EMRFD Message Archive 15017

Message Date From Subject
15017 2018-08-18 12:17:18 Nick Kennedy 70 W Chinese MOSFET linear amp kit
I read about these things on one of the QRP or homebrew lists and had to give it a try, at < $19 postpaid.  I think I got this one:

A big part of the challenge is that there are no instructions or even a schematic supplied with the box -o- parts. But of course, some brave souls have built them and reported on what they learned on-line.

The PD7MAA blog has a schematic and nice summary of his experiments and successes with the kit:

DK9JC also offers some good info:

And a YouTube video from KC2IRV shows testing at power and some good views of construction and of the schematic:


OK ... I've built mine and done a bit of testing. It's not working yet so maybe I can get some suggestions. Part of the problem is that I'm moving to a new QTH in a month and most of my parts and test equipment are in boxes.

Anyway, I adjusted the bias as shown on the schematic at first, for 2.7 VDC, which didn't put me near the threshold. Based on the experience of others, I turned it up until current draw started moving, at around 3.6 VDC with the whole board drawing 130 mA.

But when I apply 2 W of carrier from my KX3 with the amplifier connected to a dummy load (no LPF yet), I saw very little RF output. The DC current went to 7.4 A for an input power of 96 W. Where's it going? I don't think I've blown the MOSFETs, but I'm not 100% sure. I didn't keep the key down for long, for sure.

This circuit looked on the surface like pretty standard topology, but on closer inspection and comparison with published push-pull MOSFET amps, I see that there's no resistance at the input, on either side of the transformer to "swamp" reactances and otherwise provide a load to work into.

The only significant thing the driver is connected to other than the gates is the two feedback resistors. Note that they are shown as 150 ohms on the schematic, but my supplied resistors were 100 ohms.  I suppose that this is negative feedback and that the smaller resistors would reduce the gain(?)  And also that the feedback circuit has a great effect on the impedance seen by the driver.

OK, that's my tech talk bit. Some other notes on the kit:  First, the MOSFETs have had their markings sanded off. Others said they're IRF530 types, but I can't be sure of these. I bought some IRF530s expecting to have some failures, but haven't tried them yet.

Also of note - my kit had all the surface mount parts pre-soldered, even though the ad didn't mention that. That was a nice bonus. It also included insulating pads and shoulder washers for the MOSFETs,  which some  others didn't get with their kits.

One has to mention that even if this turns out to be a good amplifier, the user must supply the big heat sink and the LPF or LPFs, depending on the band or bands to be used. It includes a switch-around relay though, which is nice.

Finally, I always enjoy a little Chinglish with my eBay purchases. One note on the carton said, "Warning: the output is forbidden to open circuit, short circuit, the Prohibition of Bobbi 2 or more."

I assume Bobbi is SWR.


Nick, WA5BDU

15019 2018-08-20 10:48:36 kb1gmx Re: 70 W Chinese MOSFET linear amp kit
If you getting high DC input power and no RF for the RF drive stop!

The devices are amplifying but you have some kind of output circuit issue.
The likely case is a short!  Its possible you have a short in the  transformer
or the transformer is in wrong.  

Also 5W is way too much drive without an attenuator at the input.

I have the WA2EBY amp (look in files)  under my call HF-USE-FET.

That uses the IRF510 push pull.  Mounted and wire up right with all the 
though hole (for rivits or z wires) plus the edges wrapped to insure 
top ground and bottom ground are one.  Since the device makes 
power at 6M and has gain of about 13db there treat it like a VHF 

For 2W I get 55-60W out at 40M at 28V,  5W would fry the fets.
with 4W drive during testing I got 85W which is too close to 
the max for the devices and the output was over driven.
At 10M its still doing 37W.  Its proven reliable for the local use
with my QRP radios.   The attenuator at the iput does two 
things allows radios with 4W fixed output to not over drive the 
amp and it presents an acceptable SWR to radios like the 
FT817 and Argonaut 505.

15238 2019-02-19 14:08:28 Nick Kennedy Re: 70 W Chinese MOSFET linear amp kit
Responding to my own post six months later. I build the little amplifier right before moving the household to a new QTH 150 miles distant, so it's taken a while to get back on it.

I think part of my problem may have been with power measuring instrumentation, because I decided to give it another go before troubleshooting and I was pleased to see good power output.  

My initial tests involved driving it with my KX3 using the ATU to get a decent match. It was pretty obvious that the circuit lacked some type of resistive loading or swamping on the input so I followed advice from OE1CGS and installed 22 ohms on the MOSFET side of the input transformer - i.e. gate to gate.

That dropped the SWR to well under 2:1 on bands 20 m and below (all I checked) but the gain stayed surprisingly high. 

1 W input, Vcc = 13 V:
















I still needed to tie in an LPF and I did hook up a 5-element 40 m LPF and still see ~45 W out with 1 W drive on 40.

Getting close to QSO time!  Of course I need to drive it with something homebrew.


Nick, WA5BDU