- This topic has 25 replies, 8 voices, and was last updated 5 years, 2 months ago by
.
-
Posts
-
Yesterday I played Jolly Park few games until I noticed burning smell. Switched off the system and quickly found it was back left solenoid in my 10 solenoid setup that had stuck on and melted. Apparently it had caught on fire as the wood near it was burned too. Pretty worrying sight.
I remember reading something about DOF monitoring that no solenoid is stuck on, but couldn’t find that discussion with google anymore. Sorry if there’s a discussion about this already and please share a link if there is.
Can anyone tell if DOF has such safety mechanism or not? If it doesn’t I really suggest adding one as this is quite of a fire hazard.I use Pinscape and mosfet board to drive my 12V solenoids so thought that one option to protect my ass is to set all solenoid outputs to flipper logic mode and set the minimum power for them after a delay. Another option is to build a separate monitoring circuit to check if mosfets (other than flippers) are outputting something for longer than a second or two.
————————————————————————
Update / summary
- There might be bugs (or weird features) in tables the cause the solenoid output to stay on. Jolly Park seems to have one.
- Avoid burning solenoids with some sort of HW output monitoring or use devices that don’t care if they’re left on (industrial contactors).
- I used what I had; Arduino Mega (Mega cause needed 10 analog inputs) to monitor all 10 solenoids individually by measuring the drain voltage with a simple voltage divider circuit and Arduino analog inputs. If voltage of any other than flipper solenoids stays under set limit (I have 12V solenoids so set the limit to 11V so that solenoids won’t stay on even with a mild PWM signal) for too long (I set the time limit to 5s for now), the system cuts 12VDC supply to my mosfet board with a relay and must be manually reset.
You need to login in order to like this post: click here
The power board does not protect the coils from burning!
there are 3 options available:
- Use fuses ;) using them is ALWAYS a good idea
- build or buy a Pinscape chime board. They have a hardware protection timer installed so the coils will automatically shut off after 2 seconds of continuous use.
- Set “flipper logic” to your solenoid outputs in the Pinscape config tool. Define a lower “current” (in fact it’s pwm modulation) after eg. 100ms
if you need further assistance feel free to ask!
and btw:
youre lucky using a Pinscape as it has a electrical separation to your computer asa lack of that might have blown up your Mobo!
greetz,
Dom
You need to login in order to like this post: click here
Also what kind of solenoids are you using? If your using those cheap car starter solenoids then that’s going to happen! FYI real solenoids in Pinball’s stick and the same thing can happen
********************************************************
Messing with the VPinball app and push notifications.
So if you haven't downloaded app yet what are you waiting for!?
for IOS and Android********************************************************
You need to login in order to like this post: click here
Thanks for replies but I guess there was some misunderstandings:
I dont use the Pinscape power board. I use other mosfet driver boards and I didn’t at any point say I assumed they protect me. Question was more of the software monitoring the output controls for bugs.
- Fuses wouldn’t protect me as nothing failed on HW side as such. Software was controlling the solenoid on and it used 1-2A it needs but that @ 12V is enough to melt everything.
- I know chime board would help, but as stated I wonder why this backup timer function cant be inside the software? It would be a really easy addition to DOF / Pinscape to have ability to set the max time and have an alert if that exceeds.
- Like mentioned I know one way to avoid hazardous overheating is to use flipper logic. That also makes my coils whine nicely so I can easily hear if something is stuck.
I’m using cheap chinese open frame solenoids mounted on a aluminum bracket for some extra cooling. I’ve run those for several months without a problem. It was the Jolly Park table that caused this. Can’t say if it’s the table script, ROM or DOF that causes solenoid to get stuck.
My plan is to change all solenoid outputs to flipper logic and I’ll also add a simple feedback loop circuit to monitor if any any solenoid output gets stuck.
You need to login in order to like this post: click here
It’s very possible the solenoid simply failed. Your right a fuse would not have helped if that happened.
********************************************************
Messing with the VPinball app and push notifications.
