Topics

External Amp

Howard Fidel
 

I have been working on a number of upgrades for the uBitx. One is an external amp. A number of people, including myself ordered an amp like this one: https://www.ebay.com/itm/70W-SSB-linear-HF-Power-Amplifier-For-YAESU-FT-817-KX3-Ham-Radio-DIY-Kits/401280684714?epid=579537921&hash=item5d6e3152aa:m:mTAXYtpMu9k-yfYI6Z1m1hQ

For the price it is a good deal, but it doesn't include the heat sinks or the needed low pass filters for each band. I currently have the unit running with my uBitx only on 20 meters. I temporarily but an appropriate filter on the board while I work on my uBitx daughter board that will be a collection of several improvements. I have attached my  version of the original schematic and also my revised schematic. I made one major change and a few minor ones. The major change is that I use two IRF520s in parallel instead of the original IRF530s. This came about accidentally, but it is a could change. When I put the amp together and started testing it, the FETs quickly self destructed, before I could shut it off. I looked through my parts been and found that I have a number of IRF520s but no IRF530. I checked the data sheet, and found that when you put 2 in parallel, the total capacitance increases by about 20%. But the thermal impedance for 2 in parallel is lower for the two, and the temperature rise in each is lower because the current is 1/2 for each, and the on resistance is not much higher in the IRF520. The bottom line, is you get much better thermal performance and can use a smaller heatsink. To the amp itself, I found that I get better performance by changing the 150 ohm feedback resistors to 235 ohm (to 470 in parallel) and I made the output transformer secondary 4 turns instead of 3. I got slightly better output power, and I believe a better antenna match, since my antenna is closer to 70 ohms then 50. Lastly, I changed R2 to 4.7K from 10 K and put a 6.8 volt zener in series with it so that the PTT can be put in parallel with the uBitx PTT switch. If you want to use the amp for both SSB and CW, then you will need to or the two lines together with a couple of diodes. I'm not running CW now, except for testing, so I didn't bother with it.

When I finish my daughter board design, I will post it. Probably in day or so. The amp is tested, the daughter board will be just the design at this point, with some parts breadboarded. 

Howard

Howard Fidel
 

OOPs, I forgot to attach the schematics.

On 6/9/2018 5:19 PM, Howard Fidel wrote:
I have been working on a number of upgrades for the uBitx. One is an external amp. A number of people, including myself ordered an amp like this one: https://www.ebay.com/itm/70W-SSB-linear-HF-Power-Amplifier-For-YAESU-FT-817-KX3-Ham-Radio-DIY-Kits/401280684714?epid=579537921&hash=item5d6e3152aa:m:mTAXYtpMu9k-yfYI6Z1m1hQ

For the price it is a good deal, but it doesn't include the heat sinks or the needed low pass filters for each band. I currently have the unit running with my uBitx only on 20 meters. I temporarily but an appropriate filter on the board while I work on my uBitx daughter board that will be a collection of several improvements. I have attached my  version of the original schematic and also my revised schematic. I made one major change and a few minor ones. The major change is that I use two IRF520s in parallel instead of the original IRF530s. This came about accidentally, but it is a could change. When I put the amp together and started testing it, the FETs quickly self destructed, before I could shut it off. I looked through my parts been and found that I have a number of IRF520s but no IRF530. I checked the data sheet, and found that when you put 2 in parallel, the total capacitance increases by about 20%. But the thermal impedance for 2 in parallel is lower for the two, and the temperature rise in each is lower because the current is 1/2 for each, and the on resistance is not much higher in the IRF520. The bottom line, is you get much better thermal performance and can use a smaller heatsink. To the amp itself, I found that I get better performance by changing the 150 ohm feedback resistors to 235 ohm (to 470 in parallel) and I made the output transformer secondary 4 turns instead of 3. I got slightly better output power, and I believe a better antenna match, since my antenna is closer to 70 ohms then 50. Lastly, I changed R2 to 4.7K from 10 K and put a 6.8 volt zener in series with it so that the PTT can be put in parallel with the uBitx PTT switch. If you want to use the amp for both SSB and CW, then you will need to or the two lines together with a couple of diodes. I'm not running CW now, except for testing, so I didn't bother with it.

When I finish my daughter board design, I will post it. Probably in day or so. The amp is tested, the daughter board will be just the design at this point, with some parts breadboarded. 

Howard


Nick
 

This is good information. I also picked up the same amplifier and the IRF530s were quick to self distruct.

When testing into a 50ohm dummy load I was initially only getting about 10 watts. The resistors that came with the unit were 100 ohm instead of what I found in the online schematic of 150ohm . 

