It is currently Wed Oct 28, 2020 11:13 pm

Click the link below to visit the site sponsor

The Ham Radio Shop




Post new topic Reply to topic  [ 22 posts ]  Go to page 1, 2  Next
 Homebrew High Power Amplifier Project 
Moderator
User avatar

Joined: Wed Jul 16, 2008 7:01 pm
Posts: 2520
Location: Kidsgrove, Stoke-on-Trent IO83VC
Feedback: 9 (100%)
Reply with quote
Post Homebrew High Power Amplifier Project
With the onset of Winter and the countless hours of watching TV looming I decided to set myself a couple of challenges to try and bait off the inevitable for as long as possible.

For some time now I have been considering changing my 2m Triode valve amplifier,
I built the GS31b back in 2003, it’s never given a moments trouble and begs the question why replace it?
Well the Triode amp needed a high level of input power to drive it (typically 30w for legal limit out at the antennas driven element taking into consideration 1.5db of feeder loss).

During this summer I have just replaced all of the 2m system and at the heart of it now lies a DB6NT transverter, a remarkable piece of engineering on many levels including its excellent intermodulation performance. It seemed a shame to have to drive it flat out into my valve linear to achieve 400w out and so the 2m Solid state amplifier idea was born.

There are some advantages to be had over a valve using the new freescale LDMOS power mosfets.

1 . They offer incredible gain, 27db @ 144mhz which means 1watt of drive will produce legal limit in the shack with ease.
2. Instant power, no 3 minute warm up time for the heaters, a major advantage during a sporadic e opening!
3. Clean linear output with good 3rd order IMD performance due largely to 50v operation.
4. Simple power supply requirements 50v @ 30 amps.
5. 1.25kw peak output offers a nice spectrally clean signal when driven to UK power limits (400w) with enough headroom
to ensure high duty cycle modes such as FSK441a and JT65B don’t present ....a problem with regards to thermal stability.
6. No more 3000v @1amp D.C. in the shack.
7. Efficiency in excess of 70% which is remarkable and much less heat to dissipate when compared
with a 4” anode cooler on the valve amp (the GS31 is about 53% efficient for those who want to know)

So my mind was made up, all that was left to decide was do I buy or build!
A German company called Beko make the HL1000-V which sells at £2200,00 and it uses a Freescale device.
W6PQL makes a similar turnkey amplifier for $2600 USD- plus VAT on top when it enters blightly which roughly translates to £1950 plus transport cost and you still need a 50v PSU.

Those of you who know me probably know I like a bit of homebrew and the allure of saving some brass whilst at the same time learning how to do it was just too tempting.
So the winter project was born.

An article had been written for Dubus magazine by Lionel F1JRD about the implementation of the new Freescale MRFE6VP61K25H LDMOS device. The boards and components are available from a number of sources both in Europe and the USA doing some further research revealed quite a few people had successfully built amplifiers using them.
I decided to go with the W6PQL option.

Jim (W6PQL) has over 30yrs experience in the R.F. design field and his control boards have been around for some years now I used one of the earlier versions in a 6m/4m dual band amplifier back in 2008. After a few evenings studying his website I came up with a list of parts required, Jim sells them in kit form and most of the components are SMD.

The items needed from Jim were as follows.


R.F. Pallet with the transition boards, input and output matching and bias set control.
50v Solid state FET switch (to turn on the 50v at high current when sequenced).
Its nice not to have to leave VDD connected to the LDMOS device in case of PSU failure or bias leakage.
Low pass filter with direction coupler to remove 2nd and subsequent harmonics and drive the VSWR trip circuit and forward/ reflected power measurement.
Amplifier control board, this is the heart of the project and controls PTT sequencing for coax relays, fan control, temperature monitoring and VSWR trip.

I also needed 28v too so I ordered a very smart 12-28v booster that drives the high power coaxial relay at the business end.

