Power-on odd behaviour.

Patrick52
 

Posts: 18
Joined: 18 Dec 2022, 16:38

Power-on odd behaviour.

Post by Patrick52 »

I first posted this in the wrong place. Thanks for the gentle nudge Lucas66.

I have a weird problem.
When I plug in the USB cable to a PC or a 12v battery the device ( tr(u)SDX)won’t power up. If I leave it for a few hours it does , still plugged in. Sometimes then it will power up if I power cycle. Sometimes it won’t. I’ve taken it apart and checked the soldering with a microscope, no duff joints in the smd devices, I’ve ‘scoped the I2C busses and they make sense. The ‘scope can decode the activity ok.
Voltages seem to be ok.
Re-assemble and it’s fine. Power cycle…. Fine. 12v battery …… fine. Leave for a while , power on , fine

Leave over night. Power on, nothing. Connect to a PC , nothing. Leave for a couple of hours, it’s working. Connect to the battery, nothing. Connect to PC, nothing. Connect to wall charging point , nothing.

I wonder if there is a delay if the I2C commands don’t see an ack straight away ?

Any pointers/advice gratefully received.
Regards
Patrick

Update :
Just spent a great half hour with the tr(u)SDX in our garden connected to an inverted V aerial which nanoVNA gives dip at 7.078 MHz with SWR of 1.02

Listened to several QSOs and watched decodes of CW. plenty of signal.

Tried to call but not sure it's working yet. Need to check that transmission with another rx.

Best regards
Patrick
Luca66
 

Posts: 123
Joined: 16 Apr 2022, 13:02

Re: Power-on odd behaviour.

Post by Luca66 »

Hallo Patrick

These power-on behaviors are strange and random. I don't have a strategy to propose and I hope someone can give you more concrete help. To verify your transmission use a dummy load and a receiver that is not too close but without an antenna.

73 Luca
Patrick52
 

Posts: 18
Joined: 18 Dec 2022, 16:38

Re: Power-on odd behaviour.

Post by Patrick52 »

Here is an update, the problem is now diagnosed as temperature related.
I had the device in the house…. Working. Put it in the fridge at 4C, took out, power up and not working.
Power down.
3 minutes later ….. power up and working.
Posted to the Facebook group and lots of good ideas … so today I’ll re solder as many of the joints on the SMD devices as I can manage. Then maybe cold spray different chips.

During further testing I checked the frequency of the three oscillator/crystals on the main board. I found the one connected to the ATmega328P was 20MHz. The data sheet I found on the Microchip website shows a max frequency of 16MHz. This seems strange. I wonder if my problem comes from my assembler , not approved , has purchased the wrong device ?
I will get out the microscope and double check.
Patrick52
 

Posts: 18
Joined: 18 Dec 2022, 16:38

Re: Power-on odd behaviour. - update 2

Post by Patrick52 »

Today I went round the main board SMD devices and applied solder where there wasn't a decent fillet.
It seems some places where the SMD device pad was supposed to be connected to a larger pcb track or a via was close by had not been given sufficient solder paste to work properly. It is hard to identify a bad joint totally reliably so I just went round them all.
I used a binocular microscope for this purpose. The ICs all looked fine.

So back to the fridge for the temperature check .... and it seems ok. Starts on connection to the battery.

Now to use it in the field.

All the best
Patrick
Constrainted
 

Posts: 56
Joined: 20 Dec 2022, 20:51

Re: Power-on odd behaviour.

Post by Constrainted »

Awesome!
Luca66
 

Posts: 123
Joined: 16 Apr 2022, 13:02

Re: Power-on odd behaviour.

Post by Luca66 »

Did you solve the problem?
73 Luca
dl6sez
 

Posts: 399
Joined: 30 Dec 2021, 22:54

Re: Power-on odd behaviour.

Post by dl6sez »

Luca66, just reading is the solution to your (unnecessary) question :)

73 de Chris
Chris DL6SEZ, JN48XL near Ulm, Southern Germany
Luca66
 

Posts: 123
Joined: 16 Apr 2022, 13:02

Re: Power-on odd behaviour.

Post by Luca66 »

Dl6sez, being a moderator doesn't mean you have to answer unnecessary questions. And so don't waste your time with unnecessary answers. :)

Patrick, figured out that the problem is temperature related, but didn't fix the problem, I think....
73.
dl6sez
 

Posts: 399
Joined: 30 Dec 2021, 22:54

Re: Power-on odd behaviour.

Post by dl6sez »

Luca66,

it is stupid but you can not read!
He wrotes the problem is solved with reflowing and tested it afterwards with fridge...
Chris DL6SEZ, JN48XL near Ulm, Southern Germany
Luca66
 

Posts: 123
Joined: 16 Apr 2022, 13:02

Re: Power-on odd behaviour.

