Home
JAQForum Ver 24.01
Log In or Join  
Active Topics
Local Time 10:11 04 Jul 2025 Privacy Policy
Jump to

Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me.

Forum Index : Microcontroller and PC projects : CMM2: Bug reports

     Page 11 of 17    
Author Message
CircuitGizmos

Guru

Joined: 08/09/2011
Location: United States
Posts: 1427
Posted: 10:23pm 17 Jun 2020
Copy link to clipboard 
Print this post

  matherp said  
  Quote   At this point the only thing that I can suspect is that the processor chip (rev Y) that JLC is stuffing has bad internal RAM.


Send me a "bad" board and I'll test. Can't see a better solution. Email me direct and I'll let you have my address


This is pretty painful. I have customers waiting for shipments.

I have to make the decision to either refund their payment or ask them to wait perhaps weeks.

WW and Grogster - have you built boards partially assembled by JLC? Do they retain options?


  matherp said  
  Quote  I have had a similar auto-repeating [ENTER] key in the past too, although only after using the CTRL-W command in the editor. (Circuit Gizmo Waveshare version.) No problems with a random micro-sd card in an adapter though.


Completely unrelated, should be fixed in latest releases?


I saw this in release "p".
Edited 2020-06-18 08:26 by CircuitGizmos
Micromites and Maximites! - Beginning Maximite
 
GregZone
Senior Member

Joined: 22/05/2020
Location: New Zealand
Posts: 114
Posted: 10:30pm 17 Jun 2020
Copy link to clipboard 
Print this post

  CircuitGizmos said  I do OPTION LIST and get:

OPTION USBKEYBOARD UK

With battery still in, I type:

OPTION USBKEYBOARD US

After the reset, I type OPTION LIST and get:

OPTION USBKEYBOARD UK


It is a Y processor, if that matters.

Interesting that this is the same thing I'm experiencing with my troubled board that I just received from you (as discussed in the main CMM2 topic).

In my case, not only the USBKEYBOARD option not being retained, but also other OPTION settings like the SD TIMING.

ie. OPTION LIST appears to always prints the same unchanged initial 3 lines.

So, although my show stopper problem is not being able to get SD Card functional, this OPTION issue was viewed as the primary problem to focus on / understand.
 
GregZone
Senior Member

Joined: 22/05/2020
Location: New Zealand
Posts: 114
Posted: 10:39pm 17 Jun 2020
Copy link to clipboard 
Print this post

  CircuitGizmos said  
  matherp said  
  Quote  Correct me if I'm wrong, but with no battery shouldn't the setting survive the reboot?


No: all settings are stored in a battery backed area in the RTC domain on the chip. Without the battery there is no power to that RAM


I see. I would have thought that the battery is the backup source for that RAM, not the only source.

Someone mentioned that the VBAT jumper on the WaveShare board is needed if you remove the battery.

This coincides with the WaveShare documentation:
VBAT: short the jumper to use system power supply, open it to connect external power, such as battery

ie. It appears you either have a battery connected, OR, you have the VBAT jumper in place, otherwise there would be no power to the RTC / backup?
 
CircuitGizmos

Guru

Joined: 08/09/2011
Location: United States
Posts: 1427
Posted: 10:39pm 17 Jun 2020
Copy link to clipboard 
Print this post

  GregZone said  
  CircuitGizmos said  I do OPTION LIST and get:

OPTION USBKEYBOARD UK

With battery still in, I type:

OPTION USBKEYBOARD US

After the reset, I type OPTION LIST and get:

OPTION USBKEYBOARD UK


It is a Y processor, if that matters.

Interesting that this is the same thing I'm experiencing with my troubled board that I just received from you (as discussed in the main CMM2 topic).

In my case, not only the USBKEYBOARD option not being retained, but also other OPTION settings like the SD TIMING.

ie. OPTION LIST appears to always prints the same unchanged initial 3 lines.

So, although my show stopper problem is not being able to get SD Card functional, this OPTION issue was viewed as the primary problem to focus on / understand.


The boards that I can't get to stick to "US" also cannot access the SD card. They all have 3V batteries and the path from the battery to the processor pin is fine.

The difference is that they are not waveshare boards. They are the version 1.5 board with the surface mount parts put in place by JLC.
Micromites and Maximites! - Beginning Maximite
 
CircuitGizmos

Guru

Joined: 08/09/2011
Location: United States
Posts: 1427
Posted: 10:46pm 17 Jun 2020
Copy link to clipboard 
Print this post


Micromites and Maximites! - Beginning Maximite
 
GregZone
Senior Member

Joined: 22/05/2020
Location: New Zealand
Posts: 114
Posted: 10:47pm 17 Jun 2020
Copy link to clipboard 
Print this post

  CircuitGizmos said  The boards that I can't get to stick to "US" also cannot access the SD card. They all have 3V batteries and the path from the battery to the processor pin is fine.

The difference is that they are not waveshare boards. They are the version 1.5 board with the surface mount parts put in place by JLC.


