Xiegu XPA125B 100W amplifier support

4,455 views
Skip to first unread message

scumball contractor

unread,
Jun 6, 2020, 9:51:01 AM6/6/20
to Hermes-Lite
Hi Group,

It's cold and wet, so no farming today and time to catch up on things. I came across this 100W amplified and after reading up on it, I decided to buy it. It's more money than I would like to spend, reality has to kick in and I can't get time to build something from all the bits I have gather up over the years. Anyway I can sell a cow.......


For automatic band change, it uses a voltage to select the frequency range. I think this is similar Yaesu (They do mention the FT817) so that got me think on how to interface to the HL2.

The most obvious solution would be to add an I2C DAC and feed it's output to the amp. The other option is to create a PWM voltage output using a spare pin on the FPGA. I'm not sure what is available but CN9 look like it wasn't doing much. It's output could be taken to an end plate and filtered with a simple RC filter. The max voltage required is less than the 3.3V the FPGA can provide.

I can modify PowerSDR to output the required signals over the protocol and other software providers could add the functionality as well. I think this would be a useful addition to have as the interface is common to other bits of kit.

Cheers

Reid
Gi8TME/Mi0BOT

PS. If you know that this amp is S%^&, shout now and I can cancel the order .We could still do the interface if other people require it.

James Ahlstrom

unread,
Jun 6, 2020, 10:16:48 AM6/6/20
to Hermes-Lite
Hello Reid,

I think the best approach is to use I2C. There are many amps out there and trying to support them all with gateware changes is going to be troublesome. There are end plates that support some amps, but I have lost track of all the variations. Every amp seems to be different. There is already support in the gateware to send arbitrary I2C bytes.

Jim
N2ADR

Alex Pressl

unread,
Jun 7, 2020, 12:14:00 PM6/7/20
to Hermes-Lite
I also expect you can do that without any change to the system. I made some experiments to interface my homebrew amp and can see its working out of the box.

Just get yourself any arduino, smallest one is fine. I used this guy: https://store.arduino.cc/arduino-nano-every

your volatge on that interface requires 2.5 or 5V I think maximum scale works for the arduino at 5V. 

listen to the I2C bitmask you configured in SDR console or Quisk you are free to send what you like from there through the HL2 to your interface module to the bus.


Greetings
Alex

Alex Pressl

unread,
Jun 7, 2020, 9:09:27 PM6/7/20
to Hermes-Lite
Hello Probir,

Also Lou asked for where to hook up the bus in HL2. I made an effort to get you a minimal code sample and a schematic. I also tried the analog out thing for the XIegu Amp as I didnt try that really before, the value is sufficiently stable for that purpose with a simple Lowpass on its output. One line of code extra....

see the schema here: https://bit.ly/3h4oTr1


well, I would like to connect it to DB13, but with the LPF connector attached, I dont see space for this on the board therefore see the picture I simply doubled up the connector and used some other ground. this is now listening in parallel to the LPF on address 0x20.

if you use SDR console and roll through the ham bands you will see on entry and exit a switch as described in the SDR console instruction page. 

to switch off the LPF out of band is quite a risk in my eyes. my PA would be toast if I would transmit into a LPF. Normally it should swich as the LPF is designed. i.e. everything below the upper boundary frequency is fine to send. Even so, I am using marine frequencies which are of course outside the Ham bands.

But this is purely SDR console special interpretation for explicit this very special LPF filter bank.I assume and nothing HL2 is doing.

I need time to try other stuff like Quisk and see which desktop tool works for me. maybe the pihpsdr is the best for me (no limits to tweak:)

Greetings
Alex

Prabir Debnath

Sun, Jun 7, 6:30 PM (8 hours ago)
to me
Can you Please post the block diagram and the connection details as I am not able to understand the interfacing with the nano to switch bands and what about the code. Kindly post the detailed diagram in the forum as many will be benefited

Probir VU2BQF 
2020-06-08_02-44-07.png

Steve Haynal

unread,
Jun 8, 2020, 12:25:33 AM6/8/20
to Hermes-Lite
Hi All,

I agree with Jim and Alex on this. Your best bet is a small MCU with DAC or PWM output plus enough pins and nice library to be a i2c slave. You should be able to find a complete small board for under $5. Unfortunately the FPGA pin connected ton CN9 is input only and can't be used as an output.