I made a low pass filter for it but I think the toroids are a bit small as they got really warm. I also made an attenuator pad at the input as I read that the pa does not seem to like more then about 2.5watts in.  The pad was a 220, 22, 220 with about 4db calculated loss. 

I found that with my o'scope the output waveform from the bitx was very distorted into the amp which could explain the low power and heating in my filter. In an attempt to figure out what was going on, I modeled the amplifier in ltspice .It did not work well as per the schematic I found. I had to increase the bias voltage and put a 10 ohm resistor across the two mosfet gates to get a nice input swr and a somewhat nice 70ish watt waveform out of it. This was simulated at 7.2Mhz with the ltspice model of the irf530 and all other components were ideal.

Since it looked good in simulation I attached the resistor and started increasing the bias voltage. I was not using a very small current limited power supply and while doing this there was a small puff of smoke and my supply voltage dropped to nothing. Oops...

I suspect not paying attention to the drain current may have been a problem. I think maybe they turned on and "shorted" at dc. Not sure if I trust the components that came with the kit though as they were in standard bubble wrap with no esd protection for the MOSFETs. They could also be fake MOSFETs.  

I don't currently have replacements available but I might try your suggestion of using the 520s in parallel.

I appreciate your post,
Nick

Howard Fidel
 

Nick:
You reminded me I left out an important piece. I added 10 ohm resistors in series with each of the gates. The circuit was oscillating at 120 MHz, which is probably why the IRF530s blew. I revised schematic is attached. For the 20 meter band, I made 2 air core coils I wound on a 1" D  piece of schedule 40 pipe using 9 turns 14 AWG wire. The pipe is just a winding form, not used in the circuit
That is about .9 nH each. The caps were 240 pF at the input at output and 360 in the middle.
I posted another schematic yesterday with the LPF designs for all the bands.
I adjusted VR1 in the Ubitx to get the correct drive level. Of course the uBitx's output is not uniform across the bands, so you need to set it properly for each band. I had to drive the amp at 28 VRMS to get the full output. I'm not sure what the input impedance is, so I can't measure the drive level in watts. However, at the same setting the uBitx was putting out 24.4 volts RMS into 50 ohms, or 12 watts.
Howard

On 6/10/2018 10:08 AM, Nick wrote:
This is good information. I also picked up the same amplifier and the IRF530s were quick to self distruct.

When testing into a 50ohm dummy load I was initially only getting about 10 watts. The resistors that came with the unit were 100 ohm instead of what I found in the online schematic of 150ohm . 

I made a low pass filter for it but I think the toroids are a bit small as they got really warm. I also made an attenuator pad at the input as I read that the pa does not seem to like more then about 2.5watts in.  The pad was a 220, 22, 220 with about 4db calculated loss. 

I found that with my o'scope the output waveform from the bitx was very distorted into the amp which could explain the low power and heating in my filter. In an attempt to figure out what was going on, I modeled the amplifier in ltspice .It did not work well as per the schematic I found. I had to increase the bias voltage and put a 10 ohm resistor across the two mosfet gates to get a nice input swr and a somewhat nice 70ish watt waveform out of it. This was simulated at 7.2Mhz with the ltspice model of the irf530 and all other components were ideal.

Since it looked good in simulation I attached the resistor and started increasing the bias voltage. I was not using a very small current limited power supply and while doing this there was a small puff of smoke and my supply voltage dropped to nothing. Oops...

I suspect not paying attention to the drain current may have been a problem. I think maybe they turned on and "shorted" at dc. Not sure if I trust the components that came with the kit though as they were in standard bubble wrap with no esd protection for the MOSFETs. They could also be fake MOSFETs.  

I don't currently have replacements available but I might try your suggestion of using the 520s in parallel.

I appreciate your post,
Nick


ajparent1/kb1gmx
 

Nick,

That amp with IRF530s was created back in 1999 and known as the "hints and kinks"
amp.  When Wa2EBY tried to copy it he got your result blown transistors.  Its unstable into
reactive loads (mismatches) and the transformers were poor.  It also presents a rather
reactive load to the driver (QRP rig) and is known for making that final oscillate as well.

IF the low pass filter is heating either the design Q is too high or the harmonic content is.
assuming the cores ar T37 size.  IF T50 something is wrong as I've run 60W though those.

FYI I also tried that amp (two IRF530s), it tried to blow!  I used a current regulated power supply
so it could never exceed the blowup current.  The only things that helped was a better output
transformer and a 3DB pad at the input to keep the input tame.  I abandoned it.

The 100/150 ohm resistor are negative feedback in an attempt to stabilize the amp. 

For the 'EBY amp 2W is max power as at that point that amp is cranking serious power.
That and more power will likely fry the gates.

