DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)

[prasimix]

Prasimix, is there a list to join just in case you do end up deciding to do another run?

Crowd Supply announced that they are going to take care of second batch. I have no details about it then to say that they (as anyone else!) is more then welcome to do that and can count on my support. Anyone still interested could subscribe to campaign to get notifications in future.

[prasimix]

I just finished uploading of all files for next revision r5B13 that is addressing few issues reported from few backers (i.e. USB isolator speed issue). List of fixed issues are available here. I'll make it as new release in coming days if no issues are found.

Please note that r5B12 that is used for crowdfunding revision is removed from GitHub master branch but, of course will remain accessible through Releases page.

[prasimix]

I'm working on specification for implementing data logging/acquisition of the channels outputs that is planned for forthcoming firmware v1.1 (after v1.02 that is addressing few minor issues). Idea is to provide the following functionality:

• Recording two (out of three possible, i.e. VOLTage, CURRent or POWer) output values. Actually, if that makes any sense, we could also offer logging of temperature sensors.
• Source selection (CH1 or CH2) for each destination. If channels are coupled in SERies or PARallel, only one destination is allowed
• Destination could be SD card file, which max. size is only limited with SD card capacity (i.e. installed 8 Gb, or more if you replace it with bigger one) or data stream available over serial (USB) or Ethernet port
• Sampling rate per second or hopefully in hundreds or tens of miliseconds
• Definition of data logging duration time in seconds, and possibly in file size (e.g. until size is smaller then 100 Mb)
• Data logging start event: output is enabled, digital input pin, software TRIGger, start date/time, SCPI (remote) command
• Data logging stop event: defined duration time is expired, output is disabled, protection trip (i.e. OCP, OVP, OPP or OTP), digital input pin, stop date/time, SCPI (remote) command, destination file size
• Definition of data format, e.g. comma-separated values, binary/"raw", etc.
• Definition of data structure, for example (not all items are mandatory):
• Data format info (e.g. version, type)
• Power supply (source) info like model, firmware
• Channel's settings: e.g. set voltage, current, power range, enabled protections, calibration remark, etc.
• Start of logging event type (e.g. output is enabled, digital input pin, ...)
• Time stamp, start of logging line/mark
• Time stamp, measured value 1, measured value 2 (this is the most important item!)
• Time stamp, end of logging line/mark
• End of logging reason (e.g. OCP tripped, output is disabled, digital input pin, ...)
• Total duration in seconds
• Energy delivered in Ah and Wh
• User remark/note (defined on start of data logging session)

The latest two bullets are where I need your assistance. I cannot found any data format and structure used by commercial vendors. All what I found until now is SCPI command set that they are using to perform "internal" and/or "external" data logging. For example, Rigol has RECorder SCPI subsystem and related GUI options. Keysight has ELOG (from "external" logging) subsystem but without single word about how logged data are formatted. If anyone have power supply from mentioned or any other vendor and use data logging, please let me know. We can continue independently, but it could be nice if we can follow some existing format. Hopefully in that case other software for logged data presentation can be used.

[jleg]

sounds promising! In case you didn't already find this - the specs for Rigol's recording format are here:
https://www.eevblog.com/forum/testgear/rigol-dp832-convert-the-rof-files-(recorded-data)/msg488462/#msg488462

[prasimix]

sounds promising! In case you didn't already find this - the specs for Rigol's recording format are here:
https://www.eevblog.com/forum/testgear/rigol-dp832-convert-the-rof-files-(recorded-data)/msg488462/#msg488462

Great, that is a good start.

[prasimix]

A new firmware minor release v1.02 is now available on the GitHub. Nothing spectacular can be found there, except maybe incorrect displaying of values below 1, in YT-view mode (issue #170).

[prasimix]

There is a few possible candidates for next firmware release (planned v1.1) and I'd like to hear your feedback on them or, if you found it useful just like/vote for it on the Github:

• #66 Adaptive protections
• #67 Output voltage offset when RSENse is not active
• #89 Measure output voltage before output is switched on

We also spent some time on #165 (Prepare .HEX/.bin for direct firmware upload without Arduino IDE) but without working solution that could be simpler then installing and using Arduino IDE to compile and upload firmware. I'd like also to ask if someone knows about possible solution for #96 (Arduino start-up 10 ms transition) which is really frustrating and makes AC soft-start unusable in case when PSU is powered up mechanically (i.e. front standby or rear power switch).

Of course, feel free to suggest new functionality or improvement of what you already experienced with your PSU (please check first if such thing is already reported).