Alex, there are two places on DB13 to connect to the i2c. One should be available even if a daughter board is plugged in. See SDA2 and SCL2 on the Rear IO strip and note that there are two through hole pads for each:

I've been tempted by this PA:
Another 9 dB on TX would really help me with FT-8. It also uses the voltage level band select. I'm not sure of the quality and filtering.

73,

Steve
kf7o

Reid Campbell

unread,
Jun 8, 2020, 3:41:55 AM6/8/20
to herme...@googlegroups.com
Everybody, thanks for all the feedback.


On 08/06/2020 05:25, Steve Haynal wrote:
Hi All,

I agree with Jim and Alex on this. Your best bet is a small MCU with DAC or PWM output plus enough pins and nice library to be a i2c slave. You should be able to find a complete small board for under $5. Unfortunately the FPGA pin connected ton CN9 is input only and can't be used as an output.

Ok, as there no available output I'll have to consider some of the options. I'll look at the current end plates and see if I can't do something there. Maybe bodge together a Arduino as I have some laying about.



Alex, there are two places on DB13 to connect to the i2c. One should be available even if a daughter board is plugged in. See SDA2 and SCL2 on the Rear IO strip and note that there are two through hole pads for each:

I've been tempted by this PA:
Another 9 dB on TX would really help me with FT-8. It also uses the voltage level band select. I'm not sure of the quality and filtering.

I looked at a few other amps and most are based of no name transistors, probably IRF510. The XPA125B uses proper 100W RF transistors. The interface options are a bit limit, no remote ATU tune start for one but it gets me going with more power.

I'll let you know how it goes.

Cheers

Reid
Gi8TME/Mi0BOT 

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/7c2a2310-0524-4c26-9e68-f315d959ce1eo%40googlegroups.com.

W. Jozef

unread,
Jun 8, 2020, 6:41:19 AM6/8/20
to Hermes-Lite

pinssdascl.jpg


Hi all,
Pins for i2C marked on picture. (I know..... little space but it fits 1 x 5 pin if you don't have to use other connections offered by DB13 2x5)

73, Joe
lb1hi

Lou Scalpati

unread,
Jun 8, 2020, 9:19:56 AM6/8/20
to W. Jozef, Hermes-Lite
Thanks!

73
Lou KI5FTY



--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

scumball contractor

unread,
Jun 16, 2020, 4:19:51 AM6/16/20
to Hermes-Lite
Just a quick update on the Xiegu XPA125B amp. I received it last week and used it for our club's weekly net. So far, so good. It's well constructed but you have to get used to the the button presses. I think the software is on a slow polling loop. I got good reports on the audio and I was getting about 80-90W for about 1W in. Well, that what the indications on the front panel of the amp says. I have the drive on PowerSDR set to 1 and that appears to be enough to produce nearly full power out.

I'm just using the buttons to select the band and do ATU tunes. Interface options are just PTT, Band voltage and ALC. I have requested more information from the UK supplier but I'm not holding my breath. There is a 3.5mm TRS connector and they supply a lead with a USB on it for PC to amp comms. They say it's for firmware updates but could be used for control in the future.

That's it so far. First thing is to sort out a band voltage, still thinking about the options.

Cheers

Reid
Gi8TME/Mi0BOT

Sid Boyce

unread,
Aug 3, 2020, 7:58:59 AM8/3/20
to herme...@googlegroups.com
Hi Reid,
Did you receive any further information?

I am now reviewing the posts on the topic.

