Pinscape

PTB9 (Joystick 13) configured as my up arrow in PInscape

PTB8 (Joystick 14 configured as my down arrow in PInscape

11 is my normal Exit, 13 is my Ex. Ball, 14 is my Coin button(s)

 

When I use the button tester in Pinscape, the held down 11 followed by either of the 13 or 14 shows as a shifted up arrow, but won't react that way to change the volume on a table, I have to reach for my keyboard.  It's the Pinscape designated SHIFT that isn't working as I can hold down SHIFT on my keyboard and use the shift designated Joystick 13 or 14 (extraball and Coin) and change the table volume.

 

Could you post a couple of screen shots?

 

- Your button mapping list in the config tool

 

- Your PinVol config window

 

I think I see what you're trying to do, but I'm not completely sure.

 

You've read through the little blurb in the config tool on what the Shift Button does, right?  I just want to make sure you noticed the weirdness about how "Shift Button" and "Shift Key" are two completely different and unrelated things - the terminology is really confusing (but I haven't come up with anything better).

[JLPicard001]

I got it working, sorta.  I changed which button I used for the Shifter, taking your suggestion to use the Extra Ball and then made my flippers the UP and DOWN.  I also realized reading the above that the SHIFTER and the SHIFT key were indeed different things, so going back to my PinVol config I didn't use the keyboard, just the buttons on my Cabinet.  The SHIFTER (now my Extra Ball) didn't register, but the flipper button pushes registered as UP and DOWN.  Makes sense now since the SHIFTER isn't really any key at all, just ahhh SHIFTER......

Here's the snaps of my configs as requested:

 

The odd thing is that now I can get this to work however the EXTRA BALL button and the COIN button register all by their-selves as a volume changer.  I think they might be mapped as up and down arrow pushes somewhere maybe in the key mapping software I also have running for FX3 games JoytoKey.

UPDATE:  I just went and checked and yep that was it, JoytoKey EXTRA BALL and COIN were mapped as up and down arrows so I could navigate around FX3 menus.  Once I unmapped those two they no longer worked as VOL keys.  Now I have to figure out how to have these co-exist.  Hell I need more buttons on my cab anyway, never did install any magnasave buttons maybe I'll put a few more buttons on the cab.  Thanks MJ!

[mjr]

Glad you got it worked out.

 

One note, in case you're not aware: you can also map Pinscape buttons to keyboard keys if you want.  Unless there's some separate reason you want to be mapping them as joystick buttons, you can cut out the middleman with FX3 and just use whatever key mappings you need.  When you click a button to set its mapping in the config tool, click the little keyboard icon to switch it to keyboard mode.  Then you can select whatever keyboard key you want to assign.

[JLPicard001]

Hmm. Switch over to everything keyboard instead of joystick inputs to pinscape. Ima try that.

[JLPicard001]

Ok well on my way to this switch over!  Got Pinscape/PinballY/PinVol/FX3/PinballX/Pinball9.9's sorta kinda set up and keys coordinated.  Hell this even fixed my VPX plunger problems.  Still have a couple issues with the keyboard ESC key working the way I need it to and can't move around in FX3 menus without the keyboard but I haven't managed to hopelessly break anything yet, so that's progress!

[Ciceronic]

Hi. Sad times, my Pinscape refuses to cooperate.

 

First thing happened: As soon as I power on the cabinet a flipper solenoid pulls. At power off, it releases.

Tested solenoid separate, no problem.

 

Now I've resoldered and searching for days for shortcuts. Found nothing.

So I decided to strip things down, measuring directly at KL25Z outputs, nothing connected.

 

Without even activating any ports, I'm measuring about -3,3V ??? at each port. Minus WTF?

Something is really wrong.

 

Could it be ANYTHING software related (like a plus/minus operating level switch), or is it time to buy a new KL25Z board?

[mjr]

Ciceronic - some questions:

 

1. I take it this was a previously working setup, right?  In other words, something broke in a working setup.  Just making sure we're not talking about getting something new working.

