Solid state linear for AM service

[ka1tdq]

To build a solid state linear amplifier capable of 1.5KW+, you could buy 4 of these amplifier kits:

http://www.communication-concepts.com/eb104-1/

and put them in parallel using this splitter combiner:

http://www.communication-concepts.com/psc-4h5/

Correct?

Jon
KA1TDQ

[K1JJ]

Hola Jon,

I did just that about 10 years ago. Four EB-104s modules combined.  It worked, put out the normal power, but was about the dirtiest amplifier I ever tested. After all that work, I ended up using it as a class C driver for the 4-1000A AM rig. It was about -22 dB 3rd order IMD in linear operation which will get us into trouble on the band.

The MRF-150 FET itself is cleaner, like -30dB or better.  I thought maybe the output transformers and combiner were poorly designed.   Frank / WA1GFZ feels that old Motorola design has problems, probably related to the ferrite transformers design used.  He later consulted with some very smart RF engineers. He  built up a solid state amp that needed lots of transformer changes from the old school techniques to clean it up. He finally succeeded resulting in a clean amplifier and I think he was going to write (or has already done so) an article for QEX mag about his techniques.

If you plan to use a "Pure Signal" or whatever pre-distortion technique, you will be OK. Otherwise look around and demand actual IMD data from the kit manufacturers, or whatever design you use. (IMD data for four units combined using their own transformers)

BTW, I once heard a well known US DXer put eight of those CCI boards (combined) on the 75M band one night. He was at least 8 KHz wide on LSB. Nasty. I never heard him on with it again... :-)  Or maybe he fixed it, dunno.

I see they now sell a different combiner than I bought years ago.  Did that help? Maybe.

T

[ka1tdq]

Oh wow.  I suspect that even if they built a better combiner, the numbers wouldn't be at where they should be in terms of spectral purity.  

Here's my thing.  I might come into possession of a nice 7 foot cabinet with a plexiglass front.  It has plenty of room to build something big.  I don't want to duplicate my dual 3-500 amp, and in fact, I'm trying to get away from the HV stuff.  I now have very small children and I don't want to endanger them in any way.  

That would then suggest a big solid state transmitter, but the complications at 40 meters make things just that much more difficult.  I've built lots of stuff that works, but I really don't think I'm up to the task of anything big class E. I'm not looking for an "ah, come on, you can do it" thing... It really is difficult.  If it were easy, everyone would be doing it.  My single FET rig for 40 meters does a nice job, and I'm just looking to boost power.

Jon

[Opcom]

CA1000HF+6 a 1KW 50V MOSFET pallet is available from Communications Amplifiers commamps.com here in Texas. It was about $750 last time iIchecked. I don't know how really clean it is, the specs are not so impressive, but they will also make combiners for any of their products so that it can run at partial power where it is likely cleaner.

There is also the idea with any of the solid state products that feedback and so-called 'pre-distortion' can be used to make a cleaner signal.

[steve_qix]

If you can live with 40 meters and lower in frequency, constructing a solid state linear is neither expensive nor difficult.

I built and ran a fairly hefty linear on 75 meters made with 6 FQA11N90s in a single module.  Didn't experiment with two modules, but I suspect things would have worked just fine with that, too !

You need small value resistors in series with each source lead, to ground.  This both stabilizes the amplifier, and balances the current between the MOSFETs.  As I recall, the amp only required a few watts to drive it.  I was able to get around 150 watts carrier out.  With two modules, you should be able to get around 300W carrier.

Standard class "b" design - with a tuned circuit in the output, etc.  Just like a tube linear, except solid state.

[VE3ELQ]

There are some amazing new LD MOSFET devices available now for use in HF linear amps. The BLF188XR dual FET by NXP will easily provide 1KW OP and is extremely rugged, thus the XR in the part number. There are application notes and plenty of amp designs using this FET or the Freescale equivalent on the web and Epay has a number of RF boards from many designers.

The Amp board is really the easiest part to build.  You will also need a 50V well regulated PS, a bunch of protection circuits, and a low pass filter for each band. Then there is the input attenuator, metering and Power/SWR bridges etc to do also so it becomes a complex project. If you are not into designing or building PCBs then much of this is available in board level kit form. A good place to start is with W6PQL here:
http://www.w6pql.com/1_kw_sspa_for_1_8-54_mhz.htm
Jim has most of what you will need from parts right up to finished boards.
For a power supply there are some excellent switchers out there to do the job and yes the good ones are RF quiet.
Mean Well makes great switching supplies that will do the job, look for the RSP-1500-48 series. They do this exclusively for a living so you wont build one better for the money.

I have recently completed designing and building a 1KW 6 band HF Linear Amp for my ANAN-10E using the BLF188XR. I dont want to contaminate your thread with my stuff so if anyone is interested I can start a new thread and post some pictures and details. Im on 1975 Khz LSB most evenings if any wants to hear it or chat.

Good luck with you project,  73s  Nigel

[ka1tdq]

There are commercial solid state amps out there that go to 1.2KW, but I wanted to be able to do around 300 watts carrier (like I can do with my current linear with a pair of 3-500's).  I wouldn't dare run more than 100 watts with those amps. 

I only operate on 40 meters, so Steve's option sounds pretty good.  You wouldn't happen to have a schematic of that module would you?

As for a regulated power supply, I was thinking about using 4 car batteries in series.  That way you can just charge the whole thing from a 120 volt plug and not need 240 volts in the shack.