I also wonder if anyone came up with an interface circuit diagram for HL2.
73 ... Sid.
> --
> You received this message because you are subscribed to the Google
> Groups "Hermes-Lite" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to hermes-lite...@googlegroups.com
> <mailto:hermes-lite...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/hermes-lite/8eedb3a7-6c69-437a-9cf1-90137c2aa8eeo%40googlegroups.com
> <https://groups.google.com/d/msgid/hermes-lite/8eedb3a7-6c69-437a-9cf1-90137c2aa8eeo%40googlegroups.com?utm_medium=email&utm_source=footer>.


--
Sid Boyce ... Hamradio License G3VBV, Licensed Private Pilot
Emeritus IBM/Amdahl Mainframes and Sun/Fujitsu Servers Tech Support
Senior Staff Specialist, Cricket Coach
Microsoft Windows Free Zone - Linux used for all Computing Tasks

Don [N5SKT]

unread,
Aug 3, 2020, 9:37:24 AM8/3/20
to Hermes-Lite
I have the MX-P50M and I am OK with setting the band switch myself. The automatic switching Amps tend to be a lot more expensive and I have a 100W K900 Amp on the way. BTW, on the MX-P50M, look at this: https://www.youtube.com/watch?v=1f3VL0C_cHY and also the follow up videos where he shows compression of the Amplifier. It is a decent 35W Amp but does splatter a bit at full power.

Best Regards,
Don

Reid Campbell

unread,
Aug 3, 2020, 10:23:46 AM8/3/20
to herme...@googlegroups.com
Hi Sid,

I tried the UK supplier, the EU supplier and direct to the manufacturer
and have received no replies. I found the FCC application which has a
request that the circuit would not be released, so I'm frustrated that
they only want to sell the amp with their own rigs. They also sell an
interface for the FT817 but I'm confused, as the UK site said it doesn't
do band selection but on other sites, it appears it does.

I have now set the drive control to zero for all bands. I found that 17m
& 12m was getting near 120W and the power out LCD was flashing. I also
read that unless you were using the amp with the suppliers own rigs,
power in should be less than 1W. The amp has an "ALC input" which I
think should be an "ALC ouput", well that how ALC should work with
feedback to the driving rig to control the power. Maybe I'm getting this
wrong?

I had a look at repurposing the PWM fan control as a band select
control. I got the Verilog written and it compiles but I need to do more
reading on how to simulate the new functionality. This is my first
venture into FPGA programming.

Maybe if other people requested information for the amp, the
manufacturer would realise that there is a demand but I wouldn't hold my
breadth.

Cheers

Reid
Gi8TME/Mi0BOT

Sid Boyce

unread,
Aug 3, 2020, 12:16:46 PM8/3/20
to herme...@googlegroups.com
Hi Reid,
Thanks for the update. Amazing the manufacturer who advertise use with
other QRP rigs is so reluctant to divulge details to a customer, either
that or they are just a sales outlet.

The manual appears to state a maximum of 5W input on CW, so I expected
at least that on SSB.


I'll get mine hooked up with a PTT switch wire to the Accessory port and
set the HL2 to zero.
73 ... Sid.

Robert Nickels

unread,
Aug 3, 2020, 12:34:58 PM8/3/20
to herme...@googlegroups.com
On 8/3/2020 11:16 AM, Sid Boyce wrote:
>
> The manual appears to state a maximum of 5W input on CW, so I expected
> at least that on SSB.

When it first came out a few of us hit them with the full output of the
X108G and they sounded horrible.  Then we characterized them by watching
the output on a wattmeter and scope and found that 2-3 watts was
sufficient to develop full output without excessive IMD and spurs.  The
amps "meter" is worthless and thus far I've not found any information
about how to calibrate it. Good hardware, poor customer service, and
non-existent technical documentation.    Let the buyer decide if that's
acceptable or not.