[Pjoms]

I have some thoughts and wishes about the firmware.

1.
Then changing values in step mode the size or “gain” of the steps follows between different settings.
E.g. I choose to change the voltage in 0.5V steps (next lowest step size) and then choose to change the current in 250mA steps (next highest). If I then go back to set the voltage again, the step size now is 2V instead of 0.5V as it keeps the setting to the next highest value.
This may be a little surprising, especially then the rotary encoder is used.
Maybe the step size can be remembered for each value, or always be set to the lowest value then entering a new screen?

2.
The indication for the currently selected current range flicker between 500mA and 5A if the current limit is changed with the rotary encoder in the range under 500mA.
I don't know if this is just a graphical fenomen or if the range actual is changing.

3.
I miss the opportunity to simultanly switch both channels on and off with a single press then they are in “normal” uncoupled mode.

4.
I miss the opportunity to switch on and off each channel with an external signal, e.g. a physical switch or button then you don’t can/want to look at the psu.
Maybe the trigger input can be used?

[prasimix]

1. Then changing values in step mode the size or “gain” of the steps follows between different settings.
E.g. I choose to change the voltage in 0.5V steps (next lowest step size) and then choose to change the current in 250mA steps (next highest). If I then go back to set the voltage again, the step size now is 2V instead of 0.5V as it keeps the setting to the next highest value.
This may be a little surprising, especially then the rotary encoder is used.
Maybe the step size can be remembered for each value, or always be set to the lowest value then entering a new screen?

Yes, I'm aware of that, but if was never implemented. Please create a new issue for it that is not forgot once again.

2. The indication for the currently selected current range flicker between 500mA and 5A if the current limit is changed with the rotary encoder in the range under 500mA.
I don't know if this is just a graphical fenomen or if the range actual is changing.

You're right. Maybe doesn't looks nice but it just show that autoranging is doing its job! It's happen when current range is set to Best (default), Autoranging is ON and without load. When you connect load it will stop switching current ranges from low to high over 0.5 A.

3. I miss the opportunity to simultanly switch both channels on and off with a single press then they are in “normal” uncoupled mode.

You have two possibility to do that when channels are not coupled: using trigger step or put channels in tracking mode. Former allows you to set different output voltage and current, in later case they will be the same.

4. I miss the opportunity to switch on and off each channel with an external signal, e.g. a physical switch or button then you don’t can/want to look at the psu.
Maybe the trigger input can be used?

Yes, you can set PIN1 Function under System settings (1)... Digital I/O as Trigger input when channels are in step mode, or you can use Inhibit. For improved usage of the latest one you need firmware v1.02 (issue #166).

[Pjoms]

Yes, I'm aware of that, but if was never implemented. Please create a new issue for it that is not forgot once again.

Done. #176

You have two possibility to do that when channels are not coupled: using trigger step or put channels in tracking mode. Former allows you to set different output voltage and current, in later case they will be the same.

That feels a little bit complicated to be useful in practice.
E.g. if I set CH1=5V and CH2=12V and like to switch both channels on or off at the same time because the circuit under test needs the two supplies without “any” delay.

Yes, you can set PIN1 Function under System settings (1)... Digital I/O as Trigger input when channels are in step mode, or you can use Inhibit. For improved usage of the latest one you need firmware v1.02 (issue #166).

That could work, but will require changed setting to control the output either on or off with the trigger, if I understand correctly? Not possible to toggle the output that way?
I would like an easy accessible way to by a physical switch be able to control the output both on and off for each channel separately. 
I’m find it handy to be able to look at my circuit then I turn on the psu. It’s a little bit tricky to press the right spot on the touch screen while not looking. With the feel of a physic button it’s much easier.

[prasimix]

You have two possibility to do that when channels are not coupled: using trigger step or put channels in tracking mode. Former allows you to set different output voltage and current, in later case they will be the same.

That feels a little bit complicated to be useful in practice.
E.g. if I set CH1=5V and CH2=12V and like to switch both channels on or off at the same time because the circuit under test needs the two supplies without “any” delay.

In practice that means that for output enable you need two click (to initiate trigger) and for disable still just one since they are "coupled" with trigger. Current architecture do not allow simultaneous switching of both channels but sequential with delay well below 1 ms (about 260 us) as shown below. "Step" values of voltages and currents could be any combination.

Yes, you can set PIN1 Function under System settings (1)... Digital I/O as Trigger input when channels are in step mode, or you can use Inhibit. For improved usage of the latest one you need firmware v1.02 (issue #166).

