Close

The problem with real life

A project log for Soldering pen

Cheaper fine-pitch soldering setup with Wellers RT tips

hp-banjohatHP (@banjohat) 06/14/2016 at 12:147 Comments

The problem with real life is that it tends to get in the way of your projects.

This update is far overdue - sorry for that! The paying projects (my job amongst other things) has been taking up most of my time. Also I'm leavning for a much needed vacation in a day. This all means that I won't be back until July to continue work on the soldering pen.

So what have I managed to get done?

Last time I wrote I was ordering PCBs. I have those - I've had them for a good month now. The trouble is programming the MCUs! I feel that I have tried everything. I have tried to solder all pins under a microscope securing all connections. I have assembled 6 boards but NO luck!

I have only put on the MCU and the ICSP header. That should give me both SPI access and power for the MCU. But I must have been doing something wrong on the design.

I have looked through the schematics but again - no luck. Everything seems to be in order - except is isnt. Could it be the MCUs? Maybe. I have them stored in a tray without any control of moisture. the tray is of the type with a small indent to keep each chip in place. a chip tray. it should be good enough, however I can't vouch for the chips. but that would mean that all my chips are fried. not a pretty thought.

I know that this project has gotten a fair amount of followers since the start. Maybe some of you would like to have a look at the design? If you find an error I will send you a working PCB :)

Discussions

tillz wrote 07/04/2016 at 12:10 point

Maybe that's too obvious, but did you checked your fuse settings when programming? Especially the clock source fuses? What's the exact problem with programming? Does it not return a device signature?

  Are you sure? yes | no

HP (@banjohat) wrote 07/04/2016 at 17:23 point

That's not obvious at all and is a really good comment! The board (from boards.txt) is the following (added an entry to the pro mini section):

pro.menu.cpu.Custom=ATmega168p NO OSC

pro.menu.cpu.Custom.upload.maximum_size=14336
pro.menu.cpu.Custom.upload.maximum_data_size=1024
pro.menu.cpu.Custom.upload.speed=19200

pro.menu.cpu.Custom.bootloader.low_fuses=0xe2
pro.menu.cpu.Custom.bootloader.high_fuses=0xdd
pro.menu.cpu.Custom.bootloader.extended_fuses=0xfe
pro.menu.cpu.Custom.bootloader.file=atmega/ATmegaBOOT_168_pro_8MHz.hex

pro.menu.cpu.Custom.build.mcu=atmega168
pro.menu.cpu.Custom.build.f_cpu=8000000L

Any comments to this is greatly appreciated! I have tried both atmega168 and atmega168p as mcu type...

  Are you sure? yes | no

tillz wrote 07/04/2016 at 17:30 point

Ok, and you're trying to install the arduino bootloader via ICSP, or directly uploading a sketch? Did you select the correct programmer (Arduino as ISP, I guess?) I can't see any hardware problems as the design is really the absolute minimum, so I'd definitely guess it's an software issue. I think the fuses settings look good.

  Are you sure? yes | no

HP (@banjohat) wrote 07/04/2016 at 17:39 point

I'm skipping the bootloader since i'm not having any serial communication anyway. Yes, I'm using the arduino as ISP. 

Thanks for verifying the fuses! It could even be my arduino installation. I should try and reinstall it all (maybe run directly from a zip file..) My boards.txt is located in a weird folder anyway :/

  Are you sure? yes | no

René Arts wrote 06/15/2016 at 09:20 point

Taking your words literally; if you only mounted the MCU and ICSP header, you might have forgotten R6 to pull the reset pin to high as a low (/floating) level on that pin triggers a reset? Or is the reset pin always actively controlled by the ICSP programmer when connected? Possibly it is simply a drain on the other side which cannot drive it high :-)

To be honest, I can't imagine much more faults as your design is quite simple (which is a good thing!) and it did function in your earlier version. 

  Are you sure? yes | no

HP (@banjohat) wrote 06/30/2016 at 18:21 point

Ah, This was the comment you mentioned!
As far as I know the reset pin should be push-pull controlled by the ICSP (an arduino in my case)
However, I have also tried to mount a 10k pull-up between reset and VCC just to see if that made any difference. It didn't.

Yeah, I have been baffled by my problems with this design as well! I have some 16MHz MCUs I can mount because maybe the other ones I have had lying around for a long time have been destroyed by moisture..


  Are you sure? yes | no

René Arts wrote 06/30/2016 at 19:18 point

Ah, that's a pity... as I said I really can't imagine what else could be wrong then. Good to hear from you anyway, good luck with the further debugging!

  Are you sure? yes | no