In fairness, Xiegu has been much more responsive to customers with the
SDR-based G90, in part I suspect because they now have the ability to
change things in software compared with the previous radio.   The G90
uses a QSD/QSE architecture and is a very good little radio with active
online forum support and periodic firmware updates.  The XPA-125 has a
serial connection that was touted as "for firmware updates" but as far
as I know there never have been any, nor any information about how to do
so.   I understand the manufacturer not wanting users to muck around
with firmware of an approved product, but to have limited configuration
capability such as calibration would be in everyone's interest.

73, Bob W9RAN

Reid Campbell

unread,
Aug 3, 2020, 3:31:37 PM8/3/20
to herme...@googlegroups.com
Hi Bob,

I'm running it with the lowest power I can (PowerSDR drive level 0) and
it is showing 80-90W on the output power LCD. At some stage I will get
out the old HP SA and have a more detail look.

If you hold the PA and Band buttons at power on, the serial line becomes
active and the current firmware version is show on the LCD screen. You
can erase the firmware at that point using the front panel buttons and
use Tera Tterm and xmodem to download the firmware. There is a
downloadable firmware version on Xiegu's website but appears to be just
the current version 1.0.0, even through it says 2.0.0 and no, I have
been brave enough to try an update.

There is a Facebook page for Xiegu but the main discussion is for the
transceivers but it's where I got the above info.

Cheers

Reid
Gi8TME/Mi0BOT

Matthew

unread,
Aug 3, 2020, 3:55:41 PM8/3/20
to Hermes-Lite
Hi Reid,

If while you have your SA out you could characterise the HL2 IMD at powers less than 5 W that would be great :)

Past discussions have commented on the unknown distortion properties of the HL2 at lower power. If people are going to start using the HL2 at 2 W and then amplifying to get very high powers it would be good to understand if people are amplifying an already distorted signal. The MRF101-HL2 PA has an attenuator for this reason, I would like to have the confidence to get rid of this attenuator.

Sadly I lack an SA for such tests and last time I looked the Rigol SA wasn't an attractive enough price for me.

73 Matthew M5EVT.

Reid Campbell

unread,
Aug 4, 2020, 3:39:50 PM8/4/20
to herme...@googlegroups.com
Hi Matthew,

I'll try and give it a go. It's an old HP 8558B of dubious calibration but will maybe provide a comparison if I use the same scales.

Cheers

Reid
Gi8TME/Mi0BOT
--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/86478a6c-8644-46ed-81a5-7514c41b0a1co%40googlegroups.com.

Reid Campbell

unread,
Aug 5, 2020, 10:23:40 AM8/5/20
to herme...@googlegroups.com
Hi Matthew,

Unfortunately the SA doesn't have the resolution to see the two tone IMD products but PowerSDR has a spectral display of the transmit signal. From what I can see the IMDs go down as you reduce the power.

I was also asked to characterise the drive slider of PowerSDR, which surprised me with result. Below is a table of drive readings when the power increased and the mapping on to the Tx attenuator. My PA power calibrations are set at 38.8.

0         -7.5dB  
1         -7.0dB
2         -6.5dB
4        -6.0dB
6        -5.5dB
10       -5.0dB
14      -4.5dB
18      -4.0dB 
24       -3.5dB
30        -3.0dB
37        -2.5dB
45        -2.0dB
53        -1.5dB
63        -1.0dB
72        -0.5dB
83         0.0dB

The other thing that I found out was mouse wheeling the drive slider to zero drive would result in no power but cycling the tune button while drive was zero would again produce power, so lowest power setting is drive zero.

Cheers

Reid
Gi8TME/Mi0BOT
--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/86478a6c-8644-46ed-81a5-7514c41b0a1co%40googlegroups.com.