That could work, but will require changed setting to control the output either on or off with the trigger, if I understand correctly? Not possible to toggle the output that way?
I would like an easy accessible way to by a physical switch be able to control the output both on and off for each channel separately. 
I’m find it handy to be able to look at my circuit then I turn on the psu. It’s a little bit tricky to press the right spot on the touch screen while not looking. With the feel of a physic button it’s much easier.

The easiest way is to define PIN1 function as Inhibit. In that case you can for example connect foot pedal to front panel CH1's push-in connector pin 7 and 8 and control both channel outputs simultaneously!

[texaspyro]

We also spent some time on #165 (Prepare .HEX/.bin for direct firmware upload without Arduino IDE) but without working solution that could be simpler then installing and using Arduino IDE to compile and upload firmware.

I was hoping that we could use XLOADER to download .hex files to the power supply,  but it looks like XLOADER does not support the Due.  BTW, one feature that would be nice to add to XLOADER is showing the progress of the upload.  Currently it uploads "blindly" and uploads can take a lot of time.  Without a progress bar or indication you don't know anything is happening until / if the upload completes.

Perhaps the author of XLOADER can be persuaded to part with the source code or modify it to support the Due...

[texaspyro]

Here's a thread that talks about adding an XLOADER type program to the Arduino package.  It mentions there is a devices file in XLOADER that might be able to be used to add due support:

http://forum.arduino.cc/index.php?topic=225263.0

[texaspyro]

Another possible staring point:

https://github.com/TheokieZA/TRUpdater

[prasimix]

It seems that both XLOADER and TRUploader are using avrdude that is also used by Arduino IDE when AVR (8-bit) boards are in question. For Due, Arduino IDE is using bossac, and it works from command line but cannot work alone, you need before uploading to reset serial port that bossac don't know how to do.

[Pjoms]

What’s the plan for the temp sensor on the push-in connectors?
As discussed earlier a useful battery charging funktionalley could be a little tricky to implement well.
One simple function meanwhile could be just an external over temp protection.
I found it useful sometimes to have an “emergency stop” for high temp on e.g a heatsink.

[prasimix]

It's planned for next firmware release v1.1 (issue #143) but still without clear idea what to do with it (hence no description there yet). Feel free to add any suggestion directly into mentioned task.

[Pjoms]

Added to #143

Idea for temp sensor functionally: External overheat protection.
Protection can be set in two levels: Warning and Trip.

Regular overtemp protection:
Warning or Trip occurs then temp goes over setpoints.

Delta T protection:
Warning or Trip occurs then temp rises faster than setpoints.

[cs.dk]

Don't know if it's a bug or not.
If i couple the supply in series @ 57V/5A, and start to charge my forklift batteries, the amps and powermeter are displaying the actual output. But the voltage displays 57.00V - When going to settings > Over voltage protection, then the real value like ie. 54.60V is displayed.
I expected the display to show the measured output, not the programmed setting. (Newest firmware, downloaded yesterday)

[prasimix]

There is two possiblity why you have such high difference (when both channels are calibrated):

• PSU enters CC mode
• Your cables/connections are so bad.

Problem with second one is that you cannot benefit from remote sensing when outputs are coupled in series.

[cs.dk]

Yes it is in CC mode. The large batterypack can take much more current.
But even in CC mode, I expected to see the actual output voltage.

EDIT: I did not use remote sense.

[prasimix]

Think once again, you're in CC mode, your output voltage could be anything from almost 0 to programmed voltage (i.e. 57 V in your case).

[cs.dk]

Yeah I know. But why would you not display the real output voltage?
I've got some cheap Tenma's and some more expensive EMI, TDK Lambda, etc. They all show the actual output voltage, also in CC mode. I really thought all PSU's were like that.

[prasimix]

OK, once again, since it seems that I missed something: outputs are coupled in series, Vset = 57 V, Iset = 5 A, and when battery is connected, PSU enters CC mode and you can see Vout that is lower then 57 V, or not?

[cs.dk]

OK, once again, since it seems that I missed something: outputs are coupled in series, Vset = 57 V, Iset = 5 A, and when battery is connected, PSU enters CC mode and you can see Vout that is lower then 57 V, or not?

Maybe a language barrier, English isn't my native
Yes, series coupling, Vset = 57V, Iset = 5A, when connected, the PSU enters CC. I can see 57V on the main screen, even though the output voltage is lower. However, if I go to Settings > OVP and Voltage limit, then I can see the actual output voltage in that menu.