We had a mains spike/glitch because of rain

The 150W linear blew the gates on the MRF9180 in my Chinese board, lucky I have a complete spare linear.

I replaced the MosFet with one from Aliexpress for €9.19

It will bias up perfect, drive it with RF as normal and it consumes 8A at 26V as normal at 96.8.mhz...
Everything looks Normal

But... the RF output is only 20W

Looks like it could be a fake, I bought 3... all are the same

Even the original faulty one I took out will give me 50W if I drive it very hard, so it must be bad Aliexpress MRF9180's

This is one of my Linears

Did the devices look brand new or were they supplied cut away from an old PCB?

Bad weather and lightning is why we always use valve PAs in locations like yours. A 4CX250 would give you the power you want and be almost indestructible. The downside is the expense - bottle-based PAs and their power supplies are several times more expensive than the same power from semiconductors, but their MTBF is a totally different story.

I had an almost direct hit to an antenna on a high building in the Middle East. Lots of the antenna vapourised - the elements were reduced to gases and metal globules all over the rooftop - but the rig kept on delivering power into this bizarre load! There had even been a flashover in the valve "chimney" assembly between the anode line and the case, but it kept on working!

As an aside - back in the 1990s when the Americans finally got hold of one of the late Soviet Migs, they laughed at the valved radio gear inside the aircraft. It wasn't until someone pointed out that the glassware-based electronics would survive a nuclear EMP - and the American semiconductor-based stuff wouldn't - that they stopped sneering.....!

teckniqs wrote: ↑Mon Nov 25, 2019 6:44 pm Did the devices look brand new or were they supplied cut away from an old PCB?

Brand new with a Motorola Logo, looks genuine.

BriansBrain wrote: ↑Mon Nov 25, 2019 11:54 pm

teckniqs wrote: ↑Mon Nov 25, 2019 6:44 pm Did the devices look brand new or were they supplied cut away from an old PCB?

Brand new with a Motorola Logo, looks genuine.

I heard of other people having problems with what seem to appear to be brand new devices but the old second hand devices that have to be removed from the old cut up PCB boards all work fine.

I am wondering if it is another old different device which has had all the solder removed from the connectors and had MRF9180 wrongly printed on to it, to con people...

teckniqs wrote: ↑Tue Nov 26, 2019 7:44 am

BriansBrain wrote: ↑Mon Nov 25, 2019 11:54 pm

teckniqs wrote: ↑Mon Nov 25, 2019 6:44 pm Did the devices look brand new or were they supplied cut away from an old PCB?

Brand new with a Motorola Logo, looks genuine.

I heard of other people having problems with what seem to appear to be brand new devices but the old second hand devices that have to be removed from the old cut up PCB boards all work fine.

I am wondering if it is another old different device which has had all the solder removed from the connectors and had MRF9180 wrongly printed on to it, to con people...

They are advertised as...
MRF9180 MRF 9180 RF TRANSISTOR ORIGINAL NEW
Seller: AIICKING Store
The advert also states: 9978 pieces available
The picture used in the advert is this... And that is what I received, they looked... and are Brand New and Unused

Only difference was the production number printed on the case under the MRF9180
Which in my case is... RR1627E

I did a search on Google for MRF9180 FAKE and found this...
https://ask.oberlo.com/d/34-aliexpress- ... acklist/33

Then I did a search on that page for MRF9180 and found this... Different company name but who knows

The fake ones are usually easy to spot, they have poor amateurish printing of the logo and code.

I have one of those "Aiicking" ones somewhere (although I cant find it now), I remember when I bought it, I tested it with the meter on diode test and the readings where slightly different from the original used ones I had.

If I find it, I will pop it in an amp and see what wattage I can get. Usually with an original device fitted in a chinese kit (with modified feeders and reworked low pass filter). I set the Idq to 0.8A and get 150W @ 25v input.

s2000 wrote: ↑Tue Nov 26, 2019 2:34 pm The fake ones are usually easy to spot, they have poor amateurish printing of the logo and code.

