electronic lead screw

[GunsOfNavarone]

What caused your LED display to go south? Is there any chance the same issue damaged something else as well? I hate those INVISIBLE gremlins.

 

[kb58]

Thanks, Travis! You guys are helping to put my mind at ease. That said…

I made up a very short bit of ribbon cable, probably not even 3" and it's still no dice.
I also checked the gpio connectors on the booster, all 40 pins, for good solder joints and no shorts.
Still dead except for the LED, which I guess is a power indicator. (?)
The digits and the eight status LEDs have never so much as flickered.

The code is still driving the servo with encoder input. Correctly? Well, who knows.

What's next?

Buy another display board, they're really cheap.

 

[luxige]

Buy another display board, they're really cheap.

I did, neither board works. I tried the new one using the stock, top-mounted header to eliminate the possibility of a bad solder job on the first one.
Both boards have shown no sign of life from either the digital array or the status LEDs.
It seems like the problem is somewhere else and I don't think it's the cable.
I've thoroughly checked the cable for continuity, and replaced it with a short test cable to eliminate interference.
More info in post #379.

 

[ttabbal]

Thanks, Travis! You guys are helping to put my mind at ease. That said…

I made up a very short bit of ribbon cable, probably not even 3" and it's still no dice.
I also checked the gpio connectors on the booster, all 40 pins, for good solder joints and no shorts.
Still dead except for the LED, which I guess is a power indicator. (?)
The digits and the eight status LEDs have never so much as flickered.

The code is still driving the servo with encoder input. Correctly? Well, who knows.

What's next?

Well, that sucks.

My guess is something is off in the PCB from Clough. That one will be difficult to diagnose if you don't have an oscilloscope or logic analyzer though. You're 100% positive you don't have a pin swapped on the cable? It's easy to do, been there, done that. You can get decent little USB LAs on ebay for <$20. Well worth having around if you are going to mess with electronics.

I think the LED you are seeing is for power. So power/GND are probably good. If you have an arduino board around, you could try testing your display board with one of those.. https://blog.3d-logic.com/2015/01/10/using-a-tm1638-based-board-with-arduino/ I think it's unlikely you got two bad display boards.

 

[luxige]

Well, that sucks.

My guess is something is off in the PCB from Clough. That one will be difficult to diagnose if you don't have an oscilloscope or logic analyzer though. You're 100% positive you don't have a pin swapped on the cable? It's easy to do, been there, done that. You can get decent little USB LAs on ebay for <$20. Well worth having around if you are going to mess with electronics.

I think the LED you are seeing is for power. So power/GND are probably good. If you have an arduino board around, you could try testing your display board with one of those.. https://blog.3d-logic.com/2015/01/10/using-a-tm1638-based-board-with-arduino/ I think it's unlikely you got two bad display boards.

I heartily agree with your first and last statements!

I'm pretty sure about the cable. I hooked it up and checked continuity between header pins on the display board and booster. CLK to CLK, etc. Also, I would have to have screwed up the short test cable too.

I'll order a new booster board. Cautiously optimistic…

 

[kb58]

Well, you have a HUGE advantage: it's already a proven design that's known to work. For that reason, the problem comes down to: bad hardware, a failed sw load, and/or a defective cable. It's just a matter of compartmentalizing the issue. I don't recall now, but does Clough42's sw light or blink any LEDs on the main board? That goes a long way toward knowing that the main code is running.

 

[ttabbal]

The servo is reacting to the encoder, so I think the software is good. It wouldn't hurt to try reflashing, can't hurt. I wonder if the booster board had a transistor go bad or something..

 

[kb58]

Ah, okay, that's good to hear. So that limits the problem to the control-to-booster board cable, the booster board, booster-to-display cable, or the display. Since you swapped in a second display and that didn't fix it, it's probably not that. You checked the booster-to-display cable, so that's probably not it, leaving the booster board or control-to-booster board cable. Do a pin-to-pin check of that cable.

I'd look over his booster board with a magnifier, as it's really easy to miss a surface-mount connection problem during assembly.

One trick I use when trying to trace down problems is to physically push on the suspect board with everything powered up. This very slightly flexes it, and usually identifies any intermittent problem. Same goes for connectors; with it powered up, give them a slight wriggle.

Oh... do the buttons do anything?

A shame that you don't have a scope, but then we wouldn't be here!

 

[RJSakowski]

Thanks, Travis! You guys are helping to put my mind at ease. That said…

I made up a very short bit of ribbon cable, probably not even 3" and it's still no dice.
I also checked the gpio connectors on the booster, all 40 pins, for good solder joints and no shorts.
Still dead except for the LED, which I guess is a power indicator. (?)
The digits and the eight status LEDs have never so much as flickered.

The code is still driving the servo with encoder input. Correctly? Well, who knows.

What's next?

If the servo is tracking the encoder, you can do a rough check of the software flash. The power on state is forward, feed, inch and the rate is .005". Depending on your lead screw pitch and if all the parameters are set correctly, you should be able to rotate the spindle manually and count the turns of the spindle for one complete revolution of the lead screw. In my case, I have a 12 tpi lead screw for a pitch of .083333. One revolution of the lead screw will require 16.6667 spindle revolutions. An 8 tpi lead screw should require 25 spindle revolutions for one lead screw revolution.

If this is happening, I would guess the the flash was good.

I noticed that your Clough board has a green power on led whereas mine has a red led. My board came from his second build. Hopefully, there were no other changes besides the led color but who knows?

 

[RJSakowski]

Good point about the buttons. Although the display leds aren't lighting, the switches may be functioning.

From the power up state, five pushes of the up button will double the lead screw rpm and 11 pushes will quadruple it. From the power up state, push the feed/thread button once and the down button 11 times to give you 8 tpi which will be 1:1 spindle/lead screw for an 8 tpi lead screw.