Topics

BITX40 output power at 36 volts to the IRF510 #bitx40help #bitx40

John Norris
 

I would like to run the IRF510 in the BITX40 at 36 volts. If I do this should I be measuring about 40-45 watts output power?
Are there components that I need to change out besides going to a bigger heat sink or fan?


John Norris
ki4ucn

Vince Vielhaber
 

No, you should be watching the 25V cap on the PA power line fly across the room.

Vince.

On 05/16/2018 04:05 PM, John Norris wrote:
I would like to run the IRF510 in the BITX40 at 36 volts. If I do this
should I be measuring about 40-45 watts output power?
Are there components that I need to change out besides going to a bigger
heat sink or fan?


John Norris
ki4ucn

Ghericoan
 

You a decent heatsink. I could post pictures of what I have for a setup of you'd like. 

Also, replace those 25vdc caps or they will make nice foreworks. I replaced them with a set of 200vdc caps cause ita what I had on hand.

On Wed, May 16, 2018, 4:05 PM John Norris <jlnorrisjr@...> wrote:
I would like to run the IRF510 in the BITX40 at 36 volts. If I do this should I be measuring about 40-45 watts output power?
Are there components that I need to change out besides going to a bigger heat sink or fan?


John Norris
ki4ucn


--
----------
N5WLF, Greggory (or my nickname, Ghericoan)
General Class, Digital Radio Hobbyist

Jerry Gaffke
 

Let us know if you ever get to anything like 40 Watts.
Might be possible on the low bands like 80m.
But:
    Board layout may mean that the extremely low RF signal levels won't happily coexist with your 40 Watts.
    The 2n3904's may run out of steam and not drive the IRF510's hard enough, already happens on the high bands..

In addition to the 25v electrolytics, I'd guess you'll need to improve a bunch of transformers and inductors and caps as well.
Especially that output transform from the IRF510's and the band specific low pass filters.
And probably should nix the choke scheme for feeding power to the IRF510 drains.
A good starting point might be the WA2EBY amp.

Better yet, leave the uBitx as is and use it to drive a WA2EBY amp.

Good luck.

Jerry


On Wed, May 16, 2018 at 01:05 pm, John Norris wrote:
I would like to run the IRF510 in the BITX40 at 36 volts. If I do this should I be measuring about 40-45 watts output power?
Are there components that I need to change out besides going to a bigger heat sink or fan?

Satish Chandorkar
 

You will have re calculate and rewind the output transformer because all the calculations of this transformer depends on the operating voltage. As you are going to increase the voltage on the IRF 510's from 12 V to 36 V the drain impedance of the IRF's changes changing the impedance ratio of the output transformer. 

On Wednesday, 16 May, 2018, 6:58:03 PM GMT-4, Jerry Gaffke via Groups.Io <jgaffke@...> wrote:


Let us know if you ever get to anything like 40 Watts.
Might be possible on the low bands like 80m.
But:
    Board layout may mean that the extremely low RF signal levels won't happily coexist with your 40 Watts.
    The 2n3904's may run out of steam and not drive the IRF510's hard enough, already happens on the high bands..

In addition to the 25v electrolytics, I'd guess you'll need to improve a bunch of transformers and inductors and caps as well.
Especially that output transform from the IRF510's and the band specific low pass filters.
And probably should nix the choke scheme for feeding power to the IRF510 drains.
A good starting point might be the WA2EBY amp.

Better yet, leave the uBitx as is and use it to drive a WA2EBY amp.

Good luck.

Jerry

On Wed, May 16, 2018 at 01:05 pm, John Norris wrote:
I would like to run the IRF510 in the BITX40 at 36 volts. If I do this should I be measuring about 40-45 watts output power?
Are there components that I need to change out besides going to a bigger heat sink or fan?

Vince Vielhaber
 

I just looked at both of my B40s, they both have a 2N2219A (metal can) with a heat sink attached.

Vince - K8ZW.

On 08/04/2018 07:30 PM, Timothy Fidler wrote:
One concern that I have is
that Farhan is publishing Q4 as a 2n2219A metal case npn but what shows
on the photos is a T39 case RF driver.. you take my point. If the
physical entity and the schematic do not agree it is pretty involved to
re-engineer a physical board. I suspect that the driver is a 2N3886.
Please inspect and advise before you contact me. TEF

Timothy Fidler
 

Thanks for that.. that means my rework idea will work .. probably .. at 7mhz it is all smoke and mirrors ... I looked at the photos and even went to an online video and I would swear what I am seeing is a T39 can in the heat sink. by comparison in size with the T220 for example. Possibly some bitx40s came out with the more capable RF Power transistor (NOT a 2n3886  as per below - I got that wrong - I will check and advise the part number) and when they ran out they went back to the lightweight device.   

Timothy E. Fidler : Engineer BE Mech(1) Auckland , NDT specialist AINDT UT /RT3 , MT2 CB #2885, 
Telephone Whangarei   022  691 8405
e: Engstr@...



----- Original Message -----
From:
BITX20@groupsio

To:
<BITX20@groups.io>
Cc:

Sent:
Sat, 4 Aug 2018 19:37:35 -0400
Subject:
Re: [BITX20] BITX40 output power at 36 volts to the IRF510 #bitx40 #bitx40help



I just looked at both of my B40s, they both have a 2N2219A (metal can)
with a heat sink attached.

Vince - K8ZW.