A 10db Input attenuator, 10w in from my DB6NT needs to be attenuated to 1w of drive for the amplifier and this fits in the tx chain after the input relay, it’s only in line during the transmit cycle, this also serves as a nice 50ohm stable load offered to the transverter.
And the all important 10 segment bargraph displays that read forward and reflected power simultaneously -you gota have some lights on it ;)

Anyway that little lot in kit form set me back just short of £700,00

Next on the list, cooling.
Although the LDMOS device is highly efficient it is also small in physical size.
The maximum junction temperature is around 225 degrees Celsius and if the cooling proves to be insufficient its life will be very short lived, at £180 that’s not an option!

The device is first soldered onto a purpose machined copper heatspreader using a hotplate (takes about 20mins on the oven to complete- note for next time XYL is best sent out shopping for the day!)

Copper has a higher thermal conductivity rate than that of aluminium and so serves better at initially transferring the heat away from the device, the copper heat spreader is about 15mm thick and weighs plenty!. The bottom of the copper heatspreader has to be milled perfectly flat and a thin layer of thermal compound applied prior to mounting onto the aluminium heatsink.
R.F. ham in France sell a rather nice ducted high performance aluminium heatsink that comes with the top and bottom milled completely flat, it’s designed for use with high power solid state amplifiers and so I ordered same at just under £100,00 delivered.

Attachment:
RF Ham heatsink with 2 x 80mm fans fitted HRD2.jpg
RF Ham heatsink with 2 x 80mm fans fitted HRD2.jpg [ 171.37 KiB | Viewed 2444 times ]

Attachment:
RF Ham heatsink with 2 x 80mm fans fitted rear HRD.jpg
RF Ham heatsink with 2 x 80mm fans fitted rear HRD.jpg [ 140.22 KiB | Viewed 2444 times ]


I chose a pair of 80mm 12v square fans to mount on one end of the ducted heatsink on rubber mounts, they operate when PTT is activated and also stay on once the heatsink temperature reaches 110 degrees Fahrenheit , if the temperature escalates to 135degrees Fahrenheit the amplifier thermal monitoring circuit locks out the PTT and will not operate until it cools to an acceptable level.


The input coax relay was sourced from Henry radio in the states and I happened to have a high power Amphenol one in my junk box to suit the output.

So about 6 weeks ago I set about building the boards up, SMD work is tedious, Jim offers a building service at very reasonable costs but this is a homebrew project right? So a glutton for punishment I spent an entire weekend building up the boards, with the exception of the low pass filter, for that I allowed myself the luxury of Jim building it as he could also sweep it with his analyser and set it up read to play.

Once the boards were built I tested each one (they run at 12vdc) amazing! all of them worked as they should- that’s a first for me.
To save space I decided to mount each one onto the heatsink- some boards such as the FET switch, 12-28v booster and attenuator need heatsinking anyway and that was the start of what was to become a drilling operation for nearly 80 off 2.5mm blind holes followed by tapping them all M3, and yes I did snap a drill off in the heatsink-but only one!

Attachment:
Wiring 2 HRD.jpg
Wiring 2 HRD.jpg [ 175.26 KiB | Viewed 2444 times ]


I wanted the project to look as professional as possible, I’m no tin basher so looking through the R.S. components catalogue I picked what I thought was a suitable enclosure, this was a mistake. I had measured and it seemed like it would fit but I never thought about the fiddle factor, and it’s turned out to be a pig to install all the goodies inside even using a sub chassis plate.
What other advantage could I have over the Beko except saving money? How about P.A. voltage and current monitoring, Yes that would be very nice and Jim’s own turnkey amplifiers have this feature too. A search on ebay revealed many front mounted cheap moving coil panel meters but almost all of them were front mounted. I prefer window mount and so an enquiry to Anders Electronics in London came up with two suitable 0-1ma meters.

Expensive? You bet, try £115.00, but then why spend good money and have the thing looking naf, so as a second luxury I bit the bullet and ordered them, R.S. components came good with a suitable D.C. shunt to drive the amp meter, and a simple potential divider was built on veroboard to drive the volt meter.

