switch on load when asking to set voltage in calibration steps

[gunthervermeir]

this is basically a power supply issue or leads voltage drop)

when the voltage is asked no load is applied, if in the voltage is then set to the 'no load value" and for the second step the actual voltage drops then it seams the "low" values for the load (20 mA etc) are less accurate and "catch up" when going to higher loads (if you test the RE:load after calibration)

the 1.9 firmware solves this to a degree but not completely

sugestion : activate the load to 2 amp when asking 'set voltage' ( added value: if you have only one decent multimeter handy you can better check the voltage the ps gives at the RE:load binding posts and that then takes also into account lead loss , then pull it out and set it in series to check the amp in the next step)

[Arachnid]

I'm not entirely sure I follow. Are you talking about voltage measurement inaccuracies or current measurement inaccuracies at the low end?

[gunthervermeir]

voltage inaccurate because there is no load in the "set voltage" step hence the "set voltage" step "sees" a to high voltage compared to the voltage on the Re:load binding posts in the "set amp' step

can be the ps dropping a bit or the leads dropping some voltage due current pulling the 2 amp load when setting "set voltage" takes this better into account

i'm not alone , see Preben Madsen in the comment on http://www.arachnidlabs.com/reload-pro/firmware.html

result seams to be that the "set" current differences from the "actual current" for low amp range

[Arachnid]

1.9 fixes this by accounting for the current draw when displaying the voltage. If it's still off by a bit, the constant just needs tweaking.