300W BLF278 PA 48V (PA0V Nanko)

[VK3PF]

Hi all,

I bought my 2 devices from "subrata1979". US$170 for 2 devices plus US$40 for P&P for the pair (he normally charges US$30 for 1 device for P&P), so US$210 total. Took just over a week for the devices to arrive - very well packaged.

Now I just need time to chase all the info and to assemble the rest of the pieces, and start with the soldering iron!

Peter VK3KAI

[VK4DD]

Got an email from David VK5AYD, and am passing on the information:

I will have a couple BLF278 for resale ..if any one wants them...
I will also need some coax.

I am in Coober pedy grid square PG70JX and you want to get into Air craft /Tropo/Es.
I use an FT857 and 11 ELement yagi
David VK5AYD and Nora VK5NYD

David/Nora. Cooper pedy is probably one of the best 2m locations in VK to work all states.
It would also be interesting to know how the path is to VK6.
Let us know when your 2m station ready.

I am very pleased to see you building the amp, and I will sponsor some coax and ATC caps.
It will be great to see you on the air. You are exactly the sort of station people want to see on the air.

73 Ron
VK4DD

[VK3PF]

Hi all,

Is anyone interested in a group buy of the BLF278?

I have received an email from the guy from whom I bought my 2 devices. He has a very interesting offer of 20 devices for USD1000, so only USD50 per FET! That would be about $65 per FET plus P&P within VK.

I can afford to spend that much, but I would need enough people to commit to purchasing the devices - I do not need any FETs, but am happy to facilitate others getting the devices.

Email me direct at vk3kai@wia.org.au

Peter VK3KAI

[VK4DD]

Hi Gents,

I will supply 20 PCB's to a radio club in VK3 for 20 x 300W 2m PA. This club buys Fets and parts and offer a kit at a crazy $/watt ratio...that will never be matched again in VK.
Note this is not for export, but to promote more and better stations on the air in VK.
In order to promote 2m activity VK4DD has sponsored this with low prices and will do the same with other radio clubs who want to do these things in high volume.
Provided that the amps will be sold and used in VK. Well done Peter !

I am happy to do this also with other radio clubs, in order to make home brew more popular.

73 Ron
VK4DD

[VK4DD]

Hi All,

I have completed the schematic from the BLF278 PA.
If there is an update, than I will modify this posting and upload it here.
except the revision date below will be changed.
One location will avoid confusion in this matter.

Please down load it here. (I will update this posting)

SCHEMATIC CORRECTED 29 July 2009 necessary revision: had input coax feeding on wrong point. Was shown correct in pics but not in schematic.
Also updated some resistor values for R2 to reduce stress on R3

SCHEMATIC CORRECTED 24 Sept 2009 necessary revision: had C6 shown as 39pf. This was a copy and paste error in the schematic. Now corrected to show correct value C6 is 120pF.
Thanks to Nico PE0NYJ for picking up this error (important update)

73 Ron
VK4DD

[VK3PF]

Hello all,

As Ron has indicated, EZARC has decided to support a kit project for the BLF278 144 MHz amplifier.

20 FETs have been ordered & paid, also pcb boards at a good price (thanks Ron). As a first step, we need to wait to see what the Tax man wants when the FETs arrive, HI!

I need to work out what else is feasible to include in a short form kit, including suitable cable for the output balun (?25 ohm). BUT work is hectic at present (lots of exam papers to mark!!). One possibility we will examine is to include a decent heatsink, possibly source in VK3. I hope that the supplier might do a deal for a bulk order of 20 or more units - prices look OK at present, based on web published RRP - certainly comparable with an eBay seller once P&P costs are included, even if buying 4 units per deal. So a possible kit may be:
FET, pcb, coax for output balun, SMDs, heatsink, maybe the other on-board bits.

What else would you need (without having checked Ron's earlier posting)?
Copper heat spreader, input & output coax leads, monitoring circuit & bits, change over relays, misc. hardware, etc.

I hope to have details available in the next week or so, so either wait for a further announcement here, or drop me an email with your potential requirements.

There were several members of the club interested (around half the FETs), so they will get first bite at the cherry, as it were..... I am sure that everyone will appreciate that club members get first option, if the club is backing the deal/s. If there are extra units available, they will be announced here.

So.... more news soon.

