search

TomWhitwell / TuringMachine

Turing Machine Mk 2 Main Module
223 stars 41 forks source link

TM2 locks with all lights steady and bright after a minute or so #201

Closed alphonsom closed 1 year ago

alphonsom commented 1 year ago

Newly built Turing Machine behaves for a minute or less, then all the LEDs light up fully bright and steady, and output ceases. It is calibrated as far as is possible within the time allowed before lockup. Even in the locked state, noise output is normal, and so is pulse output. Any hints or direction about where to look would be appreciated.

Thonk-Steve commented 1 year ago

That's an unusual one. What do you have to do to get it working again?

On Sun, 23 Apr 2023, 09:40 alphonsom, @.***> wrote:

Newly built Turing Machine behaves for a minute or less, then all the LEDs light up fully bright and steady, and output ceases. It is calibrated as far as is possible within the time allowed before lockup. Even in the locked state, noise output is normal, and so is pulse output. Any hints or direction about where to look would be appreciated.

— Reply to this email directly, view it on GitHub https://github.com/TomWhitwell/TuringMachine/issues/201, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAKXH6DEXNDN7BH3G5IUI2TXCTTIPANCNFSM6AAAAAAXIMNY5E . You are receiving this because you are subscribed to this thread.Message ID: @.***>

alphonsom commented 1 year ago

Switch off the power, and switch it back on again. Then the TM starts with all LEDs bright for a few seconds, followed by them going off in sequence from left to right in time with the incoming clock. Then there is a pause where all is dark for a few seconds. Then the normal behaviour starts for a while, and we go back to the beginning.

Thonk-Steve commented 1 year ago

Hey, what are you clocking it with?

On Sun, 23 Apr 2023, 10:26 alphonsom, @.***> wrote:

Switch off the power, and switch it back on again. Then the TM starts with all LEDs bright for a few seconds, followed by them going off in sequence from left to right in time with the incoming clock. Then there is a pause where all is dark for a few seconds. Then the normal behaviour starts for a while, and we go back to the beginning.

— Reply to this email directly, view it on GitHub https://github.com/TomWhitwell/TuringMachine/issues/201#issuecomment-1519009552, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAKXH6FSFDVMNSZUZHQ3JH3XCTYUVANCNFSM6AAAAAAXIMNY5E . You are receiving this because you commented.Message ID: @.***>

alphonsom commented 1 year ago

A square wave out of an AI 011 VCO for calibration, or a clock signal from a Frequency Central Wave Runner LFO for normal. Same thing happens with either.

TomWhitwell commented 1 year ago

That's a real mystery - is there any effect on the other modules in the case - sounds like a power problem but I'm sure you've looked at that.

alphonsom commented 1 year ago

There is no effect on any other module. I tried a different power supply and cables just to be sure, and the behaviour is the same. Could something be causing a draining of randomness on the shift register return switch? Or a gradual inversion failure?

Thonk-Steve commented 1 year ago

Hi, did you build the module yourself? Did you use the IC pack from a Thonk kit?

On Mon, 24 Apr 2023, 17:37 alphonsom, @.***> wrote:

There is no effect on any other module. I tried a different power supply and cables just to be sure, and the behaviour is the same. Could something be causing a draining of randomness on the shift register return switch? Or a gradual inversion failure?

— Reply to this email directly, view it on GitHub https://github.com/TomWhitwell/TuringMachine/issues/201#issuecomment-1520495858, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAKXH6EEXNF6X6SBPHDI3LDXC2T5XANCNFSM6AAAAAAXIMNY5E . You are receiving this because you commented.Message ID: @.***>

alphonsom commented 1 year ago

Yes to both questions.

Thonk-Steve commented 1 year ago

Ok the best step next would be to remove the front panel and post good photos of both sides of both PCBs.

On Mon, 24 Apr 2023, 17:47 alphonsom, @.***> wrote:

Yes to both questions.

— Reply to this email directly, view it on GitHub https://github.com/TomWhitwell/TuringMachine/issues/201#issuecomment-1520507404, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAKXH6F4Q33VV4KOSGI6IODXC2VBDANCNFSM6AAAAAAXIMNY5E . You are receiving this because you commented.Message ID: @.***>

alphonsom commented 1 year ago

Aha. I combed its hair and washed its face for the photos, and noticed a couple of joints that I was sure you'd be telling me to reflow. So I pre-empted - reflowed, tested, and it all works lovely. Thanks for your time.