Author Topic: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)  (Read 451892 times)

0 Members and 2 Guests are viewing this topic.

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1025 on: August 05, 2018, 10:00:47 am »
Please open a new issue in firmware repository for this. I cannot promise anything since Due is already too busy taking care of TFT display and all other things to provide 50/60 Hz (neither half or full wave) but possibly we can make some optimization if we make it as special case that not necessarily works with already implemented LIST function.

Did you perhaps tried to use remote voltage programming input to accomplish such thing?

Offline Pjoms

  • Regular Contributor
  • *
  • Posts: 51
  • Country: se
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1026 on: August 05, 2018, 10:07:50 am »



Did you perhaps tried to use remote voltage programming input to accomplish such thing?

I have thought of that but not test it.
It will probably work just fine, but need some external circuits.
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1027 on: August 05, 2018, 10:10:20 am »
Yes, you have to split AC signal into two halves. I'm agree that is not straightforward. We'll see what can be done in firmware.

Offline Pjoms

  • Regular Contributor
  • *
  • Posts: 51
  • Country: se
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1028 on: August 05, 2018, 03:41:15 pm »
Please open a new issue in firmware repository for this.

Done. #209
 

Offline Pjoms

  • Regular Contributor
  • *
  • Posts: 51
  • Country: se
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1029 on: August 08, 2018, 05:01:58 pm »



Did you perhaps tried to use remote voltage programming input to accomplish such thing?

I have thought of that but not test it.
It will probably work just fine, but need some external circuits.


I made a quick try using the remote voltage programming inputs for AC generating.

- Both negative terminals on the psu are connected together. AC out on the positive terminals.
- The control signals feeds from a two channel function generator with one channel phase shifted 180 deg.
- Each signal is half wave rectified through a diode, and the dc-level from the function generator is raised to compensate for the voltage drop in the diodes.


It works quite nice, but with some drawbacks.
1. The max limit for current/power is what the down programmer can handle as it have to sink all current from the other channel.
2. As both control signals have a common ground, the both analog Gnd (pin 2 on the front connector) also will be common which messes up the current measurements due to the low side current shunt.

Despite the limit of the down programmer I still believe that an AC feature like this will be useful if it is possible to squeeze in the firmware.
 
The following users thanked this post: prasimix

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1030 on: August 08, 2018, 05:04:17 pm »
Thanks for an update. Do you think that an option to turn off DP could improve your situation?

Offline Pjoms

  • Regular Contributor
  • *
  • Posts: 51
  • Country: se
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1031 on: August 08, 2018, 05:10:49 pm »
Thanks for an update. Do you think that an option to turn off DP could improve your situation?

Correct me if I’m wrong, but without the DP it will be impossible for the “off” channel to sink the current from the one that is doing its half wave?
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1032 on: August 08, 2018, 05:12:17 pm »
Oops, you're right :)

Offline Pjoms

  • Regular Contributor
  • *
  • Posts: 51
  • Country: se
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1033 on: August 08, 2018, 08:07:35 pm »
Can the DP protection handle a pulsing load? If it has to dissipate for example 5 W 50Hz the average will be 2.5 W but it’s never goes over 1 W for 5 sec to trip the protection?
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1034 on: August 08, 2018, 09:21:11 pm »
Hm, that could be problematic. DP MOSFET is in DPAK soldered on very limited pad and NTC sensor is not close to it (that control cooling fan), therefore if firmware cannot detect pulsed excess of power dissipation that can damage MOSFET (due to overheating).
It was a stupid move from my side to use DPAK instead of THT package that can be mounted on heatsink beneath :(. Due to that we added that firmware protection that has its limitations. I have to check with my colleague what firmware can do.

Offline kirby

  • Newbie
  • Posts: 6
  • Country: us
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1035 on: August 15, 2018, 02:35:16 pm »
Greetings! My H24005 has stopped working and before I started disassembly I thought I would see if there were any suggestions.

Thinks were running fine, putting out 14V at about 20mA connected to a recalcitrant charging circuit. As so often happens, work rudely interfered. I was away for about 15 minutes. I found my power supply with a blank screen. I powered it off, turned it back on, and the backlight on the LCD flashes once and then thinks looked like there was no power.

I plugged the USB into my computer and the Arduino failed to appear. Took the cover off, powered up, and the two lights on the USB connector side of the Arduino were on.  I thought it might be the Due, so I unseated it. Looks like the Due is fine. I upgraded the firmware to 1.1.1 while it was out. Reseated the Due, powered up, and the same behavior. When I power up and switch off standby, the LCD backlight flashes briefly (.25 secondish) and then the system appears dead except for the two lights that are on the Due.

I did measure 3.6 volts on pins 22 and 23 of the Arduino, other than that, I have not poked further. My next step is to pull the front panel off, but before I start that, I wanted to check if anyone had a different suggestion.

Thanks.
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1036 on: August 15, 2018, 03:57:10 pm »
Did you managed to establish remote connection? If yes, check if it response to SCPI messages such as *IDN? or *TST?

Offline kirby

  • Newbie
  • Posts: 6
  • Country: us
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1037 on: August 15, 2018, 06:00:19 pm »
I hope this is me who should reply!

