0

Ice tube clock... probably broken
Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.

Ice tube clock... probably broken

by IandaFox32 on Sat May 05, 2018 1:21 pm

Hello all,

So recently, I accidentally knocked my ice tube clock I built in 2012 off of my night stand while cleaning. It stopped working, so my immediate reaction was "oh, I broke the tube", so I ordered a replacement tube from the Ukraine, it finally arrived, I soldered it all together, and still nothing.

At this point I decided to do the smart thing and actually do a proper diagnosis. What I know is that:
  • Without either chip and without the tube in, the DC input voltage is 9v and the regulator output is 5v.
  • With the microcontroller in but the VFD driver (Maxim) chip and the tube out the speaker beeps on power up, the DC input voltage is 9v, the regulator output is 5v, and the output of the boost converter is 57v.
  • With both chips in but tube out, the speaker still beeps on power up, the DC input drops to 1.2v and the output of the voltage regulator goes to 0.05v, the boost output is 0.6v, and fuse F1 gets very, very hot.
[*]With both chips and tube in, the speaker still beeps on power up, the same thing as above happens
[/list]

This leads me to believe that the VFD chip (or its mount) is somehow broken, given that the troubleshooting guide lists the boost output as something around 14-16v with the VFD chip in. Seems odd that it would break on dropping it though. I've tried re-seating it multiple times to no avail.
Either that or I broke the first tube and got a bad second tube... which isn't out of the question, I'm not quite sure how I could test that though.
IandaFox32
 
Posts: 7
Joined: Fri Jul 20, 2012 6:20 pm

Re: Ice tube clock... probably broken

by adafruit_support_bill on Sat May 05, 2018 2:02 pm

With both chips in but tube out, the speaker still beeps on power up, the DC input drops to 1.2v and the output of the voltage regulator goes to 0.05v, the boost output is 0.6v, and fuse F1 gets very, very hot.

That would seem to indicate that the tube is not the problem.

I would not expect dropping the clock to 'break' the VFD chip. But the diagnostic evidence so far seems to point to a problem in or downstream of that chip. I'd check the board carefully for possible cracks, bent pins or something that may be causing a short circuit - particularly in the area around the VFD & tube header.

adafruit_support_bill
 
Posts: 69809
Joined: Sat Feb 07, 2009 10:11 am

Re: Ice tube clock... probably broken

by IandaFox32 on Sat May 05, 2018 3:47 pm

adafruit_support_bill wrote:
With both chips in but tube out, the speaker still beeps on power up, the DC input drops to 1.2v and the output of the voltage regulator goes to 0.05v, the boost output is 0.6v, and fuse F1 gets very, very hot.

That would seem to indicate that the tube is not the problem.

I would not expect dropping the clock to 'break' the VFD chip. But the diagnostic evidence so far seems to point to a problem in or downstream of that chip. I'd check the board carefully for possible cracks, bent pins or something that may be causing a short circuit - particularly in the area around the VFD & tube header.


So careful probing of the pins on the VFD socket without the VFD chip in shows that all pins are correctly connected to their corresponding spots with no shorts to anything else
Inserting the VFD chip into the socket shows that pins 1, 26, 27, and 28 are shorted to ground (pin 14)
these pins are also shorted on the VFD chip itself (aka: removed from socket).

I honestly have no idea how dropping it could break the VFD chip without breaking anything else... but as far as I can tell thats what happened... the clock was on when the drop occurred, so perhaps the old tube did break by shorting two contacts, which caused the VFD driver chip to break? seems unlikely...
IandaFox32
 
Posts: 7
Joined: Fri Jul 20, 2012 6:20 pm

Re: Ice tube clock... probably broken

by adafruit_support_bill on Sat May 05, 2018 4:15 pm

The tube itself contains a bunch of electrodes and grids suspended by fine wires. I could see how dropping the tube while powered could result in things deflecting enough to cause an internal short. I don't know how the VFD chip would handle that.

Most devices tend to 'fail open'. But I have seen some that 'fail closed' - effectively resulting in a short circuit internal to the chip. That usually results in the chip getting very hot wen powered up.

adafruit_support_bill
 
Posts: 69809
Joined: Sat Feb 07, 2009 10:11 am

Re: Ice tube clock... probably broken

by IandaFox32 on Sat May 05, 2018 4:29 pm

adafruit_support_bill wrote:The tube itself contains a bunch of electrodes and grids suspended by fine wires. I could see how dropping the tube while powered could result in things deflecting enough to cause an internal short. I don't know how the VFD chip would handle that.

Most devices tend to 'fail open'. But I have seen some that 'fail closed' - effectively resulting in a short circuit internal to the chip. That usually results in the chip getting very hot wen powered up.


Yeah, I don't see any obvious signs of deflection happening inside of the tube, but there could easily be something I can't see.
I can confirm that the chip does get warm. I don't know if I'd use hot to describe it, its definitely cooler than the fuse is.

At any rate, I'm thinking it has to be the VFD driver and maybe the tube at this point, I do have the second tube so I'll try it on that first.
I assume the ones Digi-Key sells (https://www.digikey.com/product-detail/en/maxim-integrated/MAX6921AQI/MAX6921AQI-ND/1512036) would work?
IandaFox32
 
Posts: 7
Joined: Fri Jul 20, 2012 6:20 pm

Re: Ice tube clock... probably broken

by adafruit_support_bill on Sat May 05, 2018 4:35 pm

The deflection doesn't need to be permanent to incur damage if the tube was powered at the time. An ice-tube hitting a hard floor from tabletop level is probably going to experience momentary forces upwards of 50G. That could electrodes shorting together momentarily before springing back to their original position.

Yes, that VFD chip looks like the right part.

adafruit_support_bill
 
Posts: 69809
Joined: Sat Feb 07, 2009 10:11 am

Please be positive and constructive with your questions and comments.