"Christoph v. Wüllen"

unread,
Aug 5, 2020, 11:55:11 AM8/5/20
to herme...@googlegroups.com


> Am 05.08.2020 um 16:23 schrieb Reid Campbell <scumballc...@gmail.com>:
>
> Hi Matthew,
>
> Unfortunately the SA doesn't have the resolution to see the two tone IMD products but PowerSDR has a spectral display of the transmit signal. From what I can see the IMDs go down as you reduce the power.
>
> I was also asked to characterise the drive slider of PowerSDR, which surprised me with result. Below is a table of drive readings when the power increased and the mapping on to the Tx attenuator. My PA power calibrations are set at 38.8.
>
> 0 -7.5dB
> 1 -7.0dB
> 2 -6.5dB
> 4 -6.0dB
> 6 -5.5dB
> 10 -5.0dB
> 14 -4.5dB
> 18 -4.0dB
> 24 -3.5dB
> 30 -3.0dB
> 37 -2.5dB
> 45 -2.0dB
> 53 -1.5dB
> 63 -1.0dB
> 72 -0.5dB
> 83 0.0dB
>

This is more or less what I predicted. A drive level is calculated
from the slider position P via (at PA calibr. 38.8 dB)

D = Sqrt(P/94) * 255 if P<= 94, D = 255 if P >= 94

Now you take the upper 4 bits of D and map this *linearly* to the
attenuation -7.5 to 0 in 0.5 dB steps, and you are there (this means:
at P=82 you reach D=239 which maps to the second weak attenuation.
The non-linearity (e.g. moving slider from 10 to 30 but TX signal grows by 2 dB)
could be „repaired“ in the firmware but if there is interest, I could also
do this in the piHPSDR program. However, the „jumps“ will remain since this is
a property of the hardware.


P.S.: At least with piHPSDR, a PA calibration of 39.1 dB lets you use the „full“ drive scale,
that is, D goes from 254 to 255 when moving P from 99 to 100.

P.P.S: on my HL2, I have to use about 41-42 dB for PA calibration, such that the whole
Drive slider range only has about 5 dB. So moving the drive slider to zero still gives
1.7 Watt output, and I think this is the root of the problem with the xiegu PA.

"Christoph v. Wüllen"

unread,
Aug 5, 2020, 12:24:45 PM8/5/20
to herme...@googlegroups.com


> Am 05.08.2020 um 16:23 schrieb Reid Campbell <scumballc...@gmail.com>:
>
> Hi Matthew,
>
>
> I was also asked to characterise the drive slider of PowerSDR, which surprised me with result. Below is a table of drive readings when the power increased and the mapping on to the Tx attenuator. My PA power calibrations are set at 38.8.
>
> 0 -7.5dB
> 1 -7.0dB
> 2 -6.5dB
> 4 -6.0dB
> 6 -5.5dB
> 10 -5.0dB
> 14 -4.5dB
> 18 -4.0dB
> 24 -3.5dB
> 30 -3.0dB
> 37 -2.5dB
> 45 -2.0dB
> 53 -1.5dB
> 63 -1.0dB
> 72 -0.5dB
> 83 0.0dB
>

To give a linear scale on the range where is actually doing something,
all 8 bits of the HPSDR drive level should be taken into account in the
firmware, and should be mapped to DAC attenuations as follows:

0 - 110 -7.5 dB (in current firmware: 0 - 15)
111 - 117 -7.0 dB (in current firmware: 16 - 31)
128 - 124 -6.5 dB (in current firmware: 32 - 47)
125 - 131 -6.0 dB (in current firmware: 48 - 63)
132 - 139 -5.5 dB (in current firmware: 64 - 79)
140 - 147 -5.0 dB (in current firmware: 80 - 95)
148 - 156 -4.5 dB (in current firmware: 96 - 111)
157 - 165 -4.0 dB (in current firmware: 112 - 127)
166 - 175 -3.5 dB (in current firmware: 128 - 143)
166 - 185 -3.0 dB (in current firmware: 144 - 159)
186 - 196 -2.5 dB (in current firmware: 160 - 175)
197 - 208 -2.0 dB (in current firmware: 176 - 191)
209 - 220 -1.5 dB (in current firmware: 192 - 207)
221 - 233 -1.0 dB (in current firmware: 208 - 223)
234 - 247 -0.5 dB (in current firmware: 224 - 239)
248 - 255 0.0 dB (in current firmware: 240 - 255)

So the current firmware is non-linear, but you enjoy the whole
width of the slider. In both possible implementations, it is
not possible to smoothly reduce at small powers because of
hardware limitation,
and such a lookup table would probably cost
256 bytes in the FPGA so such a possible firmware change might
not be worth the effort.

