FNIRSI GC-01: Power consumption

[1MegaSaxoN]

hi folks, i have installed Gissio's radpro firmware in my unit, but due to poor battery life i am having, i want to go back to the original firmware, however the downloded 1.5 or the 1.6 files are way to big to flash back on, i cannot find the 48k files, my unit has the apm32 mcu with the j613 tube. please help!

[alfmck]

My GC-01 uses CH32 mcu and original firmware files flashes OK. I tried to measure the currency the device uses and was surprised that no matter if the radpro or original firmware loaded, the monitor is in sleeping mode or not, device uses 200 mA.

So I'm thinking about replacing the battery to more powerful one. The problem is that on the original battery are circuits that seems for controlling charging process.

Coming back to original software isn't a good solution, GC-01 with radpro convert it from cheap 5 minutes detector into 24 hours registrator with the possibility of PC connecting for data logging. You can also use any another tube, programming to it's specifications and then calibrate GC-01. Thank Gissio for Your radpro. I was very unsatisfied with the original soft but now I have what I wanted...

[1MegaSaxoN]

i wonder if anyone has successfully upgraded the battery yet? cuz that would be awesome!

[Gissio]

1MegaSaxoN:

Regarding your first question. On GC-01's with a WCH processor, power consumption of the original firmware is:

• 100 mA@battery with display on (max brightness). Since the battery is 1100 mAh, it should last for 11 hours in this case.
• 53 mA@battery with display off. It should last for 20 hours in this case.

Rad Pro requires 5 to 10 mA less than that, so it should be slightly better.

alfmck reports that the power consumption is 200 mA on GC-01's with a Geehy processor, which should last for 5 hours. It would be interesting to know how this was measured (display on, max brightness?). It would also be interesting to know what power consumption is with the display off.

Regarding your second question. Are you using Windows to upload the file? If you are unable to upload the firmware through the USB drive, in a few days I'll post instructions for flashing the firmware with a USB dongle.

[1MegaSaxoN]

i found the proper firmware files, all good.. even though i flashed it right back to the radpro firmware because the unit was a piece of shit with the original firmware... lol going to change the Geiger miller tube next to a j321 form the stock j613 junk that is in it now.. i am still interested in knowing if anyone has been able to install a better battery?

[Gissio]

According to my knowledge, it is just a matter of replacing the large battery with a 3.7 V Li-Ion battery of more capacity.

There is also another way of reducing power consumption, but this requires help from volunteers who have some experience in electronics. In Rad Pro, you can set up custom HV profiles to lower power consumption, at the expense of more fluctuation in the high voltage (and thus measurement error). It would be a matter of reducing the PWM frequency and finding the PWM duty cycle that produces the same voltage as the original firmware (440 V). To do this, you NEED TO MEASURE the high voltage at the tube. Instructions for doing this are published here: https://github.com/Gissio/radpro/blob/main/docs/users.md#hv-profiles

I am not responsible if you break your device from overvoltage or overcurrent!

[dc1rdb]

Here are some current consumption measurements (median values, because they obviously fluctuate a little) of my GC-01 with Geehy processor, J321 tube and RadPro 2.0rc2.

1. HV profile factory default
   Display high 80 mA
   Display med 70 mA
   Display sleep 50 mA

2. HV profile custom: 2,5 kHz / 10 %
   Display high 60 mA
   Display med 50 mA
   Display sleep 30 mA

[Gissio]

Seems to match the other report for devices with a WCH processor. Thanks for your efforts!

It's also quite interesting that you measured the power consumption with 10 % PWM duty cycle @ 2.5 kHz! Does the tube tick at that setting? Were you able to measure the high voltage at the tube? If it is around 400 V, I can add an "Energy saving" HV profile with these settings.

It would also be necessary for someone with a WCH processor to find the optimum power saving profile: the WCH and Geehy boards are different.

[dc1rdb]

Regarding the HV settings, I am still experimenting to find the perfect setting for me. I am measuring the voltage across the geiger tube with a Fluke 80K-40 HV probe.

Since the HV supply of the GC-01 is unregulated (no feedback loop), the key is not to exceed the voltage at low count rates, but have enough reserve for measuring higher activites. So it is always a compromise, depending on the use case.

10% @ 2.5 kHz actually results in overvoltage at my unit. Currently, I'm at 1.5 % @ 5 kHz, which on my unit results in around 440 V at background count rates and decreases to 400 V at activities around 5 uSv/h.

Please keep in mind that there seem to be quite a few different hardware revisions out on the market, so your voltages may differ from mine.

[Gissio]

You are absolutely right.

But the idea of an "Energy saving" HV profile to use less power for background level measurements is not a bad one. We actually did this for the Bosean FS-600/FS-1000 and FS2011, and it performs quite well. There is, however, one caveat: these devices come with Zener diodes that cap the output to 440 V max, so there is no danger of overvoltage. The GC-01 does not seem to come with these.

[Gissio]

Could you measure the frequency-duty cycle pairs for the factory-default voltage? From this I could derive an "Optimized" profile which uses less power but provides the same performance as the factory default.

I've already added the 2.5 kHz/1.5 % pair as "energy-saving" profile. It would also be great if you could measure this and ensure it produces 440 V @ background, both on WCH and Geehy variants.

[alfmck]

Here are my GC-01 measurements of frequency-duty cycle pairs, after replacing HH614 to J321 and R38-51mOm not placed.
Next week I'll have GC-01 with originally installed J321 and maybe repeat measurements.
HV settings.xlsx