Regards,

Peter VK3KAI

[VK4DD]

Hi Gents,

Looks like there will be a lot of home brew activity in VK3 and that is excellent news.
Thank you for this great initiative Peter. I look forward hearing more DX from down south.
This will also boost the Es contacts during summer. This is my personal experience as well as the experience from Ken N0GZ who had the opportunity to run this summer with his 2 x BLF248 amp in the USA.
But I suppose that is an other topic, which I don't want to discuss here.

I have done some work to on how to combine these BLF278 (and other ) amps.
On my home page I run also a small forum, just to document lots of info including combiners, where to get coax cheaper etc etc.

I have one combiner in particular in mind for this project and that is this one.
It is a hybrid combiner, partially Wilkinson and partially a quadrature combiner.
Basically it is a Wilkinson combiner with 2 90 degrees delay lines. This has the advantage that less power disappears because the PA impedance's don't interact with each other.

Any way here you can find more about this topic and other combiners.
http://www.vk4dd.com/forum/viewtopic.ph ... 5&start=10

I have also uploaded the schematic on how you do this.

Please note that you can pieces of coax on the input and the output, to go to these
combiners as long as you keep the length the same.

Other than that keep also an eye on the Page of Jim W6PQL. Jim will release a DC solid state switch for 48V.
This will be good if you want to protect the amp of a bad VSWR.

This will work also with his all new amplifier protection PCB.

http://www.w6pql.com/Upgraded_Amplifier_Controller.htm

73 Ron
VK4DD

[VK4DD]

Hi Gents,

I got some PCB's developed which could be useful for this project.
A coupler...
A power supply to drop down 40V to 24V or 12V(...as long as you don't do more than 100W dissipation)
New station monitor PCB's. This is useful because normal 78XX voltage regulator have limited current capacity
or can only dissipate a few watts.

More info here.

http://www.vk4dd.com/forum/viewtopic.ph ... 1&p=84#p84

73 Ron
VK4DD

[VK4DD]

Hi Gents,

Jim W6PQL will have soon 50V Fet switches.
Ask Jim also for his 50 Amp switches they are $1 extra and certainly worth the extra dollar !!

But here is an other option that worked well with the first generation boards from Jim.
Not the second generation because he has got two active low signals.
Kill (when the SWR is high)
and the sequencer. I suppose an extra interface around the solid state could solve this.
But you have to make your own.

Any way here are these so called DC solid state relays.
I found them here also for 100V/40Amp
More than enough for a 2 x BLF278.
http://www.futurlec.com/RelSS.shtml

SSRDC100V40A SPST 0-100Vdc 40A DC Solid State Relay Yes 1 US$19.90
That is more than half the price from our Jaycar stores in VK.

73 Ron
VK4DD

[VK4DD]

Hi All,

As previously announced. Jim W6PQL has now got a FET switch for a 48V amplifier.
Advantages: small foot print, cheaper, no additional interfacing needed with Jims boards.
Disadvantage: the supply voltage is only isolated by a thin isolator sheet under the fet.
Any swarf solder or other metal in your amp can short it to gnd.

High Current DC FET Switch Kit, 48V option, US $15

Can switch 48v (at up to 25 amps)

The schematic for this 48v kit is here.http://www.w6pql.com/images/kits/48vschem.gif

This switch is also compatible with the sequencer and/or the multifunction control board (also offered here).

http://www.w6pql.com/parts_i_can_provide.htm

Ask jim also for the 50 Amp option of this switch. He does not advertise them, but they are $1 extra, so well worth the money.

73 Ron VK4DD

[VK4DD]

Hi Gents,

Here is an update for the schematic.

Metal film resistors only have to be 0.6W instead of 1W.
If you ask for a 1W metal film, you mostly end up with a big 2W metal film.
Both work, but I think most people prefer the smaller 0.6W resistors.

I will update the schematic to reflect this, in the earlier posting which has the schematic attached.

73 Ron
VK4DD

[VK4DD]

Yet... an other update of the schematic... deleted REV2. and uploaded REV 3. Same location.
Follow link for schematic REV3.
viewtopic.php?f=11&t=8497&start=24
No one noticed... but the input coax was not feeding on the junction C1 and C3 and C2 the junction C4.
Any way when you build it and follow the pictures.. you may not have noticed this.
But other wise you would.

I also increased R3 from 470 Ohm to 1000 Ohm 5W.
Or alternative if you feed the bias with 28V than 1200 Ohm 2W.

The 1200 Ohm /2W metal film fits neater on the PCB and I think is a neater option if you use 28V.
By all means this resistor is not that critical as long as you don't exceed the max power of the 500 Ohm trimmer pot or the 1W of the 7.5V zener.