 

2. Does the KL25Z still connect to USB, and does the config tool see it?

 

3. Do you have any other solenoids or feedback devices attached to the KL25Z, and if so, are any of them still working?

 

4. What are you using as booster circuits to drive the solenoids?

 

If the answer to #2 is "yes and yes", the KL25Z is probably okay, and if #3 is "yes and yes", that would make that even more likely.  I'd say the problem is probably just in the booster circuit.  If that's the case, my first guess would be that the transistor or relay that's driving the solenoid has gotten fried.  One of the common failure modes for transistor switches seems to be to short across the junction, so that it's permanently stuck in the "ON" state.  If that's the case, you should just need to replace the MOSFET or whatever you're using to drive the solenoid.

 

It still could be the KL25Z that's the problem, though, because it could be that the GPIO that's controlling the solenoid is fried the same way as I just described for the MOSFET.  I think this possibility is a bit less likely, because it's kind of hard to kill a single KL25Z GPIO port in isolation.  The GPIO ports are physically on the same silicon as the rest of the CPU, so anything that can kill a port will probably fry the whole chip.  So if the CPU is working at all (for example, connecting to USB), there's a good chance your KL25Z is still OK overall.

[Ciceronic]

Hi,

1. Yes it was a working setup. For long time.
2. Yes, everything seems ok there. I can see gyro readings.

3,4 and other:
I have burnt a MOSFET before, error was exactly what you described. Fully open in "ON". Curcuits homemade according to your guide. PC817+NPN MOSFET.
So knowing this, I built a new "MOSFET-module" and just replaced. Some other things were working, in other "MOSFET-module".
New module totally dead. I got tired and thought I'll follow the curcuit from KL25Z to find the problem.
So I'm measuring directly at the KL25Z pins with nothing connected and I still don't get anything I expect. GND >> Output pin should read +3,3V, I get -3,3V or 0V.

Also made a full firmware re-installation (2019-03-05), Same result. This is so strange.
I'll order a new board, but no feedback during the holidays. Crap!
 

[mjr]

> 2. Yes, everything seems ok there. I can see gyro readings.

 

That's a good sign!  You *might* still have a dead GPIO port, but at least the rest of the CPU is still be working.

 

 

> So I'm measuring directly at the KL25Z pins with nothing connected and I

> still don't get anything I expect. GND >> Output pin should read +3,3V, I

> get -3,3V or 0V.

 

Hmm, that is weird.  Are you absolutely sure you don't just have the voltmeter leads reversed?   -3.3V is what I'd expect to see with the red lead on GND and the black lead on the GPIO pin, and the port on/high.

 

What happens if you connect the solenoid to the booster circuit, with everything wired up EXCEPT the connection between the KL25Z GPIO pin and the booster input?  That would be a good isolation test to see if it's on the KL25Z side or on the booster circuit side.  If the solenoid is still stuck on WITHOUT the KL25Z GPIO pin input connected, it's a pretty good bet that the problem is somewhere in the booster circuit.  If it only turns on when you connect the GPIO pin, you might well have a single dead GPIO port.

 

If it does appear to be a single dead GPIO port, another thing you could try is to connect the booster input to a DIFFERENT GPIO port.  See if it's still stuck on then or if you can control it by mapping the other port and testing from the config tool output tester.  If you can successfully control it from a different GPIO port, but not from the original port, that would pretty well confirm the single-dead-port theory.

You could also do a simple test of the GPIO port directly.  You'll need a small RED LED (the little "bubble" kind that's rated for about 20mA) and a resistor - around 1K ohms.  If you have the LED specs, you can plug it into an LED calculator with these parameters to get a more exact resistance, but 1K should work okay for any small red LED - those are all about 1.2V Vf.

 

Source voltage 3.3V

Diode foward voltage <plug in your value here>

Diode forward current 3mA  <NOT the LED's actual rating - use 3mA no matter what the diode specs say>

 

Then connect it like this:

 

GPIO pin ------ resistor ------ +LED- ------ KL25Z GND pin

 

Go to the config tool output tester, and exercise the port.  The LED should turn on when the port is on and off when the port is off.  "On" might be pretty faint since we're only giving it 3mA, but it should be enough to see it light up a little bit.