Here is one of the AIICKING ones - looks genuine printing to me With all the other MRF9180's I have tried in these Chinese boards...
I get +/- 145W output measured using a Bird 4304A with 26V power supply @ 8A drain with bias set to +/- 325mA
Which = +/- 70% efficiency

These AIICKING ones (and I tried 3) only gave me +/- 20W RF output,
but looked normal as far as RF drive and a current consumption of 8A

If these are good Motorola mosfets I can't see what is dropping the power

This is the only schematic I can find similar to this design

Hi, I have just found the one I bought off Aliexpress (it was giving me an aiicking head not being able to find it )

The printing is admittedly quite good (compared to some I've seen) but there appears to be one thing in common that all the fakes have and none of the originals have. Where it says "MRF9180", on the fakes, the "0" has a diagonal line through it. All the originals I have (from the kits with used mosfet), there is no line through the "0" (both motorola and freescale). Well it seems that way on the ones I have here anyway.

Maybe if anyone else has the kit with used mosfet they could check?

That's not to say a fake wouldn't work in the circuit, if I get a bit of time I will try it out.

What i think there is a problem in the output matching or LPF (lots of current and low output)
Maybe a ATC capacitor is blown

s2000 wrote: ↑Tue Nov 26, 2019 5:35 pm Where it says "MRF9180", on the fakes, the "0" has a diagonal line through it. All the originals I have (from the kits with used mosfet), there is no line through the "0" (both motorola and freescale). Well it seems that way on the ones I have here anyway.
Maybe if anyone else has the kit with used mosfet they could check?

You are correct
Here are pics of Motorola and Freescale with NO line through the "0"

RF-Head wrote: ↑Tue Nov 26, 2019 6:01 pm What i think there is a problem in the output matching or LPF (lots of current and low output)
Maybe a ATC capacitor is blown

Nah... the rest of the boards I have checked out OK

Example...
If I put back the original mosfet (with the gates blown) and drive the input to to linear hard, about 10W I get +/- 50W output.
Drawing only 4A @ 26V
Which proves the board is OK

Here is the blown MRF9180 that gives 50W out @ 4A @ 26V if I drive it HARD
And look... ... it has a line through the "0"
Now I'm baffled

On the last picture there is a PCB track gone from the 10E resistor to the left gate from the fet
also it doesnt prove that the output matching is good! that you get 50W with 10W drive is that you burn the gate of the MRF9180 and that you get a better match with a broken mosfet and not with a good one
Please look at the gate resitor pcb track and look if you get bias voltage on both gates

RF-Head wrote: ↑Tue Nov 26, 2019 9:18 pm On the last picture there is a PCB track gone from the 10E resistor to the left gate from the fet
also it doesnt prove that the output matching is good! that you get 50W with 10W drive is that you burn the gate of the MRF9180 and that you get a better match with a broken mosfet and not with a good one
Please look at the gate resitor pcb track and look if you get bias voltage on both gates

Before the electric glitch that caused the gate problem on this mosfet the output was a full 150W.
The little bit of print has been missing for ages, well before this problem.
With this old mosfet when it was in good condition or even these *new* ones the bias voltage was OK and equal for both gates.
The bias adjust seems to work as normal with the new mosfets.

BTW: I have tried bypassing the complete LPF circuit to see if that was the problem with the new mosfets...
... it was not, still only +/- 20W output.

OK...

Received another MosFet ... this time from eBay - Seller: xinye0616
This one cost 14.66 EUR - post free
*Note* ... NO line through the "0"
I will be posting the test results later

As RF-Head noted

On the last picture there is a PCB track gone from the 10E resistor to the left gate from the fet

The print has gone = True
But it had a thin wire repair job attached to the transformer that you can't see in the picture (probably covered in flux )

OK... put the newly arrived MosFet in

And...

Perfect... @ 26V & 8A = +/- 150W output

BriansBrain wrote:As RF-Head noted

On the last picture there is a PCB track gone from the 10E resistor to the left gate from the fet

The print has gone = True
But it had a thin wire repair job attached to the transformer that you can't see in the picture (probably covered in flux )

OK... put the newly arrived MosFet in

And...

Perfect... @ 26V & 8A = +/- 150W output

IMGA0706.JPG

Right so the fakes are only giving out 20 Watts. The reals are giving you the full wack. Ill have to get one.

Sent from my AMN-LX9 using Tapatalk

*Please NOTE*
My findings are...
The fake one that only give 20W output have a production number under MRF9180 of >>> RR1627E

fake ?
https://www.aliexpress.com/i/32943270648.html
https://www.aliexpress.com/i/32816018892.html
https://www.ebay.com/itm/1PCS-New-MRF91 ... 2998904155
https://www.aliexpress.com/item/4000846835989.html