Various brackets and R.F. shields have been made at my works qth out of scrap stainless steel and the time came to cut the front panel- with a jigsaw! At this point I had been having daily PM’s with Lyndon G7IMO who’s encouragement during the project has been brilliant and I was able to draw from Lyndon’s experience a neat safe way to cut the panel without the paint peeling away- thank you Lyndon!

Did I mention the enclosure was a tight fit? Well that’s the understatement of the year.
The big issue was how to form the coax interconnects.
Normal high power coax such as LMR400 does not lend itself to staying bent at all sorts of angles for long, something semi rigid was needed.
Back to the states for some conformable stuff- expensive of course! A 2m length of the stuff with the correct “N” type connector on cost £40 delivered, but the alternative could have ripped the copper track from the PA board and put strain on the coax relays and the low pass filter as it tried to straighten itself out so this was a no brainer.

Finally the day arrived to test it.
I have only one dummy load, it’s an MFJ oil filled job which is good for 1kw for about 30 secs on H.F., it says it will do up to 400mhz but I doubt it- why do they fit those horrible SO239 sockets on??, anyway it was all I had so I pressed it into service.

Initial checks were made with low current fuses in the 50 and 12v lines, voltages were where they should be and the FET switch had not shorted to the deck!, 50v was reading on the front panel meter so I grounded the ptt line, fans kicked in and bias was set to draw 2amp quiescent current.
For just a couple of hundred miliwatts drive it was doing 100w out, 1 watt gave 500 out, and if you really want to know 2.8w gives 1250w out @ just over 34 amps, that means only around 450w of heat to dissipate and about 72% efficiency.

All in all a very nice amplifier that will, I’m sure be comfortable at UK legal limit plus a bit of headroom for feeder losses, driving it with the DB6NT transverter running at only 10w prior to the attenuator means everything is de-rated by 50% of maximum, the real test will be on digital modes and I plan to give it some stick on meteor scatter soon.

I wrote the article in the hope that someone might find it an interesting read and to prove that people in this technical hobby do indeed still indulge in homebrew projects.
Although this cannot be classed as a “from the ground up” type of project it is neither classed as a simple assembly of a couple of PCB’s.

Some careful thought has been put in deciding where to source the components and how everything should gel together. With the benefit of hindsight I should have opted for a larger case, just a few inches on the depth would have made life a lot easier, it’s well ventilated and I have no issues with the physical size now it’s complete, but when I start on the 70cm one I won’t make that mistake again.

Almost forgot, 50v, yes the power supply, well H.P used a series of switch mode power supplies in their blade servers about 8 years ago, they are just the job for this project and will give 57amps, so again lots of headroom. They are available for £50 from various sources.

It was supposed to be a winter project, it’s 27th of November and it’s complete with the exception of the lettering on the front and rear panel.
I’d better get cracking with the 70cm amp next :)

I’d like to thank both G4CBW for his advice and help in the final testing stages and G7IMO for encouragement and tips on getting the front panel cut neatly.

Anyone want to buy a big 2m valve amplifier?...........................

_________________
Building your own equipment is one of amateur radio's most satisfying elements.


Last edited by G0UWK on Tue Nov 27, 2012 7:54 pm, edited 2 times in total.



Tue Nov 27, 2012 7:48 pm
Profile Send private message

 Homebrew High Power Amplifier Project 
Moderator
User avatar

Joined: Wed Jul 16, 2008 7:01 pm
Posts: 2520
Location: Kidsgrove, Stoke-on-Trent IO83VC
Feedback: 9 (100%)
Reply with quote
Post Re: Homebrew High Power Amplifier Project
More pictures,

Attachment:
Wiring 4 HRD.jpg
Wiring 4 HRD.jpg [ 143.77 KiB | Viewed 2442 times ]

Attachment:
Rear panel of PA HRD.jpg
Rear panel of PA HRD.jpg [ 125.85 KiB | Viewed 2442 times ]

Attachment:
Front Panel 2 HRD.jpg
Front Panel 2 HRD.jpg [ 146.78 KiB | Viewed 2442 times ]