All in all a quite important update

73 Ron
VK4DD

[VK4DD]

Hi All,

Here is an idea how to make a low pass filter using a commercial product from broadcast concepts.
I made a simulation using the broadcast concept PCB and capacitors.
I made the coils a little bit smaller and guess what ... it is perfect to run 600 to 700W on 2m.

Have a look at the attached screen dump.
For the schematic and software I used to calculate this.

Off course you need to tweak the filter with an VSWR meter on low power before you run max power trough this filter.
But the simulation says it is a nice filter. Second Harmonic is down 46 dB. The third is down more than 80dB.

Here is the link to the manufacturer of the unmodified filter....
http://www.broadcastconcepts.com/index. ... ducts_id=4

Cheers
Ron
VK4DD

[VK4DD]

Hi Gents,

Here is just what you for this amp.....
An attenuator for the input.
Make no mistake, if you drive this amp with an IC910 you will kill it as it makes 100W.
Also the ALC of this rig is too slow and has 100W spikes on the output, even when it is on reduced power.
Unfortunately this is not the only rig which has this problem.

So the solution is get an attenuator.....
I found them on Ebay from seller henryradio

Try this link
http://cgi.ebay.com/New-25-Watt-Hybrid- ... .m14.l1262

I suggest the 10dB...
27.5W in gives 2.5W out.

Please note that they also sell 50W attenuators 6dB. 70W in gives 17.5W out.

If you contact henry radio outside Ebay you get a better price, shipping seems to be the same for 1 or 3 units.
I think we paid 15 US$ for 1.

73 Ron
VK4DD

[VK4DD]

Hi Gents,

Some US Hams and Brazil Hams have asked me if they can modify the BLF248 amp and run it on 225 Mhz instead.
Technically that can be done but a lot of mods to the PCB needs to be done and some maths also.
So it not that easy. Personally I got no interest in doing that because I have no use for this band in VK.

However I would like to point out that the BLF278 PA PCB is much more suitable for 225 MHz.
That's because the output circuit is not a tuned circuit and can operate on 225 Mhz.
The minimum frequency for this PCB is probably the top of the FM broadcast band. The max frequency is limited by the BLF278.

What needs to be changed....
The changes are mostly on the side of the input circuit because that is the tuned circuit.
The 1/4 coax on the input needs to be shortened to be a 1/4 wave on 225 Mhz and perhaps the input tracks need to be made shorter as well.
Alternatively you can solder some copper on the board to make them wider and thus lowering the impedance.
But that is about it. I would say that the BLF278 PCB is easy to modify to run on 225 Mhz.

The output power on 225 MHz is reduced to 250W as per the NXP data sheet.
For the rest you might have to change a few caps on the input circuit.
I would start with using 2/3 of the capacity and experiment from there.

73's Ron
VK4DD

[VK4DD]

Hi Gents.

Look a few postings back to download the updated schematic Revision 4. Date 24 September 2009. Or click here to download the file direct.
download/file.php?id=688&mode=view

Because of a copy and paste error C6 was shown as 39pF. But the correct value is 120pF.
Thanks to Nico PE0NYJ who found the error in the schematic.

73 Ron
VK4DD

[VK4DD]

Hi Gents,

I got an interesting email from Nico PE0NYJ. Nico has improved the design and added a temperature compensation.
The temperature compensation is based on a 10K Ohm NTC which can be obtained in most electronic shops.
As such I think it is a good improvement over the old design which was had noticeable drift.
Nico reports to have only 20mA drift in the bias between 20 and 50 degrees

Nico also reports he gets 280W out based on 2 x 25 Ohm coax.
He also tried 3 x 25 Ohm coax (2 x 25 parallel and 1 x 25 on top) but Nico reported that this combination was less efficient.

Perhaps the 2 x 35 Ohm coax and 1 x 35 Ohm on the top gave a little bit more output power.
But the disadvantage was that is was more difficult to solder.

Nico also reports that he is chasing 35, 50 coax and it looks like he is going to do more experiments to see which coax gives most output.
I have promised Nico to give him some 35 and 50 Ohm coax so he can do his experiments.

Please note that if you are using the Amp for SSB you may not need this and just keep things simple.
If you are going to use it on WSJT or FM than I suggest to do this mod.