I never got any response from the Due if it is installed in the power supply. I upgraded the firmware while the Due was stand-alone.

However, I just tried to see if I got a response from plugging the USB cable from my computer directly to the Due while it was in-situ. The system powered up, went through the Ethernet config (which failed because it wasn't connected) and gave a 'Fan Error'. Same behavior with either the native or programming port.

If I plug a known good USB cable into the back connector and put the black cable into the native Due port, which IIRC is the proper port, I get the same 'display flash but otherwise not working' behavior.


 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1038 on: August 15, 2018, 06:30:57 pm »
Where you got that "fan error" message? On remote console (i.e. Arduino serial monitor or similar terminal console application)?

Offline kirby

  • Newbie
  • Posts: 6
  • Country: us
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1039 on: August 15, 2018, 07:37:35 pm »
Where you got that "fan error" message? On remote console (i.e. Arduino serial monitor or similar terminal console application)?
It showed up on the LCD which works fine when a USB cable from my laptop is plugged into the Arduino.
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1040 on: August 15, 2018, 07:44:58 pm »
Ok, we are progressing here :). If you got "fan error" then you shouldn't hear shortly after power up that fan is switched on and off. One possibility for that is failure in Vigortronix 5/12V AC-DC module on the AUX PS board or that fan is stalled due to some (electro)mechanical failure.
If fan sink too much current that will also drop 5 V down to nothing and your Due cannot be powered properly any more, hence you got just a flash on the screen. When you connect Due with USB cable directly, then it is powered from your PC and you can see that error message.
Please first check Vigortronix outputs (5V and 12V).

Offline kirby

  • Newbie
  • Posts: 6
  • Country: us
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1041 on: August 16, 2018, 10:25:02 am »
Please first check Vigortronix outputs (5V and 12V).
Spot on! With the Aux power board disconnected from everything except the AC mains, the output of the 5V side is 3.7 volts, and the 12V side is 0.

I am off to Digi-Key.

Thank you!
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1042 on: August 16, 2018, 10:34:46 am »
Hm, how frustrating. That component cost a fortune (see Newark or Farnell) and is not a "no name" AC/DC module, and so far it is the only component that few people reported as damaged.

Offline kirby

  • Newbie
  • Posts: 6
  • Country: us
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1043 on: August 16, 2018, 11:31:13 am »
Yup, it seems to be only available from Newark/Farnell in the UK. However, even with the additional freight charges, the Vigortronix is cheaper that the equivalent US-sourced part. Plus, the US-sourced part would need to be wired in rather than fit the footprint. I ordered an extra one, just in case. If somebody from the US is browsing this and needs one, I will send it to them at cost ($23.58) plus US postage if it is available.

You did good with this power supply. I don't use 80% of it's capability, but for what I need, it's the cat's pyjamas.

Thanks for taking time to help me with this.
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1044 on: August 16, 2018, 11:35:03 am »
Thanks for feedback about usage. I'd like to know why you're using 20% of its capabilities: because you don't need more or you don't know how to use them?

Offline kirby

  • Newbie
  • Posts: 6
  • Country: us
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1045 on: August 16, 2018, 12:42:08 pm »
Thanks for feedback about usage. I'd like to know why you're using 20% of its capabilities: because you don't need more or you don't know how to use them?

I don't need them. My main use is to safely keep me from doing stupid stuff. I can set the current limit on a channel to a safe level and not worry (as much)! The more advanced capability (serial/parallel combinations, remote sense, SCPI, etc.) are there if I need them.
 
The following users thanked this post: prasimix

Offline Pjoms

  • Regular Contributor
  • *
  • Posts: 51
  • Country: se
Rotary encoder issue
« Reply #1046 on: August 22, 2018, 06:21:56 pm »
I have noticed a quite scary issue with the rotary encoder twice the last days.
Both times I was increasing the voltage by one step, but the voltage raised 100 times that increment!

Yesterday I was in step mode set at 0.1V but the voltage raised 10V. Channel was on.

Today I was at the “normal” home screen and the voltage raised 1V instead of 10mV. Channel was off.

Unfortunately I have not find any way to reproduce it...
- Crowd campaign unit
- Latest fw 1.1.1 with the “SD-card fix”.

 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1047 on: August 23, 2018, 05:50:06 am »
Thanks for that. I can confirm that such bug in firmware exist. I still don't know when it can exactly happen neither find a time to investigate it in more details. Please report it on GitHub that we can track it.

Offline Pjoms

  • Regular Contributor
  • *
  • Posts: 51
  • Country: se
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1048 on: August 23, 2018, 06:18:16 am »
Done. #210

This is just a long shot thought, but as I will remember the psu had been on but untouched for some time before both of my situations. Can it be something with the velocity control and the timers around that causing it to jump that exact range of 100 times one increment?
 
The following users thanked this post: prasimix

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #1049 on: August 23, 2018, 10:02:24 am »
I can confirm that bug can be reproduced if after startup you wait for some time (more then few minutes less then an hour) and first increment will be 100x higher. For example, if initial voltage is set to 12 V, first increment will be 13 V instead of 12.01 V. It's possible that is somehow connected with timer or counter "overflow". We'll continue to work on it.


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf