Mirage mx-36HP3 Q39 ?

[Hawkeye351]

Got a Mirage mx-36HP3 on the desk today. Board number EPT-DX2912Z.

Issue:
Transmit audio has constant oscillation sound, sounds like background air noise and can't be brought down with mic gain off.

Evaluation:
Q39 looked burnt under it on the board.
Removed Q39 and noticed it's the wrong part. The part someone put in that spot is a 2SC2076 fet, but normally that limiter is a 2SC945. This would explain the burnt spot on the board under it. Put the 2SC2076 in the tester and it tested fine as a BJT with an Hfe of around 120.

Anyone ever heard of switching the limiter (C945) to a different transistor? If so, what's the purpose?

Gonna place a 945 in there and try it out.

 

[Hawkeye351]

According to Alltransistors database, the 2076 will sub for a 945. Regardless, I'm putting in a 945.

 

[Hawkeye351]

That fixed it, although I think the 2076 just wasn't making a good connection.

Output:
Variable power wide open yields a 9w deadkey, swings to 40w PEP, 20w average.
Variable power all way down yields a 1w deadkey, swings the same as above, hmmm, swing mod somewhere in it.

Anyway, it's fixed of the issue.

 

[Eldorado828]

That fixed it, although I think the 2076 just wasn't making a good connection.

Output:
Variable power wide open yields a 9w deadkey, swings to 40w PEP, 20w average.
Variable power all way down yields a 1w deadkey, swings the same as above, hmmm, swing mod somewhere in it.

Anyway, it's fixed of the issue.

Good work!
I don't think yours has a swing mod, they don't need it. I have one that occasionally gets swapped in to the work truck and just as yours it swings away without help, I know because I set it up.

I enjoy reading your write ups, thanks for sharing with us.

 

[Hawkeye351]

Was gonna set the bias, can't check it in ma mode due to AM only, so had to try the voltage route.

When I hooked volt meter to test point for driver bias, it was set around 4.6v, the finals were set around 4.4v to 4.8v. Anytime I had to go the voltage route for MOSFETS, I've set them for 3.80v (mostly 520s). So I set these for 3.80v, but PEP dropped dramatically to only 28w max. These are RT1 mosfets, so what should the bias voltage be for the gate to function as it should? Obviously 3.80v is to low for an RT1. The radio wasn't running hot or anything with it set high as it was before I set it, so what's the round about voltage for the RT1 bias settings?

 

[Eldorado828]

Was gonna set the bias, can't check it in ma mode due to AM only, so had to try the voltage route.

When I hooked volt meter to test point for driver bias, it was set around 4.6v, the finals were set around 4.4v to 4.8v. Anytime I had to go the voltage route for MOSFETS, I've set them for 3.80v (mostly 520s). So I set these for 3.80v, but PEP dropped dramatically to only 28w max. These are RT1 mosfets, so what should the bias voltage be for the gate to function as it should? Obviously 3.80v is to low for an RT1. The radio wasn't running hot or anything with it set high as it was before I set it, so what's the round about voltage for the RT1 bias settings?

There is another version of that radio, the superstar ss-36hp3 which is identical other than the fets being rt1 instead of 520's. I posted about them both and compared here in a wwdx thread I'd have to dig up.

As to the bias, seems like you'd be better off with the lower bias voltage to play it safe. The rt1 info is still something of a mystery.

 

[Hawkeye351]

I set the driver bias to 4.25v and both the finals to 3.80v each. Reset carrier levels for RF power all the way up to a 9w key and all the way down to a 1w key. Replaced the modulation VR inside the radio due to other one fluctuating badly when adjusting. Added a 22k ohm resistor across the leads of C158 to prevent the AMC from being so aggressive.

Output now:
RF Power varies from 1w to 9w.
Max PEP swing is 53w
Max peak swing is around 35w
Max average swing is around 30w

Spectrum analyzer shows no harmonics, spurious emissions, etc...of any kind using span levels of 50khz, 450khz, 1mhz, 10mhz, 30mhz and 60mhz. Slight splatter when looking at a span of 50khz, and a slight tiny bump on 54mhz when looking at a span of 60mhz.

Radio runs cool to warm when talking long periods with key at 9w, finals/driver don't even come close to getting hot, and the regulator stays just warm, never gets hot. Locals say it's loud, but still crystal clear. The owner just heard it and is ready to get it back.

Thank you Eldorado, I'll have to look into that superstar version also.

On to the next one...