Interestingly though, what you just said seems to directly match with the issues I am having with the Waveshare based v2.0 board, that I just received from you!

ie. US option not sticking, and also cannot access SD card!

nb. I also recently noted the repeating key issue after arecent firmware update (I'm about to load the latest to see if that at least goes away).

So, in summary, what you are describing is identical to the issues I'm having on the Waveshare board you sent me.
 
CircuitGizmos

Guru

Joined: 08/09/2011
Location: United States
Posts: 1427
Posted: 10:53pm 17 Jun 2020
Copy link to clipboard 
Print this post

  GregZone said  
  CircuitGizmos said  The boards that I can't get to stick to "US" also cannot access the SD card. They all have 3V batteries and the path from the battery to the processor pin is fine.

The difference is that they are not waveshare boards. They are the version 1.5 board with the surface mount parts put in place by JLC.


Interestingly though, what you just said seems to directly match with the issues I am having with the Waveshare based v2.0 board, that I just received from you!

ie. US option not sticking, and also cannot access SD card!

nb. I also recently noted the repeating key issue after arecent firmware update (I'm about to load the latest to see if that at least goes away).

So, in summary, what you are describing is identical to the issues I'm having on the Waveshare board you sent me.


Exactly! The design is the "same" but the implementation is different.

OPTIONS are stored in this battery RAM. I don't know what else is. That is why I suspect that the RAM isn't right. It's a possibility. Bits stuck either high or low? That RAM is important enough that testing it on powerup might be prudent. It's only a few k.

What else could be wrong? The 'CUBE' programmer? I updated it. Residue? I used Trichloroethane to clean the board and give myself cancer. The big RAM passes the minimum RAM test.
Edited 2020-06-18 08:57 by CircuitGizmos
Micromites and Maximites! - Beginning Maximite
 
TassyJim

Guru

Joined: 07/08/2011
Location: Australia
Posts: 6266
Posted: 10:58pm 17 Jun 2020
Copy link to clipboard 
Print this post

How are you programming them?
With a USB A-A cable or through the serial?
What version of STM32CubeProgrammer are you using?

I use a USB A-A cable and V2.1.0

Jim
VK7JH
MMedit
 
CircuitGizmos

Guru

Joined: 08/09/2011
Location: United States
Posts: 1427
Posted: 10:59pm 17 Jun 2020
Copy link to clipboard 
Print this post

  TassyJim said  How are you programming them?
With a USB A-A cable or through the serial?
What version of STM32CubeProgrammer are you using?

I use a USB A-A cable and V2.1.0

Jim


A-A and v2.4.0

Edited to add: With verify on.
Edited 2020-06-18 09:00 by CircuitGizmos
Micromites and Maximites! - Beginning Maximite
 
GregZone
Senior Member

Joined: 22/05/2020
Location: New Zealand
Posts: 114
Posted: 11:02pm 17 Jun 2020
Copy link to clipboard 
Print this post

  CircuitGizmos said  
  GregZone said  
  CircuitGizmos said  The boards that I can't get to stick to "US" also cannot access the SD card. They all have 3V batteries and the path from the battery to the processor pin is fine.

The difference is that they are not waveshare boards. They are the version 1.5 board with the surface mount parts put in place by JLC.


Interestingly though, what you just said seems to directly match with the issues I am having with the Waveshare based v2.0 board, that I just received from you!

ie. US option not sticking, and also cannot access SD card!

nb. I also recently noted the repeating key issue after arecent firmware update (I'm about to load the latest to see if that at least goes away).

So, in summary, what you are describing is identical to the issues I'm having on the Waveshare board you sent me.


Exactly! The design is the "same" but the implementation is different.

OPTIONS are stored in this battery RAM. I don't know what else is. That is why I suspect that the RAM isn't right. It's a possibility. Bits stuck either high or low? That RAM is important enough that testing it on powerup might be prudent. It's only a few k.

So, given that you are having the same problem with some JLC supplied SoC chips, that I'm also having on a WaveShare based CMM2, that would seem to rule out a bad batch of SoC chips (The JLC supplied SoC chips presumably won't be from the same source that you sourced your WaveShare boards from?).

Wouldn't this then point to it being more likley a firmware issue (not a SoC / hardware fault)?

Should we be loading some earlier firmware, to see if the issues go away?
 
GregZone
Senior Member

Joined: 22/05/2020
Location: New Zealand
Posts: 114
Posted: 11:06pm 17 Jun 2020
Copy link to clipboard 
Print this post

  CircuitGizmos said  
A-A and v2.4.0

Edited to add: With verify on.

FWIW, I'm also using v2.4.0

I've used both A-A (USB) and A-B (UART) with same result (both methods verified successfully).
 
TassyJim

Guru

Joined: 07/08/2011
Location: Australia
Posts: 6266
Posted: 11:08pm 17 Jun 2020
Copy link to clipboard 
Print this post

  CircuitGizmos said  
  TassyJim said  How are you programming them?
With a USB A-A cable or through the serial?
What version of STM32CubeProgrammer are you using?

I use a USB A-A cable and V2.1.0

Jim


A-A and v2.4.0

Edited to add: With verify on.