Look at the WA2EBY amp (it is the two files [HF use fet] in files under my call).  With 2W I see
at  80 and 40M about 55W at 28V using 2 IRF510 and about 37W at 10M.  I got a board
for it from FARCircuits.  I use it with a 4db input pad as most of the HB rigs I have do
about 4W.    Do copy the mechanical layout as its critical to cooling and stability.
FYI I bilt mine back around 2006 and its still running despite various "operator errors".

One note.  one must use very good care with parallel MOSFETs regarding layout.
They are known to become unstable and fry.  Adding  a 1-4 ohm SMT (1206) resistor
right at the gate before it goes into the part can calm them.  FYI SHORT leads, the
board and the heatsink should nearly be one unit and the MOSFET leads short as possible.


Allison

ajparent1/kb1gmx
 

A note: the commercial version of the 'EBY amp is the K5OOR HF packer.
the power supply is unique as it runs only on TX and does a 12 to 24V boost
to power the IRF510s to high power.  

Allison

Nick
 

Allison,

Thanks for the reply! I will read the articles regarding WA2EBY.

For the low pass filter I am using T50 the actual Fair-Rite part number is: 5961000301 which is mix 61, lower permeability and should be lower loss at HF. It could be that the whole system like you said is not stable as I have seen the distorted waveform at the input of the amp.

I will likely order some new MOSFETs and give it another go. The gate resistors might be a good addition.

I have attached what I put together in LTspice for anyone that wants to play with it. I did not put in an accurate representation for the transformers but I think the ratio is correct. What I have could be completely wrong.

--
Nick

ajparent1/kb1gmx
 

For low pass filter either T50-2 bekow 10 mhz and T50-6 above as that is the most
commonly used.    The 61 mix has too high mu and is prone to overloading and
heating as anything but transformers at HF.

Wire, I'd use #26 if it fits or #28 as fallback.

Allison

W0PWE
 

I bought one of these out of curiosity more than anything but if it worked I was going to use it as a 40M only amp while operating mobile with my scratch built Bitx40. To my surprise it worked pretty well. Output on 40 was about 50 watts with 1 watt drive and 70 watts with 5 watts drive. I built a 3dB attenuator for the input and a 40M LPF using T50-6 cores. I also built a signal operated PTT driver based on the WA2EBY design. My heat sink was salvaged from the motor control board that died in our Bosch washing machine. It is about 50mm x 100mm and the fins are about 50mm tall. After a long winded transmission the heatsink was just a little warm and the MOSFETs were likewise.

Then one day I was speaking a little too loud into the mic and the MOSFETs shorted. New MOSFETs should be in the mailbox when I get home today so I will try again. I think if I work on keeping the drive level down it will serve it's intended application just fine. I have certainly had more than $18 worth of fun with it so far. 73.
Jerry - W0PWE

allen west
 

I bought and built the 100 watt version but have not tried it out yet.  I also bought a multi band LPF but have not integrated it yet with the amp.  Does anyone have experience with the 100 watt Chinese kit?

Al
KB4RA


On Jun 11, 2018, at 1:43 PM, W0PWE <j.b.hall@...> wrote:

I bought one of these out of curiosity more than anything but if it worked I was going to use it as a 40M only amp while operating mobile with my scratch built Bitx40. To my surprise it worked pretty well. Output on 40 was about 50 watts with 1 watt drive and 70 watts with 5 watts drive. I built a 3dB attenuator for the input and a 40M LPF using T50-6 cores. I also built a signal operated PTT driver based on the WA2EBY design. My heat sink was salvaged from the motor control board that died in our Bosch washing machine. It is about 50mm x 100mm and the fins are about 50mm tall. After a long winded transmission the heatsink was just a little warm and the MOSFETs were likewise.

Then one day I was speaking a little too loud into the mic and the MOSFETs shorted. New MOSFETs should be in the mailbox when I get home today so I will try again. I think if I work on keeping the drive level down it will serve it's intended application just fine. I have certainly had more than $18 worth of fun with it so far. 73.
Jerry - W0PWE

Richie Chambless
 

I have the same amp on my Bitx. I’ve gone through several IRF530s. I did find a suggestion to fix this issue. I found an article that said to put a 22 ohm one-watt resistor across output of the first transformer (gates of the finals). This is said to help with instability. I am also going to install a 50 ohm pad as suggested by Allison. My goal is to get to a stable 35-40 Watts out.

Sajid Rahum
 

Hi Richie

Were you able to get the 22ohm 1 watt resistor added?  can you share the exact place you soldered over from the gate to target solder point.  That is where it is not clear.

Sajid