The MRF1K50N

I was searching Mouser for a new LDMOS recently, as my trusty ole 1k80h finally went “out of round” on me. In the last few years, I’ve soldered and unsoldered this device four times and used it in countless amp designs.

I’ve certainly TRIED to let the smoke out of it but it lives on. However, through all the abuse and tests, it’s degraded and is now too unbalanced. With the exact same Vgs, there’s now a 37% difference in Idq between the two sides.

With what I’m doing here, this is not going to cut it.

So, off to Mouser I went and I quickly came across this guy, the MRF1K50N. Which is $165 UDS.

Two 8 minute round trips later, one to desolder the 80h and another to solder the 50N and I was in biz.

Testing the MRF1K50N
Fresh from Mouser: MRF1K50N

Prepping the area – I’ve used this copper just a few times…

Guides for when it drops down, including the score

Solder pattern bed under device

Heating up on the bed of solder

Soldered down and cooling off

So how does it perform?

First, soldering it down is a bit more tricky. The contact patch is pretty small compared to the flanged devices and as soon as it melted, it started moving. I had to hold it with tweezers until it solidified (5-10 secs).

The balance of the two transistors was much better but not perfect (13% difference vs 37%). This surprised me since these Gemini devices are obviously made together under all the exact same conditions and on the same substrate. Maybe there are some that match, but not this one.

It performs very well. If you put the datasheets up for the 188, 50N and 80H, you will see that this MRF1K50N device is right in the middle. I’d say that’s exactly how it acts.

Last note, the gate and drain flanges are slightly larger and slightly lower to the bottom. So it’s not exactly a drop in replacement, but pretty damn close.

That’s all I got, you guys can pull the datasheets if you want to see the specs and three capacitances.

Comments

11 responses to “The MRF1K50N”

  1. Vinny Stipo Avatar

    Hi Rob,

    How are you measuring IDQ difference between sides?

    I’m using MRFX1K80N right now, and you’re right, it’s a bit tricky, the pad is much smaller.

    Also, I’m looking into BLF189XRB, from test below, apparently extremely low harmonics, very linear.

    https://www.youtube.com/watch?v=GOAJ9KuiXJ4

    Anyways, keep it up, I enjoy following your projects.

    73
    -KM2W

    1. N4GA - Rob Avatar
      N4GA – Rob

      Thanks Vinny.

      I’ll check out those two devices, but unless the laws of physics have changed, the harmonics are not magically going to be -60dB!

      That looks like a tuned class C FM amp he’s jumping up and down about. Not going to work for a class AB broadband amp that needs to be linear for SSB, and can’t be tuned for one frequency.

      But I like to see all the links, it’s fun and you never know, you just might learn something, so thanks for sending.

      I measure Idq individually by lifting the input xformer and inserting the the ammeter inline to each drain.

      73s!

  2. Alex EW3KO Avatar
    Alex EW3KO

    Hi Rob.
    What are your successes with this transistor?
    Probably there are some statistics already.

    1. N4GA - Rob Avatar
      N4GA – Rob

      Alex – so far so good. Behaves roughly as the BLF188 does. I run my amps a little different from a lot of people that run a single device amp at 1-1.4kW.

      My max is ~733W because below here you can stay in pure class AB. Can’t really speak to maxing this device out.

      73!

  3. Tom Avatar
    Tom

    I have been reviewing all of your articles, a very interesting find for me, and have enjoyed reading them.
    However, of all the device parameters pointed out I see Pout, Pin, Eff %, etc., but no mention of IMD?
    I would think that this would be a most important parameter to test for.

    Keep up the good work and a Happy and safe NY 2021

    1. N4GA - Rob Avatar
      N4GA – Rob

      Hi Tom,

      The short answer is if you design an amp the way I do, the IMDs are as low as the LDMOS will support…and all significant IMDs are in-band.

      People like to measure them, look at the lobes and talk about them, but not one of them has a single clue how to affect them.

      I’m going to write an article on this for the site some day!

      Thanks for the note. 73!!

  4. Lee Johnson Avatar
    Lee Johnson

    I am retired, and looking into building an LDMOS amplifier.
    Still in the research mode, and look forward to hearing more from you.
    N4WYE, Lee

    1. N4GA - Rob Avatar
      N4GA – Rob

      Let me know if I can help and 73!

  5. Ed Erlemann Avatar
    Ed Erlemann

    MRF1K50N plain seems no longer available in Mouser neither Digikey, they have others but higher than $250. Can you suggest any other good option? Thanks.

    1. N4GA - Rob Avatar
      N4GA – Rob

      ART2K – sent you an email.

Leave a Reply

Your email address will not be published. Required fields are marked *