_________________
Building your own equipment is one of amateur radio's most satisfying elements.


Tue Nov 27, 2012 7:50 pm
Profile Send private message

 Homebrew High Power Amplifier Project 
Advanced Member

Joined: Sat May 16, 2009 10:37 am
Posts: 8522
Location: Preston. IO83PQ
Feedback: 295 (100%)
Reply with quote
Post Re: Homebrew High Power Amplifier Project
Thats a superb looking bit of kit Ian. I know how meticulous you are and Ive seen some of the stuff you had assembled when I was down there a few weeks ago. I wish I had the time, skills and patience you had. Well done mate. As for the valve linear? How much is it?

_________________
"Not so fast my friend!" Lee Corso


Tue Nov 27, 2012 7:56 pm
Profile Send private message

 Homebrew High Power Amplifier Project 
Advanced Member

Joined: Fri Mar 27, 2009 3:11 pm
Posts: 3510
Location: IO80BS Spreyton Devon.
Feedback: 3 (100%)
Reply with quote
Post Re: Homebrew High Power Amplifier Project
Thats a fantastic post Ian,and as usual a really well made project.


Seriously...well done,its nice to see some really nice constructional work leading to such an impressive outcome.

Are you sure that it actually worked first time.....really :shock:


Last edited by g4rra on Tue Nov 27, 2012 8:08 pm, edited 1 time in total.



Tue Nov 27, 2012 8:05 pm
Profile Send private message

 Homebrew High Power Amplifier Project 
Advanced Member

Joined: Mon Nov 02, 2009 5:30 pm
Posts: 1287
Location: IO 83 NU Elswick
Feedback: 2 (100%)
Reply with quote
Post Re: Homebrew High Power Amplifier Project
Fantastic job, Ian. ..and I greatly appreciate the time you took to publish the product on this site.

You sounded as good as ever the other night on 2m, so you have a sweet output.

...and while you were building the amp, I put 2 UHF connectors and 2 N-types on patch leads, so homebrew is not dead in Elswick!

..and now we await the tropo lift... ..and wait

73

dave


Tue Nov 27, 2012 8:06 pm
Profile Send private message

 Homebrew High Power Amplifier Project 
Contributor 2020

Joined: Mon Oct 17, 2011 2:50 pm
Posts: 1303
Location: IO81KQ GWENT
Feedback: 106 (100%)
Reply with quote
Post Re: Homebrew High Power Amplifier Project
that is a cracking job well done buddy it looks amazing fair play

_________________
ELITE GIFTS FOR ALL YOUR SUBLIMATION GIFTS PRINTED MUGS CUSHION COVERS T SHIRTS MOUSE MATS WE PRINT YOUR DESIGNS OR PICTURES ONTO THE ITEMS


Tue Nov 27, 2012 8:15 pm
Profile Send private message

 Homebrew High Power Amplifier Project 
Contributor 2019
User avatar

Joined: Tue Apr 03, 2012 11:58 am
Posts: 843
Location: Stockport, Cheshire.
Feedback: 23 (100%)
Reply with quote
Post Re: Homebrew High Power Amplifier Project
Quality!

_________________
Build a man a fire, he'll be warm for a night. Set a man on fire and he'll be warm for the rest of his life.


!!!!!!!!!!!!!!!!!!!!!!FAFFER!!!!!!!!!!!!!!!!!!!


Tue Nov 27, 2012 11:19 pm
Profile Send private message

 Homebrew High Power Amplifier Project 
Moderator

Joined: Fri Jul 20, 2007 1:02 pm
Posts: 15135
Location: IO83lb, Wrexham
Feedback: 16 (100%)
Reply with quote
Post Re: Homebrew High Power Amplifier Project
Your usual high quality Ian.
A pleasure to see.
I look forward to a melted front end some time soon. :D
Perhaps it's just as well I've left out the masthead pre-amp, since going over to the transverter. :lol:

Cheers

Tony

_________________
50MHz and above from IO83lb


Tue Nov 27, 2012 11:51 pm
Profile Send private message

 Homebrew High Power Amplifier Project 
Contributor 2019
User avatar

Joined: Sat Jun 18, 2011 4:15 pm
Posts: 14711
Location: io65ga, Donegal
Feedback: 8 (100%)
Reply with quote
Post Re: Homebrew High Power Amplifier Project
I'd love top have the ability to turn out something like that....but I don't.
Very professional looking Ian, well done.