For old Penelope boards, the TX drive is regulated by scaling the
TX IQ samples before sending them to the radio (this way you could
smoothly regulate from 0 to 5 watt all the way and with a linear
slider), but this I would not recommend since this will disable PURESIGNAL
(this only jumps in if the TX IQ samples „hit“ unity).

So this is something we have to live with, a simple attenuator between HL2
and PA would cure the problem.

Yours,

Christoph Dl1YCF.

Reid Campbell

unread,
Sep 13, 2020, 4:25:18 AM9/13/20
to herme...@googlegroups.com
Hi All,

Just a quick update on the amp. I have been running it quite hard with
FT8 producing about about 40W, and running for over an hour continuous.
The temperature as displayed on the LCD interface goes up to 60C at the
end of a transmit cycle and falls back to 55C on receive.

I have let it run at this for a couple of weeks to "bathtub" it as I
know that Sid had a problem with his. I have now put a 80mm fan on the
side, at the back, drawing air out of the unit. The temps are now 40C,
going 45C at the end of a transmit cycle.

I have some bi-metallic temperature switches which I'm going to control
the fan with and that should provide a permit addition to the setup.

Cheers

Reid
Gi8TME/Mi0BOT

Probir

unread,
Sep 14, 2020, 12:27:30 AM9/14/20
to herme...@googlegroups.com, Reid Campbell
Hi Reid,

Bimetallic switch is not accurate and not consistent temp. cutoff can take
place - So just keep this thing in mind.

Coming after a long time here - Busy in building LDMOS power amplifier with
MRF300 around 500 to 600 watt.



73
Probir VU2BQF
.........................................
--
You received this message because you are subscribed to the Google Groups
"Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/hermes-lite/0060fabc-1f7a-a85f-e84d-c2353df3f9e1%40gmail.com.

VU3NXI Siddhalingesh Basawanal

unread,
Sep 14, 2020, 12:37:45 PM9/14/20
to Hermes-Lite
Hi All,

I have just added an 80mm fan on the side, yet to make the connections. Intend to drive it continuously (direct connection to the PS). It is drawing the air in and on top of the PCB.

What should be the best orientation, drawing the air in / out? I do not have an exhaust fan yet at the back panel.

Best Regards,

Sid
VU3NXI
20200914_210109.jpg

Reid Campbell

unread,
Sep 14, 2020, 2:59:40 PM9/14/20
to Probir, herme...@googlegroups.com
Hi Probir,

On 14/09/2020 05:27, Probir wrote:
> Hi Reid,
>
> Bimetallic switch is not accurate and not consistent temp. cutoff can
> take place - So just keep this thing in mind.

I'm not to worried accuracy but placing the switch to pick up the heat,
without opening the box may be a problem.

Cheers

Reid
Gi8TME/Mi0BOT

Evan Hand

unread,
Sep 14, 2020, 4:12:13 PM9/14/20
to Hermes-Lite
Sid,
In designing cabinets for hot industrial locations we always forced outside air into the enclosure.  This was in case the inlet got blocked that we would not create a low-pressure area in the box that would reduce the heat transfer.  The rest of the design is to then put the exhaust close to the hottest components so that the heat from them is not circulated in the enclosure.

With that, I have seen cabinets plumed both ways.  As long as the flow is not restricted, there is not a lot of difference.  It is only when there are restrictions on the makeup air that you can run into problems.