Nico also reported to be very pleased with the quality of the PCB. After many hours of experimenting and soldering the PCB is still in good condition and
the tracks are still on the PCB .

More info on this project can be found here.

viewtopic.php?f=11&t=8497

73 Ron
VK4DD

[VK4DD]

IMPORTANT POSTING PLEASE READ

Hi Gents.

Over time the contact between the bottom of the PCB and the copper heat spreader degrade because of oxides, metallic reaction between different metals etc.
An other problem could be that some heat paste are not electrically conductive or simply too much heat paste. Resulting in poorly grounded FETS.

The result of all these poorly grounded FETS can be Oscillation. The amps takes off and is no longer stable.
This can not only damage the FET but also cause interference.

The higher the gain of the FET, the quicker you run into this problem.

Lessons learned:

Use many screws to bolt down the amplifier to the PCB. Do not try to save money or time on screws.
Use these toothed washers to improve grounding between screw and PCB material.
Have sufficient torque (but not too much that you kill the thread).
I would also recommend ground the top of the FET. Not the white part, but the flanges on the side.
This will help grounding the FET if you used non conductive heat past.
Using less heat past will help... also because the metal of the Fet is in closer contact with the heat spreader.

Note that the amp may work now without these measures but over time it might develop a grounding problem because corrosion kicks in.

How do you notice the amp is not stable...?

If you see one of these symptoms in your low voltage transistor or Fet amplifier.... clean up the oxides and improve grounding. Add more screws, remove excess heat paste.
Ground the flanges of the FET with some copper.

That is easy, just monitor the current and you will see that the amp is drawing more current than the idle current you have setup.
Or ... it is almost impossible to setup the bias current. It is either to high or to low. At a certain point it just takes off.

Information is shared because ...at the end of the day we don't want to reinvent the wheel every time.... as such it is good if we pass this on.

73 Ron
VK4DD

[VK4DD]

Hi Gents,

More balun experiments have been completed by Nico PE0NYJ.

As you know there are three dots on the PCB. With a "dot" I refer to the Island were we solder the ATC or the ARCO (variable caps) on.
It is preferred to mount the short balun leaving the first Island (closest to the BLF278) not used.

Nico calls this mounting the coax like this...the short balun.
Any way here are the results of the Experiments from Nico PE0NYJ.....

3 x 25 Ohm => Pout = 250W (short balun). (note 2 x 25 Ohm parallel and 1 x 25 Ohm on top)
2 x 25 Ohm => Pout = 270W (short balun).
2 x 1/4 inch Teflon coax 50 Ohm => Pout = 310W (short balun). @ 10.5A.and efficiency 61%.

Nico reports is changing the balun, means you have to change the ATC or the setting for your ARCO trimmer.
With the thick 50 Ohm coax he reports only 8pF.

A pretty good result from Nico showing that there is plenty room for experiment with the coax.
It is a pity that the 35 Ohm coax was not tested, to see if that gave any other results.

A warning from Nico:

Clean up your solder flux. It can get black and cause the transistor to short !
Use alcohol (or nail remover is the dirty method) with a cotton ear stick (the ones to clean your ears ).

Indeed the PCB still looks to be in good nick after all your experiments.
Thank you for sharing the results of the coax balun with us Nico.
If you would like to use the 50 Ohm coax than please remember to use 1/4 Inch Teflon.
This has to be sourced from the flea markets because new it is quite expensive. (that was sort of the reason why I used other coax )

I think Hams will also be using 25 or 35 Ohm coax to make the amplifier.
At the end of the day you can only use the coax you can get hold off...

73 Ron
VK4DD

Please contact VK4DD if you happen to have some excess stock 1/4 Inch Teflon coax.

[VK4DD]

Hi Gents.

Yet an other amp has been finished. I got an email this morning from Ralph PA3GRY.
He reports

300W @ 8A. Ralph uses a DAIWA power meter and has a fast 8A fuse in circuit.
I am using a big 35pf air spaced trimmer which looks a bit big on the circuit board but Ralph reports "it works perfect".

The coax he uses is "6mm thick 50 Ohm teflon coax"

He reports :

I am planning to use two of these boards and combine them and reduce the output power to 400W to stay within license requirements.
As such the amps don't have to work real hard.
The picture shows the amp all by it self. I had a temporary heat sink but I need to finish my big heatsink on which I still have to do some work.
I don't have much experience with high power and learn a lot from this project.

Thanks Ralph PA3GRY for emailing your results.

73 Ron
VK4DD