TheBackShed.com

Forum Index : Microcontroller and PC projects : Excessive Current Drain on Some MX170 Micromites- Mystery

Author

Message

jwettroth

Regular Member

Joined: 02/08/2011
Location: United States
Posts: 79

Posted: 05:18am 30 Aug 2021

Copy link to clipboard

Print this post

I have experienced a few MX170 Micromites (5%'ish) loaded with the latest firmware etc. that draw very high currents- like 120 mA at 3.3v even with the CPU Speed set to 5 Mhz- it doesn't seem to matter what the CPU Speed is set at. If I leave everything else the same in the circuit, Xmodem in the same code etc and just replace the Micro with another goes down to something around 6 mA for CPU 5 MHz and around 30 mA for full speed as expected. I've got several devices that do this. I'm building some flea power stuff where power matters a lot and I probably would never notice it in many projects but its kind of weird. The parts seem to function fine they just draw about 90 mA too much current all the time. I'm doing this all with developmental hardware that I'm debugging as I go but I've never had any really egregious HW errors where I've applied 5v to Vcc etc. The parts function fine except for this mysterious 1/4 watt of excess power. Any help appreciated. I have two three Micromites that exhibit this issue. I don't have time to full investigate right now but plan to.

John Wettroth

CaptainBoing

Guru

Joined: 07/09/2016
Location: United Kingdom
Posts: 2170

Posted: 06:00am 30 Aug 2021

Copy link to clipboard

Print this post

I have not seen this, but I have had weird stuff happen when the flash gets corrupted. I have a few hundred 170s out there (all with version 5.0408 which is old but operational reasons...) and they are solid as a rock. But I get bread-boarded ones that have thrown wobblers in the past. Not at all common but it has happened.

re-flash the PIC (what version are you using?) and then just let it sit at the prompt. Do you still get the high draw?

The re-flash will ensure you haven't got any pins or options defined that might be causing internal shorts with your wiring.
Edited 2021-08-30 16:12 by CaptainBoing

Mixtel90

Guru

Joined: 05/10/2019
Location: United Kingdom
Posts: 7938

Posted: 06:22am 30 Aug 2021

Copy link to clipboard

Print this post

A possible cause is that pins defined as inputs but not connected to anything can, in some cases, cause oscillation and current draw. It's not usual on Schottky input pins though. I suppose it *might* apply to console input. You could try pull-down (or pull-up) resistors on any likely pins.

Mick

Zilog Inside! nascom.info for Nascom & Gemini
Preliminary MMBasic docs & my PCB designs

Volhout
Guru

Joined: 05/03/2018
Location: Netherlands
Posts: 5091

Posted: 06:38am 30 Aug 2021

Copy link to clipboard

Print this post

The behaviour you describe fits a few failure modes.

1/ IO pin stuck: On these 3 boards there is an IO pin stuck to ground, that is programmed "high" in the code. Fix is easy: apparently the pin in question is not used (the boards run fine), program the pin low.

2/ MX170 core voltage regulator: The ceramic cap between pin 19 and 20 stabilizing the internal voltage regulator is causing oscillation of the internal voltage regulator. This can have several causes.
2a/ the value is too low (a 10uF ceramic can be as low as 4uF depending the voltage applied). Parallel a second one.
2b/ The layout is not optimal. The capacitor has 1 or 2 long (1" or more) connections to the chip.
2c/ The capacitor is bad (internally broken as result of soldering/board handling).

3/ Overclocking: You are running 40MHz chip at 48MHz. Some may work, some won't.

4/ Latchup: Maybe your design has 5V chips and 3.3V chips (the MX170). At power ON these voltages change from 0V to 5V and 3.3V. The order in which they do can be causing this. When there is 5V before the 3.3V, during a (very) short period of time the MX170 can get power through it's IO lines from the 5V chip. This is called "backpowering". It is possible that backpowering causes a latchup in the low voltage chip (any chip connected to 3.3V can cause this). IF your system is 3.3V only this is not the cause.

5/ RESET: The MX170 uses an internal reset circuit. Most reset circuits require a "fast rise" of the 3.3V to do a proper reset. In low power applications sometimes the 3.3V rises to slow for a good reset, and part of the MX170 is not initialized correctly.

6/ USB backpower: When you still have USB connected and have the laptop switched off (sleep) the connected USB cable can load the circuit.

Enough idea's.

IF all this is not the case, you may have faulty chips...?

PicomiteVGA PETSCII ROBOTS

jwettroth

Regular Member

Joined: 02/08/2011
Location: United States
Posts: 79

Posted: 04:52pm 30 Aug 2021

Copy link to clipboard

Print this post

Thanks guys- great ideas for testing. The best part about the whole Maximite/Micromite world might be this smart friendly group- no ego's, just straight help. I'm trying to get two prototypes out to my customer so I just put these duds aside for now- I'll report back to this thread in a week or so when I have some time. Somewhere in that list of suggestions has to be the answer. I hadn't considered corrupted firmware, I've had that happen before and break other things. Thanks again.

John Wettroth

Print this page

To reply to this topic, you need to log in.