forum.hobbycomponents.com

Just to confirm (and summerise) by no change in behavior I assume this still means the following...

1) There is no activity from any of the cube LEDs.

2) The 4x 10mm base LED's do not flash then illuminate after about 15 seconds have past from applying power.

3) When touching certain pads you can make the base LEDs change state.

4) The voltage on Pins 7 & 20 of IC U2 is ~ 5V

5) The voltage on pin is ~ 0.75V and on pin 10 ~ 0.5V


If that's still the case then 1 to 4 suggest that it's likely the microcontroller isn't running for some reason. Number 5 suggests that the microcontroller is at least trying to start up and so isn't completely dead. This suggests the problem is related to the crystal or one of the 22pF capacitors.

The only things I can suggest is just rechecking all the pins on IC U2 again by using your meter to check continuity between the PADs on one side of the PCB to the pins of the IC on the other. Also can you recheck the voltage on pins 7 and 20 again by this time by doing the following:

Put the -ve probe of you meter on pin 8 and the +ve on pin 7.

Put the -ve probe on pin 22 and the +ve on pin 20.

In both cases you should measure ~ 5V

I wouldn't bother removing the 22pF capacitors, just check that the soldering is ok and there is no physical damage to them.

If you still can't find anything wrong after checking the above I think we'll have to assume there is something more fundamentally wrong with one of those parts and take it from there.

Hi there, before I start I need to say my cube is working and looks really good, I've really enjoyed building the kit (although I wouldn't race to do another one anytime soon!).

Just wanted to say that I have seen what sounds like the exact same problem as seen by Ramshackle, namely:

1. I was only getting snippets of the demo, i.e. the cube would run for a few seconds then stop completely
2. Seemingly not a power problem since lower 'display' leds always on.
3. When I touched the cube with my hands, that seemed to jolt it into action for a few seconds (sometimes this wouldn't work).
4. When I touched it with a soldering iron, that would reliably start it for a few seconds again, sometimes longer. Tapping around the structure with the soldering iron would repeatedly work in seeing snippets of the routine.
5. The routine would seem to fail at the same points every time, i.e. at the same point in the routine. It would do some effexts better than other, some not at all, but never function for more than 10 seconds, and more often 4-5 seconds.

My first thought was a loose connection or bad solder (I will admit to not being the neatest with the soldering iron) so I went back over the board making sure there were no dry joints or ball joints.

When that didn't work it seemed (and still seems) like there must be a grounding problem (hence working when tapping with a soldering iron or anything else conductive). The fact the routine failed at the same points makes me think there must be one grounding 'path' which isn't working.

Then I hit the forums and saw ramshackle had seen the same, and realised I also could get the board running through the whole sequence by pressing down on pins 15 and 16. I had a spare 104 as well and soldering across those pins has everything working fine - the board runs the whole routine reliably, every time. I thought it was perhaps running a little fast but after watching the youtube video of the demo I don't think that's true - it seems to match the speed pretty well.

I did test the continuity across the pins in U2 but everything seemed ok.

I'm happy to leave things as working now with the 104 fix but wondered your opinion on whether or not this was safe? Presumably if there is a problem with one (or several) grounding paths, others are available and there's no real problem?

Thx.

@bluethunder22

I think the comments I made in repose to Ramshackle's last post are also relevant to your issue. Adding the 100nF capacitors won't hurt anything but I think it's likely that although it's making the cube work, it's probably just making the actual issue rather than fixing it.

If you want to investigate the problem further you may want to try what I suggested to IBI above. Although different symptoms I still think in both cases it's a bad connection somewhere, most likely a ground connection. So measuring across the GND and 5V pins in IC U2 would be the first place I would suggest.


I may add a list of known causes to the first post but every confirmed problem to date has been caused by one of the following (in order of most common)

1) Pins of IC sockets bending under the socket and so not getting inserted into the through-hole pad. This results in either an intermittent or no connection with he pad. It's also very difficult to spot visually.

2) A bad solder connection either caused by a dry joint or damage to the pad/track from previously soldering a re-soldering a component.

3) A bad batch of M54562P ICs in a few kits sold late last year (see first post). The symptom of this is random layers not illuminating.

4) And finally In one case a bad 5V adaptor which was over-volting the ICs which the DM13A IC's are particularly susceptible to.

Is there an easy way to dim the leds a bit? Just replace some resistors?

The brightness of (or rather current though) the LEDs is controlled by the four 1K resistors R8, R9, R10, & R11. Increasing the values of these resistors will reduce the brightness. For example, changing them to 2K will half the brightness.

Quick question for those out there more enlightened than I.

Has anyone ever tried to have a completed 8x8x8 cube respond to a music audio source? By sampling the frequency range, amplitude or beat, that kind of thing.

Can it be done at all? Or is it a case of I wouldn't start from here?

Many thanks in advance...

Hi all.

Another question for the group.

I've just noticed that the text in the default patterns library are displaying upside down? Presumably all the patterns are upside down, it's just noticeable with text.

Any ideas why?

Thanks in advance.

I've just noticed that the text in the default patterns library are displaying upside down? Presumably all the patterns are upside down, it's just noticeable with text.

Is it possible that you have the layers of your cube connected in the reverse order? Take a look at connectors L1 to L8 which run down the left side and connect power to each layer of the cube. Starting at the front of the cube nearest the push buttons the first connector (L1) should connect to the bottom layer not the top, then the next one (L2) should connect to the 2nd layer from the bottom and so on.

andrew wrote: ↑

Tue May 14, 2019 8:30 am

I've just noticed that the text in the default patterns library are displaying upside down? Presumably all the patterns are upside down, it's just noticeable with text.

Is it possible that you have the layers of your cube connected in the reverse order?

Genius Andrew! I do indeed... many thanks.

Just in case anyone is interested, I've written up what I did with this cube on Thingiverse (videos also there): https://www.thingiverse.com/thing:3696884

Came out very nice!