Jon

[WD5JKO]

    Jon,

   A high voltage exposure as you mentioned can be very dangerous. Also low voltage at high current can be dangerous as well. For example, tossing a Tonka toy across a bank of capacitors could cause flash burns, heating, and fire...even if only using 12V.

Also 1500W at 50 ohms is 273 volts RMS. If you are feeding a random length wire, it could be much more than that, i.e. doesn't matter whether amplifier runs off 12V or 3KV.

   My point is whether you have a 3-500Z amplifier, OR a bank of transistors running off low voltage, hazards exist. Since you are getting a tall rack, perhaps you can repackage the 3-500z amplifier and power supply such that nothing dangerous is exposed? Or think of the many commercial tube amplifiers where nothing dangerous is exposed. Sure if you want to build a big SS amplifier, hazards exist there too unless it is well enclosed from curious fingers.

   I have raised three kids with high voltage stuff in the hamshack. My first line of defense was to limit what was exposed, and more important to make sure the ham shack WAS NOT a kids play area. A locked door is an option while you are away. Put the key on a nail above the door, 6' off the ground.

   Some parents tend to let the kids go/do anything, even if that means climbing over the ham radio equipment. I was not one of those parents. My kids had lots of freedom, but never in the hamshack. In the hamshack it was sitting on daddy's lap making the scope twitch to the kids voice, i.e. always supervised. A living room ham shack with small kids is not going to work unless everything is in a locked cabinet.

   Kids are more likely to injure themselves falling down stairs, and with dressers that are not bolted to the wall. My daughter used dresser drawers like steps on a ladder to get on top of the dresser, and she was only 2! She also went down the stairs twice like a bowling ball, even with a 4' tall gate at the top...

Jim
Wd5JKO

[WBear2GCR]

Nigel, VE3ELQ, I am very interested in what you have on solid state linear amplifiers.

Also, if anyone has a copy of the GFZ article from QEX, I'd like to read that... (PM me?)

Do any solid state amplifiers use RF feedback for linearization?? Examples?

                            _-_-

[ka1tdq]

You make a really good point Jim.  People raise kids with all sorts of hazards around like guns, power tools, kitchen utensils, etc.  I do take precautions on my linear by having a lock on the 240 volt box supplying the amp, and another set of locks on the switches to the filaments and HV.  In theory, it should only be operated when my butt is in the chair. 

I guess the only true safe method of operation would be QRPp.  Alright, everybody, listen up... Turn your carriers down to 1 watt in the name of safety.  (I guess I sound like a Washington politician).

Jon

[KD6VXI]

John,

Single father here.   Had full custody of my kids since 1.5 and 3 yrs old.

They are now 10 and 11.   One has a bday next month.

Since they could speak and understand me (about 4 yrs old),  they knew the dangers of hv.   They also know what to do if they find me in the shop laying on the floor,  etc.   They've been taught not to touch daddy,  etc.

Children are exceptionally smart.   A lot smarter than today's feel good parents give them credit for.   

I agree with Jim.   Incorporate them into your hobby.   Inform the kids of the hazards.   Explain why each piece of DON'T YOU EVER TOUCH THIS is in fact,  don't touch.   If you don't explain,  then they will want to find out why themselves.

To put it another way.....   Your son is running toward the street.   Fixated on his THING rolling towards the street.

Do you want to be the parent that yells his name once,  and he stops,  turns and looks at you.......   Or the parent whose words are more guidelines,  and your son slows,  but keeps after said whatever.  Your shack is the same thing.

 Think of safety in your shack the same way.   Key switch on the AC interlock (running a contactor?)  preventing hv from ever being turned on without said key?   And,  all your hv behind new plexiglass rack.

--Shane
KD6VXI

[K1JJ]

Also, if anyone has a copy of the GFZ article from QEX, I'd like to read that...
                            _-_-

Frank just sent me a copy of his article from Sept/Oct 2015 QEX. It is quite informative and really shows the past problems and assumptions of the old schoolers ideas about ferrite transformer techniques. No wonder the amp was not as clean as it should be. In fact one of the original proponents of that Motorola design sent him an email saying he wished he could go back and redesign that old EB-104 circuit.

Problem is, I don't know if I can upload the QEX article here legally... anyone know?


It is not easy to make a multi-module SS linear amplifier to test as clean as a tube that was designed for linear service, like an 8877, etc.   The military uses RF negative feedback and Frank is suggesting using balanced transformers rather than shunt fed stuff. It's all in his article.

As for using 11N90 MOSFETS in linear service, I know of several guys who have built mobile linear amps using them and say they work well. But, I have never seen actual IMD result tests and have my doubts if one would be able to run a big 1500 w 11N90 linear amplifier into a full size dipole without getting complaints. I'd need to see the IMD test data.  IE, it's tough to do with expensive linear designed MRF-150s, so 11N90s would be a bigger challenge.

Nigel, your suggestion of using the more expensive LD devices is probably the best approach for a minimum of modules, less transformers, less clutter, latest technology, best IMD etc..

And, yes, pre-distortion, Pure Signal, etc., solves all linear amplifier problems. We could use the dirtiest abomination of an amplifier chain and end up with pristine IMD approaching -50dB +.  Like spark gap getting booted out by CW, there will come a day on the bands when even the now popular open-loop 3-500Z amplifiers are looked down upon as trash generators - when pre-distortion techniques finally become mainstream and rule.  

T