Post by Luca66 »

I hope for him that he has solved the problem, given that he has redone many welds and none of them, however, was a sign of a defect.
A few hours of trouble-free operation will tell if he's solved it.
Patrick52
 

Posts: 18
Joined: 18 Dec 2022, 16:38

Re: Power-on odd behaviour.

Post by Patrick52 »

Further news...

i was getting ready to use the radio, (tr)uSDX, and thought i'd just power it on to check before going out ...

And it failed ...... again ..... so now it's back to the 'scope to try and find out why it stops.

Please ... somebody.... could I have a glimpse at the software, even just the outline sequence on power-up with a bit about what happens if errors occur ?

Please ..... i promise i won't tell.

Thanks
Patrick ( frustrated as everybody else's works so well !)
Patrick52
 

Posts: 18
Joined: 18 Dec 2022, 16:38

Re: Power-on odd behaviour.

Post by Patrick52 »

I've done some more investigation.
I put 'scope probes onto the I2C signals ( SCL, SDA ) for the display and powered up the radio. It worked ok.
But then it didn't but when I twiddled the knob ( changing the frequency I think ) it started to fill in the display and 'noise' was heard from the speaker.

The 'scope I have is a Siglent 4 channel 100MHz/1Gs/s with the ability to decode I2C. I couldn't detect any bad transactions at radio power-on when the radio didn't work. It may well be they are there but I haven't found them yet.

I'll keep updating.

Patrick
Luca66
 

Posts: 123
Joined: 16 Apr 2022, 13:02

Re: Power-on odd behaviour.

Post by Luca66 »

Patrick52 wrote: 12 Jan 2023, 20:05 Further news...

i was getting ready to use the radio, (tr)uSDX, and thought i'd just power it on to check before going out ...

And it failed ...... again ..... so now it's back to the 'scope to try and find out why it stops.

Please ... somebody.... could I have a glimpse at the software, even just the outline sequence on power-up with a bit about what happens if errors occur ?

Please ..... i promise i won't tell.

Thanks
Patrick ( frustrated as everybody else's works so well !)
I'm sorry Patrick that you haven't solved your problem, but I was cautious about considering the problem solved.

Someone else doesn't. Strange!

I hope someone with expertise can help you.

73 Luca
dl6sez
 

Posts: 399
Joined: 30 Dec 2021, 22:54

Re: Power-on odd behaviour.

Post by dl6sez »

Hello Patrick,

measuring makes not really sense on I2C, the other (tr)uSDX work and haven't had such a strange behaviour like yours.
Software (bootloader and firmware) has been reprogrammed by you as i understood.
Cold solder joints are still a possibility, but there could also be a problem with your Atmega328P.
I would change it simply now. Not really nice to do but also not hard ;)
There are chinese copies of Atmega328 on the market.
Maybe it has a problem with the 20MHz, sounds like this is possible from your explanations.
Also Microchip bought Atmel and changed the spec of Atmega328P to only work up to 16MHz today, so please use a new Atmega328PB which can do 20MHz according to Microchip specs still as former both types :)
This all happened due to chinese Covid shortings in semiconductor market.

Sorry i have no other idea, the I2C components on (tr)uSDX can't really be the reason, because it works sometimes flawlessly.

73 de Chris
Chris DL6SEZ, JN48XL near Ulm, Southern Germany
Patrick52
 

Posts: 18
Joined: 18 Dec 2022, 16:38

Re: Power-on odd behaviour.

Post by Patrick52 »

It still defies me…..
I agree that it may be the wrong version of the 328 but I can’t find information about the chip markings that would definitively show me which chip is in place.
I’ll start another post in this troubleshooting I think.
Maybe purchase a kit from china which will definitely have all the correct parts. After all the rf board is fine.
dl6sez
 

Posts: 399
Joined: 30 Dec 2021, 22:54

Re: Power-on odd behaviour.

Post by dl6sez »

Hello Patrick,

if there is not clearly "328PB" on the chip
yes there is PB and on former ones only P than it is a fake copy from China.
You won't start another thread...

73 de Chris
Chris DL6SEZ, JN48XL near Ulm, Southern Germany
Patrick52
 

Posts: 18
Joined: 18 Dec 2022, 16:38

Re: Power-on odd behaviour.

Post by Patrick52 »

Well... finally I think i've tracked the problem down with help from Chris.
I have checked the ident on the ATmega chip with a microscope and it reads as follows
ATMEL
MEGA328
U-TH
354730
2129H67

So it looks like it's a chinese copy ?

I'll try and source a proper one and persuade my friendly surface mount guy to give me a hand to replace it.

Any thoughts on the best supplier ?
dl6sez
 

Posts: 399
Joined: 30 Dec 2021, 22:54

Re: Power-on odd behaviour.