So if you haven't downloaded app yet what are you waiting for!?
for IOS and Android********************************************************
You need to login in order to like this post: click here
It’s not a flipper solenoid so it must have had the control stuck on for a long time to end up like that as pulsing solenoids dont normally heat much. It had melted the inner plastic casing so that the shaft was stuck so it couldn’t be just a wiring short on the surface either. I took it apart and coil itself didn’t look burned.
You need to login in order to like this post: click here
Does the table have a controller stop sub?
********************************************************
Messing with the VPinball app and push notifications.
So if you haven't downloaded app yet what are you waiting for!?
for IOS and Android********************************************************
You need to login in order to like this post: click here
Jolly park :
Sub table1_exit:Controller.stop:End Sub
So, yes it does. And all DOFOn has a corresponding DOFOff.
You need to login in order to like this post: click here
Oh poo… KL25Z has only 10 PWM channels + 3 for onboard LED, and 2 of those need soldering from e.g. LED legs so that’s kind of a let down for the original workaround idea for my 10 solenoid setup as I’d really would want to keep my cabinet buttons pulsing softly (need at least 4 PWM’s for those). Guess I just need to wait and order some TLC5940’s and start expanding this.
I noticed that controlling IRF540N mosfet was killed in action too (stuck in closed state to help with the bonfire). Of course there’s the possibility it was faulty mosfet causing this all, but these are rated to 33A@100V so I doubt it. My super duper solenoids draw ~1.7A each. I also have 1N4005-1N4007 diodes on each solenoid to protect the drivers. Anyhow, replaced that mosfet with what I had, FQP30N06L (32A@60V).
You need to login in order to like this post: click here
For now, while waiting for spare parts I just made a simple safety circuit with Arduino and a relay board:
If any of solenoid outputs (other than flippers) stay on over 2s it just cuts the whole 12V supply to mosfet boards.I did add another smaller fuse before mosfet boards too as in this case it would have maybe saved the fet.
I’ll continue testing the same table to see if I get that back left solenoid stuck again.
You need to login in order to like this post: click here
and btw:
youre lucky using a Pinscape as it has a electrical separation to your computer asa lack of that might have blown up your Mobo!
Forgot to answer this.
Didn’t quite get how burned solenoid relates to risk of blowing mobo? Overloaded solenoid PSU yes, burned fuses yes, broken fets yes but why u mentioned mobo? That would require my mosfet board to fail in such way it connects +12VDC to logic level circuit -> KL25Z -> USB 5v.You need to login in order to like this post: click here
Well, other controllers don’t have that electrical separation through optocouplers.
I’ve seen controllers burning up Mobos because the owner wired some toy wrong and directly grilled the controller and as he had the shaker in that line as well, the 12V from that went directly through the usb wire and grilled everything.
wont happen with Pinscape and expansion boards ;)
You need to login in order to like this post: click here
I understand but nothing in this topic was related to a wiring fault or conducting 12V to USB line so was a bit confused about this comment.
The whole point is simply that if the software has a bug and it keeps solenoid output on for too long, the solenoid will burn no matter how many fuses or opto couplers there are.
You need to login in order to like this post: click here
I use Siemens contactors for my DOF setup, these can be left on with no issue. They are designed to be left engergized for weeks or months continuously
You need to login in order to like this post: click here
@sniiki the software does not have a “bug” Ive been running feedback well on 6 years and never smoked anything. BUT lets say you exit wrong or an error in table script i suppose that could leave a output stuck on in some weird state but really would be more of a problem with solenoids you are using or wiring issue but your right a solenoid should never stick on normally.
@arngrim could anything be wrong with this table to cause a stuck solenoid? I have not tried the table so cant report any issues with it********************************************************
Messing with the VPinball app and push notifications.
So if you haven't downloaded app yet what are you waiting for!?
for IOS and Android********************************************************
You need to login in order to like this post: click here
I noticed that controlling IRF540N mosfet was killed in action too
It could very well be that a failed mosfet is what CAUSED the problem, as opposed to being harmed by the issue.