73
Evan
AC9TU

Probir

unread,
Sep 15, 2020, 12:24:50 AM9/15/20
to VU3NXI Siddhalingesh Basawanal, Hermes-Lite
Hi SID,
 
Drawing the Air out will be a better choice.  Had this been more than 500 watts it should have been forced cooling.
 
Also drawing the air out wont attract environmental dust inside the enclosure over a period of time
 
 
 
73
Probir, VU2BQF
 
Sent: Monday, September 14, 2020 10:07 PM
Subject: Re: Xiegu XPA125B 100W amplifier support
 
--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

Reid Campbell

unread,
Sep 15, 2020, 2:28:08 AM9/15/20
to herme...@googlegroups.com
I have the fan orientated to draw air from the enclose at the back of the amp. I would image, with time, that dust will be draw in through the front of the amp, so I may have to look at some filter arrangement.

Cheers

Reid
Gi8TME/Mi0BOT

VU3NXI Siddhalingesh Basawanal

unread,
Sep 17, 2020, 2:22:27 PM9/17/20
to Reid Campbell, herme...@googlegroups.com
Thank you Chris, Evan, Probir, Reid !! That makes sense to get the air out. I oriented the fan to draw the Air out and completed the connections. As of now, 2 mins key down is taking it to 34 C and release drops back to room temperature (28 C) within a few seconds.

73,
Sid VU3NXI

You received this message because you are subscribed to a topic in the Google Groups "Hermes-Lite" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/hermes-lite/Yzv8hG0Jy_8/unsubscribe.
To unsubscribe from this group and all its topics, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/68503f99-811b-c574-0869-cff84d8d4501%40gmail.com.

Reid Campbell

unread,
Oct 10, 2020, 3:35:01 AM10/10/20
to herme...@googlegroups.com
Hi Group,

Just a quick update on where I have gotten to with the Xiegu XPA125B amp. I was able to modify the Gateway code to repurpose the Fan PWM output as a band select voltage. I have a simple RC filter (10K/1uF) on the output which is stable enough for the amp to accept to switch bands.

The band voltage input resistance of the amp is about 90K, so that is limiting the max voltage, so it's a bit lower than the XPA125B is expecting for the higher bands but it not causing any problems so far. I may tidy it up and use the fan PWM switching arrangement with 3.3V to beef up the voltage but will need to order the components.

I got the bimetallic thermal switch (40C switch) working working by placing it under the amp and holding it against the case with a bit of cardboard. I had to be carefully and put the switch in the ground line for the fan as the metallic case of the switch was connected to one of the leads. The amp will get to about 50C and then the fan kicks in and takes it back to about 45C. This is with about 50W output running FT8.

I plan to make a L shaped bracket to support the fan with the switch on the long section on a bit of foam to insulate it both electrical and thermally from the bracket and hold it tight against the case bottom.

Cheers

Reid
Gi8TME/Mi0BOT

William Pickles

unread,
Feb 19, 2021, 9:33:45 AM2/19/21
to Hermes-Lite
Reid,

I have an HL2 and was also thinking about the Xiegu XPA125B to give it a bit more presence. I am NOT an electronics person and was confused, and concerned, at the statement about the HL2 EXTTR being " This is an "open collector" style output to turn on an external power amplifier for transmit. The centre pin is pulled low by the HL2 during transmit. The ring is ground. An external PA should pullup the centre pin with a resistor. The pin may be pulled up to 28V. " . This statement together with the note in the Xiegu manual in section 3.1.3 saying “Note: If a voltage of 3.2V is applied to the PTT port of the XPA125B, the internal processor will be permanently damaged!”.

 This made me stop and investigate further. I have also read concerns with this amp from IC705 users, or should I say rumours.

I am very interested that you have got the amp to work OK  with the HL2. Do I need any extra circuitry or is it just a case of connecting the EXTTR phono to pins 2 and 6 on the XPA125B?

Very interested too with your efforts to get the HL2 to switch bands automatically.

