How to increase thermal shutdown from 50ºC to 60ºC for the Radioberry Preamp?

156 views
Skip to first unread message

Samuel Lourenço

unread,
Nov 4, 2024, 6:18:32 PM11/4/24
to Radioberry
Hi,

I've modified the Radioberry Preamp, adding two aluminum bars as head spreaders and an heatsink. Although the thermal performance improved a lot, it still shuts down after a while. However, it seems that I have the conditions to increase thermal shutdown, as thermal conductivity between the output transistors, and from those to the thermal sensor is much improved.

The question is, how I can increase the thermal shutdown to 60ºC? Also, what compiler tools are required? What code should be changed? How can I recompile?

Kind regards, Samuel Lourenço

Yado-san

unread,
Nov 8, 2024, 8:57:26 PM11/8/24
to Radioberry
Hi Samuel,

I would recommend using a suitable heatsink, but I also changed the firmware temporarily.


* The compile and install process is described in radioberry_install.sh

https://github.com/pa3gsb/Radioberry-2.x/blob/master/SBC/rpi-4/releases/dev/radioberry_install.sh#L127

sudo systemctl stop radioberry
git clone  --depth=1 https://github.com/pa3gsb/Radioberry-2.x
cd Radioberry-2.x/SBC/rpi-4/device_driver/firmware
sudo make
sudo cp radioberry  /usr/local/bin


* The temperature settings are in firmware/radioberry.c

https://github.com/pa3gsb/Radioberry-2.x/blob/master/SBC/rpi-4/device_driver/firmware/radioberry.c#L501
line 501-521  ( 60C: 1256 -> 1382 )

static void *rb_measure_thread(void *arg) {
// temperature == (((T*.01)+.5)/3.26)*4096   if pa temperature > 50C (=1256) switch pa off! (pa_temp_ok)
int measured_temp_ok_count = 0;
while(1) {
sem_wait(&i2c_meas);
if (i2c_measure_module_active) read_I2C_measure(&pa_current, &pa_temp);
if (pa_temp_ok && (pa_temp >= 1256)) {
fprintf(stderr, "ALERT: temperature of PA is higher than 50ºC; PA will be switched off! \n");
pa_temp_ok = 0;
}
// PA recovery after high temperature; switch PA on again if PA temp is in range for 10 seconds.
if (!pa_temp_ok && (pa_temp < 1256)) measured_temp_ok_count++;
if (measured_temp_ok_count == 100) {
measured_temp_ok_count = 0;
pa_temp_ok = 1;
fprintf(stderr, "PA temperature is ok; PA can be used! \n");
}
}
fprintf(stderr,"rb_measure_thread: exiting\n");
return NULL;
}

hope this helps
Yado-san, jg1twp

2024年11月5日火曜日 8:18:32 UTC+9 Samuel Lourenço:

Samuel Lourenço

unread,
Nov 9, 2024, 11:15:02 AM11/9/24
to Radioberry
Thanks, Yado-san,

I was under the impression that the steps required to install Radioberry from Git would already require the compilation of source code, but I have to revisit that and do as you indicated. I'll also have to clone the SD card and try that.

As for suitable heatsinks, do you have any suggestions? I've had to makeshift one, as you read, but although the improvements were clear, it is still underwhelming.

Kind regards, Samuel Lourenço

Joey G. (Bonehead)

unread,
Nov 9, 2024, 8:23:07 PM11/9/24
to Radioberry
Has anyone tried putting bigger transistors to see if they are better at handling the heat?  I have put heat sinks on mine as well but still get the temp shut down.  It doesn't trasnmit long enough to do a FT8 transmission.  I have not changed the temp settings though.

wim

unread,
Nov 10, 2024, 3:08:41 AM11/10/24
to radio...@googlegroups.com



Van:
Verzonden: 10 november 2024 09:03:07 CET
Aan:
Onderwerp:
Re: How to increase thermal shutdown from 50ºC to 60ºC for the Radioberry Preamp?


I think I have a Solid sulotion for it.
See the pictures of mij pre amp.
Never have any breakdown by temp.
Wim
PE1PWR
73


"Joey G. (Bonehead)" <jcgre...@gmail.com> schreef op 10 november 2024 02:23:06 CET:
wim
wim
IMG_20241110_085950.jpg
IMG_20241110_085930.jpg

Wim PE1PWR

unread,
Nov 10, 2024, 3:11:06 AM11/10/24
to Radioberry
I have done this solution.
My pre amp works normal without temp problem.
See pictures.
Wim
PE1PWR 
73


Op zondag 10 november 2024 om 02:23:07 UTC+1 schreef Joey G. (Bonehead):
IMG_20241110_085930.jpg
IMG_20241110_085950.jpg

Joey G. (Bonehead)

unread,
Oct 13, 2025, 9:52:50 PMOct 13
to Radioberry
I know this is an old thread but I'm revisting the 5w amp hat and do have a good chunk of aluminum on the two output transistors but was curious, what componant actually reports the heat?  There is a small transistor next to the output transistors that gets really hot during transmit and have been playing around with heatsinking that one but was curious, what componant is actually reporting the heat because I think it's getting false readings.  The output transistors are not getting that warm.
Reply all
Reply to author
Forward
0 new messages