[Ciceronic]

I built that LED-thing yesterday. Helped a lot to find the problem.
I'm not absolutely certain, but I might have found the problem.

While playing around with the output port tests I noticed a flickering at "Night Mode" button.
That made me find a small crack in the Night mode button soldering.
Resolder that and some other GND connections seems to have eliminated the problem.

Now I'll continue and connect the booster curcuits again.

Thanks a lot for your support!

[rickh]

I have burnt a MOSFET before, error was exactly what you described. Fully open in "ON". Curcuits homemade according to your guide. PC817+NPN MOSFET.
So knowing this, I built a new "MOSFET-module" and just replaced. Some other things were working, in other "MOSFET-module".
New module totally dead. I got tired and thought I'll follow the curcuit from KL25Z to find the problem.
So I'm measuring directly at the KL25Z pins with nothing connected and I still don't get anything I expect. GND >> Output pin should read +3,3V, I get -3,3V or 0V.
 

Ciceronic, 

Are you placing EMF (AKA Flyback) diodes across your solenoids?  I find it difficult that MOSFETs are burning up on an inductive load which would have me believe that there is no EMF diode across the coil.   Also, it is highly suggested that these diodes do not reside on your driver board, but rather on the coil itself.  This practice eliminates annoying 'pops' and possible USB lock ups due to the high energy transients in the wires when the magnetic field collapses.

[mjr]

Ciceronic - glad you tracked down the problem.  Rickh's suggestion to double-check diodes is a good one if you've been having repeated MOSFET problems.  Another thing to look into is cabinet grounding.  MOSFETs are sensitive to static electricity, which cabinet grounding will help suppress.  The build guide has sections on both subjects:

 

http://mjrnet.org/pi....php?sid=diodes

http://mjrnet.org/pi...id=cabGrounding

[rlewis19]

Hi,

 

I just have the KL25Z board to handle my table nudging and love it.  Recently I upgraded from V1 to V2 of the firmware.  I'm noticing in the config tool there is a fair amount of jitter for the accelerometer.  I was just wondering if this is normal?  To be honest I don't remember if that happened with V1 or not. The x,y of the accelerometer mostly is at 0,0 but I can see it twitching to different values.  (It's too fast for me to see what the other values are)

 

Rick

[mjr]

I just have the KL25Z board to handle my table nudging and love it.  Recently I upgraded from V1 to V2 of the firmware.  I'm noticing in the config tool there is a fair amount of jitter for the accelerometer.  I was just wondering if this is normal?  To be honest I don't remember if that happened with V1 or not. The x,y of the accelerometer mostly is at 0,0 but I can see it twitching to different values.  (It's too fast for me to see what the other values are)

 

A little bit of that is perfectly normal.  The KL25Z's accelerometer is extremely sensitive - enough to pick up little bits of random vibration, like just walking past the machine.  You'd have to install it on a really solid, well-damped foundation to get it completely still, below its threshold of detection.

 

I'm always keen for a little numerical crunching to back up hand-wavy answers, so let's try to see if I'm in the right ballpark here.  The MMA8451Q (the MEMS accelerometer chip integrated into the FRDM-KL25Z board) has 14-bit resolution at +/- 2g, which is to say that one quantization step equals 1/4096 of a g ("g" stands for the acceleration due to Earth's gravity).   So its threshold for detection, modulo noise, is 1/4096 of a "g".  But what does that mean for force detection?  I for one don't have a very good intuitive sense of this, but let's see if we can translate to something everday-ish.   A "g" is a unit of acceleration, or equivalently force-per-unit-mass, which we can express in Newtons per kilogram.  That unit means nothing to my benighted Imperial Units intuition.  Pound-force (lbf) is probably the closest thing that I'd find intuitive, so let's do the conversion.  1/4096 g = 9.8m/s^2 / 4096 = 0.0024 m/s^2 = .0024 N/kg (Newtons per kilogram).  If we say a virtual cab weighs in at about 300lb = 136kg, then to exert .0024 N/kg on this mass, we'd have to exert 0.32 N, which translates to .072 pound-force, or 1.15 ounce-force.  So if I'm doing this right, exerting a force on the cabinet that feels like the weight equivalent of about an ounce would register.  That fits pretty well with my sense of it - my sense is that the slightest breeze is enough to set this thing off.

