IceTube Clock acting weird

[MGaddict]

Okay, so I built this kit to the best specs I could. My soldering skills aren't great but all the points look good. I've been looking thru the forums and several different people seem to have this same problem for different reasons. Here is a quick vid of whats going on. https://www.youtube.com/watch?v=EzdgsftEkow As you can see, if I connect the top pin to the right (looking at the side of the clock) to the ground, I get the correct display. I've checked there is good continuity from the wire exiting the IceTube thru that pin and to the solder point under the PCB. One person suggested a different Q3 transistor. I don't have a different one to try. Some other people tried cleaning the PCB. Yep, just finished doing that. No help. I know software, I'm just getting into hardware because it seems like a cool hobby. Still love electronics, but I'm getting bored of just code. So I really don't know what to check. Any suggestions would be appreciated.

[adafruit_support_rick]

Can you post some clear, detailed pictures of both sides of the main circuit board and of the tube board?

[adafruit_support_bill]

Probably not Q3. Sounds more like a battery-sense circuit problem. Post photos and also measure the voltage on pin 13 of the Atmega chip.

[MGaddict]

Here are the best shots I could get. The best I got is my cell phone for a camera, sorry. I'm almost wondering if I over soldered the parts.

[MGaddict]

Probably not Q3. Sounds more like a battery-sense circuit problem. Post photos and also measure the voltage on pin 13 of the Atmega chip.

I've got a solid 1.2 volts.

[adafruit_support_bill]

1.2v is marginal. Is the display steady if you jumper that to 5v?

[MGaddict]

1.2v is marginal. Is the display steady if you jumper that to 5v?

Sorry, I know I may sound stupid asking this, but how, and where from?

[adafruit_support_bill]

Much like you did with the bias pin on the tube. Touch one probe to pin 13 of the processor and touch the other one to 5v (the rightmost pin of the voltage regulator is a good place to find 5v).

The pin with the dot next to it is pin 1. Pin 13 will be the second-to-last pin on that same side of the chip.

[MGaddict]

That didn't change anything. That was my first instinct, I just didn't want to short something.

[MGaddict]

May help, since posting my original message, I've come to notice if the capacitors are fully drained before I plug it in, the display will give me the majority of the digits for a few seconds.

[adafruit_support_bill]

Hmmm. What voltage do you measure on pin 5?

[MGaddict]

Hmmm. What voltage do you measure on pin 5?

0.0013 That seems really low to me.

[adafruit_support_bill]

No, that's right. It controls the gate of Q3 to turn on the bias voltage to the tube. Since Q3 is a P-channel device, you need to pull the gate low to turn it on.

One thing that sounds strange:

if I connect the top pin to the right (looking at the side of the clock) to the ground, I get the correct display.

If that is the bias pin, it should have some voltage on it to turn the tube on. Grounding would normally turn the tube off.
The bias pin is fed from R3. What voltage do you measure on each end of R3?

[MGaddict]

What voltage do you measure on each end of R3?

2.02 and 3.06 and of course as I check both of those, the display fully comes on.

[adafruit_support_bill]

2.02 and 3.06

That is on the low-side. It should be closer to 4.7 on the end nearest Q3.

and of course as I check both of those, the display fully comes on.

That sounds almost like an intermittent connection somewhere. Can you get a clearer shot of the soldering in the vicinity of Q3 and R3?