_________________
Even duct tape can't fix stupid, but it can muffle the sound.


Wed Nov 28, 2012 12:30 am
Profile Send private message

 Homebrew High Power Amplifier Project 
Contributor 2019
User avatar

Joined: Sun Sep 21, 2008 4:58 pm
Posts: 353
Location: Stoke on Trent IO83va
Feedback: 1 (100%)
Reply with quote
Post Re: Homebrew High Power Amplifier Project
Wow, nice neat job Ian! :thumbsup:
Looks like my front end is going to get some stick hee hee :D
I'm not worried buddy as I know you and Tony will have checked it out and made sure it has a spotless and pure output.
I will look forward to the youtube videos of all the dx worked.

How are you finding 70cms? I actually heard the Swiss beacon on there last week, have you been working a pile up?

Great job as usual.
Thanks for sharing.
:thumbsup:

_________________
http://www.clublog.org/league.php?club=69
http://www.clublog.org/league.php?club=115
https://secure.clublog.org/league.php?ukcw=1


Wed Nov 28, 2012 3:35 am
Profile Send private message WWW

 Homebrew High Power Amplifier Project 
Moderator
User avatar

Joined: Wed Jul 16, 2008 7:01 pm
Posts: 2520
Location: Kidsgrove, Stoke-on-Trent IO83VC
Feedback: 9 (100%)
Reply with quote
Post Re: Homebrew High Power Amplifier Project
Hello Lads,
Thank you all for your kind words,
Paul, yep it worked first time- honestly, I thought the bargraphs were not functioning but a quick tweek of the input sensitivity potentiometers bought them both to life.
It’s great to have both forward and reflected indicators visible without the need for a 43 inline.

Terry thank you and re Valve amp, it was said tongue in cheek and I have had two enquiries a month back for it but as ever I need to be 100% certain the new PA carries the same reliability before I decide to sell.

Mark, yep it’s clean, the reported limit with the usual two tone test before it hits 1db compression is around the 1000w level (realistically it’s about 850w) and it’s well within spec so no worries at our power limits. Dynamic range is very good, I used the NXR variant of the Freescale device (BLF578XR) it is documented to be able to withstand 65:1 VSWR abuse at all phase angles! I won’t be testing out that claim hopefully.
70cm was really good a couple of weeks back, OE and HB9 amongst others, I nearly missed it due to the 6 solid weeks spent working on this amp- thanks for the text alert by the way Tiny if you are reading this.

Dave, thanks as ever for the tests! I see you are progressing in technical prowess :good:

Gerry, I bet you could do it, all the drilling was done on a workmate with a hand drill and I used a small adjustable spanner to drive the 2.5mm tap into the holes with some grease as lubrication.

All the front panel square holes were cut to within 1mm of the required sizes and then needle filed to the final dimensions, painful and time consuming but satisfying when complete, mounting bezels hide any imperfections ;)

Tony, it should be as clean as the GS31 so hopefully you won’t notice any difference, in fact the DB6NT should be a bit cleaner that my original transverter.
Coming from a non technical background (RF wise) as ever these sort of projects are always a bit daunting but if I can do it then it’s possible most others can, with an interest in learning and of course some spare time. It comes back to having someone there in case of difficulties and in my case having Mr CBW at the end of the phone makes projects like this one less risky of failure.

I know I’m very lucky to have that help and eternally grateful for it.
It works both ways of course, I spent quite some time cutting all the elements to his new pair of 70cm LFA’s to within 250micron tolerance as payment for last Sunday’s test sessions, and he bought the curry and beers!