Many thanks for any reply,

Bill,   M0XJL

Jim Ancona

unread,
Feb 19, 2021, 9:45:31 AM2/19/21
to William Pickles, Hermes-Lite
I'm not Reid, but "open collector" means that you must supply the voltage that the HL2 will switch. The warning means that whatever voltage you supply must be less than 28V. If the Xiegu doesn't want to see more than 3.2V you should use a power supply that is a lower voltage than that. I don't have a Xiegu amplifier, but presumably it has an appropriate supply voltage which you could tap into through a resistor. But I'll leave it to Reid or someone who has done it to fill in the details.

Jim N1ADJ 

Reid Campbell

unread,
Feb 19, 2021, 2:12:11 PM2/19/21
to herme...@googlegroups.com
Hi Bill,


On 19/02/2021 14:33, William Pickles wrote:
Reid,

I have an HL2 and was also thinking about the Xiegu XPA125B to give it a bit more presence. I am NOT an electronics person and was confused, and concerned, at the statement about the HL2 EXTTR being " This is an "open collector" style output to turn on an external power amplifier for transmit. The centre pin is pulled low by the HL2 during transmit. The ring is ground. An external PA should pullup the centre pin with a resistor. The pin may be pulled up to 28V. " . This statement together with the note in the Xiegu manual in section 3.1.3 saying “Note: If a voltage of 3.2V is applied to the PTT port of the XPA125B, the internal processor will be permanently damaged!”.

I seen that as well and thought, OK, the pull up voltage within the XPA125B is to the 3v3 line. To confirm this, I measured the voltage on the PTT pin (pin 2) and it was about 9 volts! I don't know where that fits with the warning in the manual. So I just pull the PTT line to ground with the EXTTR line.


 This made me stop and investigate further. I have also read concerns with this amp from IC705 users, or should I say rumours.


One thing you have to be careful of with the XPA125B amp, is the drive level. I run the power slider at '1' and that gives enough input to drive the amp into the range 60 - 110 watts. 12M seems too produce the most gain, with well over the 100W mark. I used to run at drive level '0' but noticed that sometime that produced full output.

I think some of the bad press for the amp comes from overdrive. Sid might comment on his experience with the amp as his went faulty. I also noticed that MFJ had an offer on them of $100 off but I wasn't sure if they were off loading their remaining stock. 

I am very interested that you have got the amp to work OK  with the HL2. Do I need any extra circuitry or is it just a case of connecting the EXTTR phono to pins 2 and 6 on the XPA125B?


That is all the connection you need for basic operation where you select the band manually and remember to reduce the power to level '1'


Very interested too with your efforts to get the HL2 to switch bands automatically.


I have updated the HL2 Wiki with my modifications to the end plate to create band select voltage and it is working well for me. The amp is connected to a 40M HWEF and I have gone through all the bands, 80 - 10M, manually trigging the ATU. The amp stored the solutions for each of the bands, so QSY'ing is all automatic. 80m is the only problem child as the tune range is limited and the amp will not self tune unless the SWR goes over 3:1. I just move to either end of the band, tune, then to the required frequency and retune again.

When the amp goes out of guarantee, I will look at using one of the two unused pin on the ACC connector to initiate the ATU tune cycle. It would have been nice if it had been included in the first place.

Cheers

Reid
Gi8TME/Mi0BOT

 

William Pickles

unread,
Feb 22, 2021, 11:00:29 AM2/22/21
to Hermes-Lite
Reid,

Many thanks indeed for your quick response what puts my mind at ease regards the PTT on the XPA125B.

You do raise a very important further concern with regards to overloading the front end of the amplifier. My initial thoughts were to perhaps introduce an attenuator between the HL2 and the XPA125B - but of course I don't want it in circuit during receive.

Perhaps the better solution would be the MX-P50M together with an external antenna tuner. Cheaper but perhaps not as elegant.

Thanks again,

Bill - M0XJL
Reply all
Reply to author
Forward
0 new messages