 

I know this "slightest breeze" realization makes a lot of people go into a panic about all of the random stray vibrations you get from solenoids, subwoofers, and trucks rolling by in the next county.  But this extreme sensitivity is a good thing!  The trick is to remember that this device is not only extremely sensitive, but also extremely linear.  Yes, it picks up those small vibrations, but it reports them as small vibrations.  It also picks up the big vibrations and reports them as big vibrations, and it picks up the medium-sized vibrations and reports them as medium-sized.  If you have things set up right in VP, VP applies the accelerometer input linearly as well, so it'll treat little inputs as little inputs and big inputs as big inputs.  Every vibration registering on the accelerometer should be handled in VP with an effect proportional to the actual vibration.  What trips people up is that they want to think about this as an all-or-nothing input, like the bad old way where you press the space bar to input a single, fixed quantum of nudge into VP.  It's not like that, because the accelerometer sends all of these different size nudges in proportion to the physical force it registers.  

[mjr]

I just posted updates to the Config Tool and firmware.  As usual, they're a la carte, so you can update either or both as desired, and as always you can just stick with your current setup if you don't want to risk destabilizing anything.  (The past versions of the firmware are on the site as well, so you can always back out changes if you do ever run into a problem after an update.)

 

The main change in the Config Tool is that it now includes direct support for the Oak Micros All-In-One board, so you no longer have to fake it with the regular Expansion Board options.  It also has support for a couple of new plunger sensor types that I have in the works, although they're still works in progress, so they're not quite ready to use yet.  

 

The firmware has support for the new sensor types as well, as well as some improvements for people using the potentiometer sensor.  The way that it reads the analog voltage input on the sensor has been changed for better accuracy and less noise.  If you've been using the "Jitter Filter" to deal with shaky readings, you might want to go back and fine-tune that again after updating - you should be able to use a much smaller jitter window, or even eliminate the filtering entirely.  I've seen rather dramatic improvements in it on my test rig.

As always, you can find the downloads at:

 

http://mjrnet.org/pi.../swversions.php

[MikePinball]

The main change in the Config Tool is that it now includes direct support for the Oak Micros All-In-One board, so you no longer have to fake it with the regular Expansion Board options.  It also has support for a couple of new plunger sensor types that I have in the works, although they're still works in progress, so they're not quite ready to use yet. 

Thank you for accepting my code changes into the Config Tool and making it available. This is a nice usability improvement for existing and new customers. I will be posting an update procedure for my existing customers. It is quite straightforward but requires you to edit a file with notepad. This modification is only necessary if you want to see the configuration represented as a Pinscape All-in-One versus the traditional Pinscape expansion boards.

 

Here is a screen shot comparing before and after. This picture doesn't tell the whole story because other changes permeate through the application such as help text and pinout diagrams to show the exact I/O pin placement on the Pinscape AIO board.

 

 

Little known fact: you can add regular Pinscape MOSFET power boards or Chime boards by connecting to the builtin expansion headers just like you would for the regular Pinscape expansion boards.

Edited by MikePinball, 23 January 2020 - 04:56 AM.

[cenutrio]

mjr congrats for the new proyect the fronted pinball y  looks nice, awesome proyect ,vey configurable and build in open source

next week i try to install, i hope are intuitibve instalation, the section pinbally help are very completed, but my english is like a chimpancee.

thanks again for share without economic profit  the hundred of hours of work with the pinscape software,resolving thequestions in the forums .

the guide its a big add and have many tricks for make the body of the cabinet i make my cabinet autodidact and reading  (but my english are bad)some post and the outcame are a cabinet desastrous  with many mistakes and many hours  reparing fudges