_________________
Building your own equipment is one of amateur radio's most satisfying elements.


Wed Nov 28, 2012 11:39 am
Profile Send private message

 Homebrew High Power Amplifier Project 
Moderator
User avatar

Joined: Wed Jul 16, 2008 7:01 pm
Posts: 2520
Location: Kidsgrove, Stoke-on-Trent IO83VC
Feedback: 9 (100%)
Reply with quote
Post Re: Homebrew High Power Amplifier Project
There was one photo missing, the amp actually producing the R.F.
This was taken during the final test stages into the dummy load with 2.8w of drive, the oil was bubbling out of the top within about 15 seconds :D

Link here http://www.flickr.com/photos/64948567@N ... otostream/

_________________
Building your own equipment is one of amateur radio's most satisfying elements.


Wed Nov 28, 2012 1:44 pm
Profile Send private message

 Homebrew High Power Amplifier Project 
Advanced Member

Joined: Sat Jan 22, 2011 6:16 pm
Posts: 818
Location: stoke on trent
Feedback: 0 (0%)
Reply with quote
Post Re: Homebrew High Power Amplifier Project
Well chaps... isnt it nice to see a "amateur" doing such a professional job...
I met Ian through this forum after posting some comments on another topic and funnily enough I think its true to say that we hit on a friendship that will last, it was plain to see that we immediatey had similiar interests and ideas and Im proud to call Ian a friend.
Ian had mentioned that he had started this project and was a little sceptical about mentioning amps to a operartor who enjoys qrp, but as I said at the time, its good to see what others are upto, it was so nice to see him advancing through the project and Im more than pleased that a few words of encouragement and the odd tip here and there helped him to achieve his goal...

Ive said to Ian before that the amp is something that he can be really proud of and Im sure he will have many years of joy using it..

Well done again Ian.

Rgds

Lyndon

_________________
ELECRAFT K2 ICOM706MK2G X1M
Antennas now that's a different question all together...
Think string that is wet....


Wed Nov 28, 2012 2:27 pm
Profile Send private message YIM

 Homebrew High Power Amplifier Project 
Top Poster
User avatar

Joined: Sat May 23, 2009 1:46 pm
Posts: 21139
Location: Shropshire UK IO82TN
Feedback: 5 (100%)
Reply with quote
Post Re: Homebrew High Power Amplifier Project
Ian, I cant claim to build amplifiers of this ones power capabilities, the biggest to date being 100 watts solid state for 2 meters!.........pales into insignificance really! :D
But being someone who also likes the "finished" job to be something remotely resembling proffesional, I can say to you, that this is an EXCELLENT project mate, I applaud your attention to asthetic, as well as electrical soundness! :good:
With a bit of "silk screen printing" on the front and back panels, that would hold its own in the commercial realms of amplifier production!
A VERY good result, and one I am sure will reap dividends for all the "time and effort" you have obviously put into it! :good:
Very well done Ian, a credit to the hobby!
All the best...........

_________________
Dave G0BHD.
Been there and got the T shirt, sadly like the T shirt the memory of it has long since faded, gone rather tatty, and finally been consigned to the bin!


Wed Nov 28, 2012 3:11 pm
Profile Send private message

 Homebrew High Power Amplifier Project 
Contributor 2019
User avatar

Joined: Mon Oct 19, 2009 8:22 pm
Posts: 1270
Location: Monteneuf - Brittany IN87VV
Feedback: 108 (100%)
Reply with quote
Post Re: Homebrew High Power Amplifier Project
Now't much I can add Ian but Congratulations on the finished result. Looks excellent and by your test results meets what you set out to achieve.

Here's looking forward to working you again in the future :good:

73's
Kelvin (in wet Wales !!!)

_________________
Kelvin F4VPC / GW4TTU, feeling a bit like Bonnie Tyler - Lost in France !


Wed Nov 28, 2012 4:16 pm
Profile Send private message WWW
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 22 posts ]  Go to page 1, 2  Next


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
Powered by phpBB © phpBB Group.
Designed by Vjacheslav Trushkin for Free Forums/DivisionCore.