Can you try V2.1.0 ?
That is the only difference I can see that could be doing something so strange.

I can put V2.1 up on my website for you to grab if needed.


Jim
VK7JH
MMedit
 
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 1865
Posted: 11:10pm 17 Jun 2020
Copy link to clipboard 
Print this post

The PIN("BAT") function returns 3.1 volts. So MM Basic is seeing the battery, and previously it was found that it does accept Options with the Waveshare VBAT link installed (without battery).

So the options should be able to be set and read back, but they are not?, so either the writes to RTC RAM are not happening or the Read from RTC RAM is returning nothing, hence the same MM basic defaults are displayed when you read from RTC RAM, I don't know if any of the coms bus is exposed on the Breakouts or Motherboard  connections. I would think that is the only way it could cause this option problem.      

If you had a small USB to serial convertor, I'm using a CP2012, you could unplug the Waveshare board, install VBAT link, and with 4 leads from the serial board tacked on top (5v TX RX GND), you could test the Options RTC RAM read write via a PC terminal and rule out the Motherboard as the cause. Which means the Waveshare boards have a problem.

Mike.
NANO Inverter: Full download - Only Hex Ver 8.1Ks
 
CircuitGizmos

Guru

Joined: 08/09/2011
Location: United States
Posts: 1427
Posted: 11:11pm 17 Jun 2020
Copy link to clipboard 
Print this post

  GregZone said  
So, given that you are having the same problem with some JLC supplied SoC chips, that I'm also having on a WaveShare based CMM2, that would seem to rule out a bad batch of SoC chips (The JLC supplied SoC chips presumably won't be from the same source that you sourced your WaveShare boards from?).

Wouldn't this then point to it being more likley a firmware issue (not a SoC / hardware fault)?

Should we be loading some earlier firmware, to see if the issues go away?


I loaded rc17 (just grabbed an early one) and the same thing happens. This is one that only gives 0=UK and 1=US as options.

No battery installed. I reset it (button), and it didn't ask for keyboard type. So the reset code is different from the power up code? Not having the battery in there should have had it ask again for keyboard type...
Micromites and Maximites! - Beginning Maximite
 
CircuitGizmos

Guru

Joined: 08/09/2011
Location: United States
Posts: 1427
Posted: 11:14pm 17 Jun 2020
Copy link to clipboard 
Print this post

  TassyJim said  
  CircuitGizmos said  
  TassyJim said  How are you programming them?
With a USB A-A cable or through the serial?
What version of STM32CubeProgrammer are you using?

I use a USB A-A cable and V2.1.0

Jim


A-A and v2.4.0

Edited to add: With verify on.


Can you try V2.1.0 ?
That is the only difference I can see that could be doing something so strange.

I can put V2.1 up on my website for you to grab if needed.


Jim


Yes, please. Or email to me.
Micromites and Maximites! - Beginning Maximite
 
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 1865
Posted: 11:20pm 17 Jun 2020
Copy link to clipboard 
Print this post

I'm using STM32CubeProgrammer v2.4.0
NANO Inverter: Full download - Only Hex Ver 8.1Ks
 
CircuitGizmos

Guru

Joined: 08/09/2011
Location: United States
Posts: 1427
Posted: 11:21pm 17 Jun 2020
Copy link to clipboard 
Print this post

  KeepIS said  I'm using STM32CubeProgrammer v2.4.0


Then its probably not the programmer...
Micromites and Maximites! - Beginning Maximite
 
GregZone
Senior Member

Joined: 22/05/2020
Location: New Zealand
Posts: 114
Posted: 11:25pm 17 Jun 2020
Copy link to clipboard 
Print this post

I've just reinstalled STM32CubeProgrammer v2.2.0 (earliest version still available on their website).

I can report no change / same result.

I also reverted back to v5.05.02 firmware, and strangly I'm still getting the random key repeat issue (along with the OPTION / SD Card failing issue).  :(
 
TassyJim

Guru

Joined: 07/08/2011
Location: Australia
Posts: 6266
Posted: 11:28pm 17 Jun 2020
Copy link to clipboard 
Print this post

Here is V2.1
https://www.c-com.com.au/stuff/cube21.zip

There could be something to do with a combination of chip revision and cube version.
We had some issues with saving options between versions during development.
V2.1 works OK for both my V and Y Waveshare boards.

Re battery backup
Pressing reset with power still applied but no battery will retain settings.
Battery only comes into play with power removed.

Jim
VK7JH
MMedit
 
GregZone
Senior Member

Joined: 22/05/2020
Location: New Zealand
Posts: 114
Posted: 11:39pm 17 Jun 2020
Copy link to clipboard 
Print this post

  TassyJim said  Here is V2.1
https://www.c-com.com.au/stuff/cube21.zip
Jim

Thanks. I have now also re-installed STM32CubeProgrammer v2.1.0, and same issue.

ie. It appears this issue is not related to Programmer version (ie. v2.1.0, v2.2.0 or v2.4.0 all have same result).
 
     Page 11 of 17    
Print this page
The Back Shed's forum code is written, and hosted, in Australia.
© JAQ Software 2025