[solved] Power on problem

[nsunier]

Hi,

I have:

• 1x core module
  
• 2x ain modules
  
• 3x din modules
  
• 3x dout modules
  

I've loaded with success the ain64_din128_dout128_v2_0 example code. All seems to work quite well. I use a DC laboratory stabilised power supply (9V). I can switch on the PSU without applying the voltage and I have another switch on PSU in order to enable output voltage. So the voltage has a very small settling time. Often when I power on all DOUT go to 1 and nothing is displayed on LCD. It seems to be freezed. I have to retry 10-15 times until it succeeds.

Has someone an idea?

Thanks in advance for your help/ideas.

nsunier

[nsunier]

I've just found the answer by myself.

I was applying a 1A current limitation which is sufficient to supply all boards and all LED turned on. But by this way the inrush current is limited and does not allow the system to boot correctly. I have increased the limitation to max (3A) and it seems to work.

Hope it can help someone...

Have a nice day!  ;)

nsunier

[nsunier]

The problem seems not to be fully solved.

The boards are connected together by this way:

CORE -> DIN -> DIN -> DIN -> DOUT -> DOUT -> DOUT

Almost each DOUT is connected to a LED (>100 LED)

Often when I power on the system (without any current limitation: max available current is 3A) all DOUT LED switch on and stay powered and the CORE refuse to boot. When I remove the last DOUT module in the chain all LED blink at startup and the CORE boots without any problem.

What can I do?

Best regards,

nsunier

[DavidBanner]

what happens if you swap the last-dout in the chain with another one?

i.e. are all the dout PCBs correctly soldered etc, and are all the interconnect cables ok?

[nsunier]

I've connected this last DOUT module before the first DIN module:

CORE -> DOUT (3) -> DIN (1) -> DIN (2) -> DIN (3) -> DOUT (1) -> DOUT (2)

It produces the same problem. Sometimes it works, sometimes it doesn't.

I can observe two behaviours:

1. it works, I switch on-off-on quickly, it works again

2. it works, I switch on-off-on slowly (10 sec off), it doesn't works

[nsunier]

I've tried to remove all DIN modules:

CORE -> DOUT (3) -> DOUT (2) -> DOUT (1)

It produces the same problem. Sometimes it works, sometimes it doesn't.

[nsunier]

I've tried almost all configuration with only the three DOUT modules (with different cables). It's only when I use three modules that it produces boot faillures. One or two modules are ok. This design allows up to four DOUT modules?!

[stryd_one]

I had this kind of thing happen when I was feeding the core slightly too little voltage.... I found that with a bare core and LCD, I needed at least 8.6V or it just locked up sometimes, but at 8.4 it would just brownout and reset. ...

It's weird I know cause it should brownout whenever the voltage is too low, so I think it's the regulator starting to drop off, staying at 5V output, and just not supplying enough current. But it only happens in a very small window between brownout reset and working fine. Maybe turn your PSU up to 9.5V or more. The 7805 should take care of it no worries, at least for a short time for testing, but you may want to heatsink it. Your thumb will tell you ;)

Edit: maybe it's something about feeding regulated power into the regulating circuit.... Really we should turn down to 5V and power via J2. I wired mine to J1 to test it, and never went back to it after it powered on...

[nsunier]

Yeah! ;D It works!

You're right. With 9.5V the problem is solved.

I can't remove the 7805 because the final application doesn't have 5V, but... 12V. And 12V also work great. I use a BIG heatsink, so it's fully secure.

Thank you! ;)

[stryd_one]

Oh awesome. This is great news for two reasons:

1) You box works

2) Now people will believe me when I tell that story :D As far as I know, only you and I have had that problem.... It's a very exclusive club!