Improving Adjustable Dual Voltage Bench Power Supply

[Frankentronics]

Now I hate myself for buying those 6 additional panel meters. It would obviously be better to have 4 digit displays.

Do you think those blue LEDs could easily be changed to green ones? Or, since this seems to be a new model, perhaps it's just a matter of time before the red/green option are available for the 10A model.

[xavier60]

Now I hate myself for buying those 6 additional panel meters. It would obviously be better to have 4 digit displays.

Do you think those blue LEDs could easily be changed to green ones? Or, since this seems to be a new model, perhaps it's just a matter of time before the red/green option are available for the 10A model.

Although I haven't searched, those LED displays would be easily available. When buying any 7 segment LED display, make sure to get the correct wiring configuration. They can be either common anode or common cathode in these meters. Ill look into it.
I have been swapping the red and green because I want the colours to match the CV and CC LEDs on my PSU. It can be tricky getting them back to the same board height because they hold the filter window in place.

[1001]

I bought two panel meters on aliexpress, they have one more chip, it seems to me to read 2057T, 6 pin. does anyone know which component it is?

[xavier60]

I bought two panel meters on aliexpress, they have one more chip, it seems to me to read 2057T, 6 pin. does anyone know which component it is?

A 5mΩ shunt amplified by an LM358 opamp would normally have too much drift to display a useful reading on a 4 digit display.
The 2057T would be some sort of CMOS analog switch controlled by the micro-controller to periodically disconnect the current shunt from the input to the opamp. During this time, the micro-controller would note the opamp's output voltage as corresponding to zero reference.
I have yet to identify the past's full part number.
 

[1001]

tnx a lot, could this be it?

[xavier60]

tnx a lot, could this be it?

Yes. Makes sense being double-throw so that either shunt or ground can be alternately selected for the opamp's input.

[1001]

ledtester asked to me to probe the "Digital Control Pin" of the RS2057, here are the pictures

[xavier60]

I found the same looking meter in my collection. If yours also has the same slow 1 second display refresh, be sure to complain to the seller about it.
The ICs control input pin is driven from one of the display's digit enable lines. The step is due the way they have reduced the drive duty cycle to reduce LED brightness.
At 5 Amps, I see 20mV at pin 3 and a 0 to 20mV chopped signal at pin 4 which is then fed to the opamp.

[1001]

 Specifications from seller

[xavier60]

Mine was advertised as 300mS refresh also.

[David Hess]

Digital panel meters typically operate at a sample rate of a multiple of 100 milliseconds so that they can reject 50 and 60 Hz noise and harmonics, but 100 milliseconds is too fast to follow so they update slower.

[xavier60]

These panel meters used to have satisfactory refresh rates. I have some 3 digit types that are 500mS.
Some I got from BangBad last year were 2000mS although they were advertised as 300mS. I sent them all back.
The only reason for slowing the refresh will be a stupid one.

[David Hess]

These panel meters used to have satisfactory refresh rates. I have some 3 digit types that are 500mS.
Some I got from BangBad last year were 2000mS although they were advertised as 300mS. I sent them all back.
The only reason for slowing the refresh will be a stupid one.

I remember panel meters from 20+ years ago where the update rate was adjustable, either with an RC circuit or by jumpers, but they were usually never faster than 100 milliseconds for the reason I gave.  The exceptions were meters that only reject 50 or 60 Hz so they did not have to rely on the least common denominator of 50 and 60 Hz to cover both.

I guess we need a good "universal" panel meter kit, but it is difficult to make such a thing economical.