MOSFET conversions for upD858 and MB8719 boards

[Robb]

Using a zener - like Galaxy radios use?
Sounds like a plan.

I run the DX-959 @4v; gets great reports for clarity and crispness w/heatsink = 24w+.

 

[Robb]

This MOSFET replacement you posted has proved to be an inexpensive replacement solution compared to the CB-20 by Eleflow ($20!). Cost to replace all parts was less than $3 for me - if you make the biasing device yourself. I made one but didn't use it - BTW.

Removing/replacing a couple of parts wasn't that hard to do. The old glue/shmegma on the board made it hard to find L40 tho; so I traced it out using a schematic from CB Tricks.

 

[Robb]

Whoa, hold on a tick....

Just as well that I didn't use the biasing device I made, read this:

http://www.worldwidedx.com/general-...anion-cbtricks-errors-drawing-correction.html

Since PsiDOC did the math, I used his formula of .000185 / (desired bias voltage) = ~47k ohm resistor for the value of R2 in the EN369FN biasing device for a 4v gate voltage. 20k ohms will give the more desirable 3.75v gate voltage, as shown in the above graphic.

Soooooo........Using a 100k ohm VR @ R2 in parallel with a 20k ohm fixed resistor will make the gate voltage variable for the EN369FN?
That's a guess BTW; but it seems logical tho . . . OR adding a 100k ohm VR between the gate and source pins using the EKL EN369FN device . . .

 

[Robb]

Sorry PsiDoc; you cannot be correct.

I experimented by trying a 100k ohm VR across the EN369FN device. It cannot be 20k ohms for R2. The VR did work; but as high as the gate voltage would go was 3.5v with max resistance from the VR. So R2 would have to be greater than 100k ohms . . .

 

[ExitThirteen]

100k for R2 would be a sufficient value. That coupled with the 100K variable will give good results. Let us know how you get on.



~Cheers~

 

[Robb]

Took out the EKL part and measured the resistance putting the neg lead on the positive pole to read just the resistance of R2. It was 19.5k ohms just like PsiDoc said; so he was right! Then I put the pos lead on the pos end of the device and read 21.5k ohms.

Now the fwd resistance thru a 1n4148 diode is ~180k ohms when tested. So in parallel with a 20k resistor, it should read 100k ohms. What is it that I don't understand? Doesn't make sense; what did I miss? What about the diode resistance; is that where I got the wrong picture?





Don't have any 100k resistors ATM . . .

 

[Robb]

Made another homebrew EN369FN using a 22k resistor for R2.
After testing it, it has the same/exact values forward/reverse as the OEM EKL part.
PsiDoc was right on the money . . .

 

[Robb]

Tested the MOSFET/Cobra 148GTL with the Autek WM-1 watt meter and 50 ohm dummy load:

AM: 5w dead key swinging to 20w+.
SSB: 24w

The MOSFET barely gets warm with short keying. If you key up for 1 min, it does get a bit warm. That is with the bias at the stock EKL device/3.85v setting. Didn't get any 100k resistors yet; prolly tomorrow. Looks fine on the scope - too . . .

 

[Heatwave1972]

Update:

MOSFET conversion for MB8719 board (Single conversion board, found in Uniden Washington, President McKinley, Realistic TRC-450, Cobra 142GTL, etc)

Parts you will need:
IRF520 MOSFET
EKL companion part EN-369FN
56k ohm 1/2 watt resistor
470pf disc or mica capacitor
68pf disc or mica capacitor (optional, see installation text)
small jumper wire

Installation:
Remove old transistor at TR41
Remove R163
Remove L34
Install jumper wire at R163
Install IRF520 at TR41
Install EN-369FN across final, positive side to gate, unmarked side to source
Install 56k ohm resistor from gate of final to far side of L35 (furthest side away from final, 13.8V)
Install 470pf capacitor across C173
Install 68pf capacitor across C148 (Optional, will bump up output in some radios, check without first)
Check voltage at gate of final, should be approx. 3.85V + or - .1V
Retune TX

Done! You should have same output as before, if not more. Did this on my Uniden Washington, was doing 21W before conversion, does 26W after conversion. ~Cheers~

exit thirteen where does the 56k go excactly?

 

[Heatwave1972]

exit thirteen where does the 56k go excactly?

from the far side of L35

 

[Birddog2009]

Exit thirteen why does the mod sound trashy in my 858 SSB rigs? PM me or post on here or text me and tell me what i am doing wrong.

 

[ExitThirteen]

What's your voltage at the gate of the mosfet? Should be around 3.75V or so.

Sounds like a low voltage condition. Is the bad audio in all modes, or just one mode only?



~Cheers~

 

[M0GVZ]

So you're doing all of this to get an extra 6W or so? What is the point? The receiving station can't tell the difference. Even if you double the power its only half a S point increase. You'd have to increase the power fourfold to be remotely worthwhile.

 

[Robb]

So you're doing all of this to get an extra 6W or so? What is the point? The receiving station can't tell the difference. Even if you double the power its only half a S point increase. You'd have to increase the power fourfold to be remotely worthwhile.

As it was stated earlier, the reason is not so much the small gain in power; but the availability, time, and cost factor. I can tell you firsthand, that using a commonly found IRF-520 and a few paltry parts and taking very little time, one can make use of that substitution for less than $3 as opposed to using an Eleflow for $12 plus shipping and the loss of time waiting for it to arrive in the post. Not to mention it works like a treat, and the power gain is just so much more icing on the cake.

 

[GnG8d]

So you're doing all of this to get an extra 6W or so? What is the point? The receiving station can't tell the difference. Even if you double the power its only half a S point increase. You'd have to increase the power fourfold to be remotely worthwhile.

That's about what your Moxon does over a dipole, and the mosfet conversion takes considerably less time.

But like Robb said, it's mostly a parts availability issue in trying to keep some of these radios going. I know some will remove a perfectly good 2sc1969 and replace it with an irf520 looking for more power, but most realize how insignificant that power gain really is.