Post by dl6sez »

Hello Patrick,

i looked on one of my two working (tr)uSDX today and it is showing:
Atmel (in Logo letters)
Mega328P
....
Sure a fake or copy on yours.
I have checked at Mouser they have Atmega328PB-AUR
R means Reel, you can get single one from a Reel (cut tape...).
Only buy at a big serious distributors like Mouser, Farnell, Digikey, in Germany: Reichelt, Buerklin and Distrelec.

73 de Chris
Chris DL6SEZ, JN48XL near Ulm, Southern Germany
dl6sez
 

Posts: 399
Joined: 30 Dec 2021, 22:54

Re: Power-on odd behaviour.

Post by dl6sez »

Hello Patrick,

for Info, mine looks like the foto of the aliexpress.com sold Atmega328PB-AU.
Article no.: 1005005023235188
without B it is an old 328P capable of 20MHz, but please buy a PB than you are shure if it's
original Atmel/Microchip and can do 20MHz.

73 de Chris
Chris DL6SEZ, JN48XL near Ulm, Southern Germany
Patrick52
 

Posts: 18
Joined: 18 Dec 2022, 16:38

Re: Power-on odd behaviour.

Post by Patrick52 »

Thanks Chris, I'll buy a new one.

I'll also contact the guy who sold me the kit.... to let him know .....

Best regards
Patrick
Patrick52
 

Posts: 18
Joined: 18 Dec 2022, 16:38

Re: Power-on odd behaviour.

Post by Patrick52 »

Order placed for new chip from Mouser. Delivery is April 2023 ...

Ah well, "good things come to those who wait" as the saying goes in the UK

I tried a german firm but MOQ was 1000.00 euro !

Best regards
Patrick
DL2MAN
 

Posts: 706
Joined: 30 Dec 2021, 19:18
Contact:

Re: Power-on odd behaviour.

Post by DL2MAN »

I bought my AtMega Stock from an austrian dealer on ebay....
They all worked fine.

And I donßt think your AtMega is the Problem. We have identified problems with some "late" batches, that were swept somewhere on a shop floor and sold in the chip crisis, but this is the exeption to the rule.... Normally it´s some bad solder connection. Bootloader corrupted, some bit lost in firmware etc.

73 Manuel, DL2MAN
Patrick52
 

Posts: 18
Joined: 18 Dec 2022, 16:38

Re: Power-on odd behaviour.

Post by Patrick52 »

Well ... that's the end for now.

I used the serial number provided by Javier, the seller, and downloaded firmware.

Then using USBASP with Avrdudess I successfully downloaded and verified to the SDX.

Then power cycle ...... and nothing at all. Even after a while...

So now it's going in a box and I'll think about it again when the replacement arrives.

Maybe I should give up on it :( and buy a commercial HF txvcr? Seems a shame really.
storyofyouandme
 

Posts: 54
Joined: 03 Jan 2022, 15:30

Re: Power-on odd behaviour.

Post by storyofyouandme »

With all due respect, I don't think your question is 328. First of all, the 328 with the TH mark is not a fake from China. The 328 marked with TH is made in Thailand, and its quality is better than the 328 made in Korea. I can't buy the products made in Thailand, but I can only use the products made in South Korea. There is no big difference in actual use. The pads of TRUSDX of DL2 version are really small, which makes manual welding very difficult. It is also for this reason that faulty soldering is not ruled out when the machine is mounted. So I think the above judgment of DL2 is correct. The first is probably faulty soldering, and the second is firmware. I have also encountered similar strange problems. The menu button does not work when powered on, and the other buttons are normal. After about 15 seconds, the menu button is normal. And in winter and summer, I checked all the welds, and the solder joints are normal. I judged that it was a capacitance problem or a grounding problem, although the grounding is normal when measured with a multimeter. Because it doesn't affect the use, I don't care about it. I hope it can help you.
73.
DL2MAN
 

Posts: 706
Joined: 30 Dec 2021, 19:18
Contact:

Re: Power-on odd behaviour.

Post by DL2MAN »

Patrick52 wrote: 24 Jan 2023, 20:13 Well ... that's the end for now.

I used the serial number provided by Javier, the seller, and downloaded firmware.

Then using USBASP with Avrdudess I successfully downloaded and verified to the SDX.

Then power cycle ...... and nothing at all. Even after a while...

So now it's going in a box and I'll think about it again when the replacement arrives.

Maybe I should give up on it :( and buy a commercial HF txvcr? Seems a shame really.
It´s up to you, if you want to give up.
I wouldn´t. Usually it´s something simple like a cold solder joint, a band inter PCB connection.
Sometimes it´s even something stupid like: OLED Mod done wrong and connected to 12V instead of 5V/VCC, so the display is dead now, because over Over Voltage....

Just stay focused and logical. If you cannot do it on your own, ask a nerd friend. It´s really not much that can be the cause.

73 Manuel; DL2MAN
Post Reply

Users browsing this forum: No registered users and 3 guests