On 08/04/2018 07:30 PM, Timothy Fidler wrote:
> One concern that I have is
> that Farhan is publishing Q4 as a 2n2219A metal case npn but what shows
> on the photos is a T39 case RF driver.. you take my point. If the
> physical entity and the schematic do not agree it is pretty involved to
> re-engineer a physical board. I suspect that the driver is a 2N3886
> Please inspect and advise before you contact me. TEF

--
Michigan VHF Corp. http://www.nobucks.net/ http://www.CDupe.com/
http://www.metalworkingfun.com



Timothy Fidler
 

On Sat, Aug 4, 2018 at 04:30 PM, Timothy Fidler wrote:
Not only will the capacitor let go but the IRF will open up and spray silicon and electron smoke about.  THe WA2EBY amp runs 40W max flat out at 24 DC supply and Kossor admits in his  notes (part 2 of that article ) that the IRFs will probably only last a couple of hundred hrs given the hammering they are getting.  Now you are trying to make one off IRF 510 in class A at say 35 perc eff outwork TWO IRFs in class B with a lot of professional design work behind them... it is not going to happen. The silicon temperature will exceed 150 degrees if you were at that power level. The IRF has a  therm resist of 3.5 W/K Junt to case and regardless of what heatsink you bolt on  (CW mica washer of  thermal resistance of ...4.7 W/C all up). you will still have that as your major thermal resistance. see https://www.fairchildsemi.com/application-notes/AN/AN-4166.pdf   If you want to  mod the design to give near 15W all out with a fair bit of rework pse contact me via PM but you will have allow say USD 25 for bits and pieces to do it (but it would be fun) . One concern that I have is that Farhan is publishing Q4 as a 2n2219A  metal case npn but what shows on the photos is a T39 case RF driver.. you take my point. If the physical entity and the schematic do not agree it is pretty involved to re-engineer a physical board. I suspect that the driver is a 2N3866.  Please inspect and advise before you contact me.  TEF 

Timothy Fidler
 

PN  is 2n3866 1.5 W RF NPN silicon but based on recent reply it seems the smaller metal cased part was used as the pre driver on many boards.

ajparent1/KB1GMX
 

A serious answer is that at 36V the IRF510 has no SWR latitude the slightest
bump or bad SWR and they disappear.  Stay  in the 24-28V is far more reliable
assuming its RF stable and doesn't oscillate which is not a safe bet from what
I've heard.

With 2W drive a pair of IRF510s in a good design should hit about 50-60W at 28V.
That's what the WA2EBY amp can do.  You need the board layout to support that.

The driver should be improved but at 7mhz the 2n2219 is a good part.

Watch for parts that can't stand the DC voltage.

Generally doing that with the BITX40 is not going to fly long as the
low pass filter parts are not up to it.

Allison

K5ESS
 

One version of the 2N2219 comes in a TO-39 package.

Mike

K5ESS

 

From: BITX20@groups.io [mailto:BITX20@groups.io] On Behalf Of Timothy Fidler
Sent: Saturday, August 4, 2018 7:07 PM
To: BITX20@groups.io
Subject: Re: [BITX20] BITX40 output power at 36 volts to the IRF510 #bitx40 #bitx40help

 

On Sat, Aug 4, 2018 at 04:30 PM, Timothy Fidler wrote:

Not only will the capacitor let go but the IRF will open up and spray silicon and electron smoke about.  THe WA2EBY amp runs 40W max flat out at 24 DC supply and Kossor admits in his  notes (part 2 of that article ) that the IRFs will probably only last a couple of hundred hrs given the hammering they are getting.  Now you are trying to make one off IRF 510 in class A at say 35 perc eff outwork TWO IRFs in class B with a lot of professional design work behind them... it is not going to happen. The silicon temperature will exceed 150 degrees if you were at that power level. The IRF has a  therm resist of 3.5 W/K Junt to case and regardless of what heatsink you bolt on  (CW mica washer of  thermal resistance of ...4.7 W/C all up). you will still have that as your major thermal resistance. see https://www.fairchildsemi.com/application-notes/AN/AN-4166.pdf   If you want to  mod the design to give near 15W all out with a fair bit of rework pse contact me via PM but you will have allow say USD 25 for bits and pieces to do it (but it would be fun) . One concern that I have is that Farhan is publishing Q4 as a 2n2219A  metal case npn but what shows on the photos is a T39 case RF driver.. you take my point. If the physical entity and the schematic do not agree it is pretty involved to re-engineer a physical board. I suspect that the driver is a 2N3866. (mark II correction)   Please inspect and advise before you contact me.  TEF 

Timothy Fidler
 

Jerry .. bit  of confusion. He is talking re uBitx40 and you are not.

jah12@...
 

Have a look at a 25 Watt amp by VK3XU at http://www.philipstorr.id.au/radio/seven/five/25%20Watt%20Amp.pdf
Looks a bit simpler to build than the WA2EBY amp.
--
John VK6JAH

ajparent1/KB1GMX
 

>>Have a look at a 25 Watt amp by VK3XU at http://www.philipstorr.id.au/radio/seven/five/25%20Watt%20Amp.pdf
Looks a bit simpler to build than the WA2EBY amp.<<

It is simpler and doesn't work near as well.  Also it bets on the mosfets having
the same gate voltage threshold or a hand picked pair.  If they are not one will
run hotter and the 2nd harmonic output will be higher (cleans up with a filter).

Allison