You need to login in order to like this post: click here
@sniiki the software does not have a “bug” Ive been running feedback well on 6 years and never smoked anything. BUT lets say you exit wrong or an error in table script i suppose that could leave a output stuck on in some weird state but really would be more of a problem with solenoids you are using or wiring issue but your right a solenoid should never stick on normally.
@arngrim could anything be wrong with this table to cause a stuck solenoid? I have not tried the table so cant report any issues with it
Good to hear. Please don’t take me wrong, I don’t mean to blame software or point fingers. I just wanted to rise this question why couldn’t we add such simple parameter to control software just in case. In the end monitoring needs to be at HW level to make sure it’s safe but SW monitoring wouldn’t hurt either.
You need to login in order to like this post: click here
@sniiki the software does not have a “bug” Ive been running feedback well on 6 years and never smoked anything. BUT lets say you exit wrong or an error in table script i suppose that could leave a output stuck on in some weird state but really would be more of a problem with solenoids you are using or wiring issue but your right a solenoid should never stick on normally.
@arngrim could anything be wrong with this table to cause a stuck solenoid? I have not tried the table so cant report any issues with it
Good to hear. Please don’t take me wrong, I don’t mean to blame software or point fingers. I just wanted to rise this question why couldn’t we add such simple parameter to control software just in case. In the end monitoring needs to be at HW level to make sure it’s safe but SW monitoring wouldn’t hurt either.
sniiki wrote:
I noticed that controlling IRF540N mosfet was killed in action tooIt could very well be that a failed mosfet is what CAUSED the problem, as opposed to being harmed by the issue.
Yeah it could. But I’d think it’s more likely that it got burned in the process when the solenoid got so hot it melted the wires, shorting it and causing shit load of current to rush through. Why my 12VDC supply overload protection didn’t kick in? Dunno, cause it’s bad quality I think. I have a new Meanwell PSU on the way anyway for other reasons.
You need to login in order to like this post: click here
@sniiki the software does not have a “bug” Ive been running feedback well on 6 years and never smoked anything. BUT lets say you exit wrong or an error in table script i suppose that could leave a output stuck on in some weird state but really would be more of a problem with solenoids you are using or wiring issue but your right a solenoid should never stick on normally.
@arngrim could anything be wrong with this table to cause a stuck solenoid? I have not tried the table so cant report any issues with it
Good to hear. Please don’t take me wrong, I don’t mean to blame software or point fingers. I just wanted to rise this question why couldn’t we add such simple parameter to control software just in case. In the end monitoring needs to be at HW level to make sure it’s safe but SW monitoring wouldn’t hurt either.
sniiki wrote:
I noticed that controlling IRF540N mosfet was killed in action tooIt could very well be that a failed mosfet is what CAUSED the problem, as opposed to being harmed by the issue.
Yeah it could. But I’d think it’s more likely that it got burned in the process when the solenoid got so hot it melted the wires, shorting it and causing shit load of current to rush through. Why my 12VDC supply overload protection didn’t kick in? Dunno, cause it’s bad quality I think. I have a new Meanwell PSU on the way anyway for other reasons.
I use Siemens contactors for my DOF setup, these can be left on with no issue. They are designed to be left engergized for weeks or months continuously
Sure they would help, but I’d like to avoid them for few reasons like that I can’t ‘adjust’ the tone of the sound like I can with solenoids, are quite big for my heightwise slim cabinet and they need another 24VDC PSU or step up converters.
You need to login in order to like this post: click here
Most likely the diode failed. About 5% of the time diodes fail to a ‘blown’ or open state, 95% of the time they fail in the ‘short’ or closed state. When this happens no software monitor will change a thing since the diode is actually shorting the circuit and it takes the MosFet along for the ride. Unless your software is controlling the source feed of the voltage to the coil, it will just keep drawing power until something in the load on the short circuit gives up its smoke.
I know so little about everything that I could tell you all there is to know about nothing
You need to login in order to like this post: click here
Forums are currently locked.
