Galaxy Dx-959 No Am Modulation/power/transmit

[ExitThirteen]

Robb, by "companion part", are you referring to the EKL companion part (EN-369FN) or the 2SA1012 that goes along with the AM pass regulator transistor?

The EKL companion part stays the same. However, if a person was going to put the 13N10's in a 949/959 and make it a dual final radio, the AM pass regulator circuit would have to be modified to accept something stronger than a 2SB817/2SA1012 pair. I don't know what an upgrade to this pair would be, since I don't play radio much anymore because of work, but I DO know that with just the IRF520N's running in dual final configuration, the 2SB817/2SA1012 pair tends to get a bit warm.


~Cheers~

 

[Robb]

Well. I was thinking about Q55 as well as Q54 (2SB817).


Why?
What did you have in mind for upgrading that circuit?

 

[ExitThirteen]

I'll try to get a little time and do some research on that this weekend, Robb. I'll post my findings. This REALLY makes me wish I had a 949 or 959 mosfet version to work on... there's so many things I have in mind that I'd like to try with the radio.

~Cheers~

 

[sonoma]

Robb I am glad you did some looking around and not just take my word on the mosfets. I have only done 2 cobra 29's so far and they swing very easy and they do not get hot so far. I bought a few more here awhile back and am planning on trying them in a 148 to see how it does and see what it sounds like on SSB. that is the one thing I have wondered about. on my older radios with the 2sb754 I change them out any way to the 2sb817 or a tip36c if it will fit. the 36c is a little larger than the 2sb parts and the caps on the board make the 36c hard to fit in place. some times I pull the 2 little caps and put them on the bottom side of the board. I was thinking about putting a driver and mosfet in my 959 and see how it acts one of these days just for fun.

 

[Robb]

After looking at the 959 schematic again, I can see that it already has a 2SA473 in there @ Q55. Can't see much room for improvement over that part, except for perhaps a TIP42C. Looking at the modulation circuit, I cannot see what else might be beefed up, unless the traces themselves aren't up to the current draw/needs. Doubtful. But then again, some of the traces on a 959 have been known to be marginal . . .

 

[ExitThirteen]

Ok guys, After doing some studying, I have come up with a pair that should work. Use a TIP36C at TR51 (Q54 in the 949/959 chassis) and a 2N6109G at TR50 (Q55 in the 949/959 chassis). This is a MUCH beefier setup and should withstand the extra current draw from a pair of 13N10's with no problems. Since I have no 949 or 959, I can't confirm that this will work in those radios, but it DOES work in the older 33/44/55/66/73/77/88/99 chassis quite well, so I don't see why they wouldn't work in the 949/959 radios.

If someone has a 949 or 959 they can perform these replacements on, do so and give me some feedback, please.

(Edit) a TIP42C will also be an upgrade to Q55 too, and is similar to the 2N6109G.

~Cheers~

 

[sonoma]

Exit13 if you read a post I did earlier you will see I already use the tip36c in Q54 and have used it now for over 2 years. my friend that has a shop I have him using them and he has not had one to fail so far. the only problem I have ran into on some units the 2 disc caps interfear with the TIP36c due to it being just a little larger than the 2sb817 units. I put the caps on the solder side of the board if this happens due to size. have only had 1 Q55 that has failed in the past 2 years so far and that was on a stock radio.. have not thought about the TIP42c for Q55. thanks that is worth a try. I will try one in that location the next time one shows up. have a bunch of them on hand to try.

 

[Robb]

I agree with those changes (TIP 36C & 42C) - that figures.
So we agree; that will work.

While we are on this subject, thought I would raise another problem area that is found in many modulation circuits similar to the 959. Q56 is a 2SC945 and it usually runs pretty hot as a rule. Seen the electrolytic cap next to it fail as well as that transistor when a local op accidentally set the 'lock on' his mic for a couple of hours.

The problem is the heat generated by Q56, as it was enough to curl the plastic cover off of that cap (C242) with failure, and the failure of that transistor itself. Replaced both parts to put the radio back into service. Cap is rated at 85C; don't think that a 105C rated cap will help much. Maybe? But the heat generated by that transistor is the cause and failure point. That to92 case doesn't exactly lend itself to using a heatsink so close to the power input of the radio. Using a NPN transistor that has the same value, has a to220 case, and mounted to the frame will . . .

 

[mb91w126]

If someone has a 949 or 959 they can perform these replacements on, do so and give me some feedback, please.

(Edit) a TIP42C will also be an upgrade to Q55 too, and is similar to the 2N6109G.

~Cheers~

send me your address, I'm certain I can find a 949 packed away that you can play with.

 

[ExitThirteen]

send me your address, I'm certain I can find a 949 packed away that you can play with.

I sent you a PM. Thanks!

~Cheers~

 

[mb91w126]

I sent you a PM. Thanks!

~Cheers~

Box is in the mail, Enjoy!

 

[Ranch55]

Usually there will be a bare wire running from the 10.240mhz xtal and the xtal filter; you can to attach the neg lead there. If there isn't one; then I take a small hand grinder and nick the corners of those two parts just enough to remove the nickel plating. The brass underneath that plating can be soldered, so I solder a piece of resistor lead between those two points in there.

If you look at the back of the board near the driver and final, you will see what looks like a resistor with a white circle silkscreened on the board around it. There is one near the driver and one near the final. That is where you want to put the pos lead of your DMM to check the gate voltage for the driver and final. Or, just put the tip of the pos test lead on the MOSFET gate directly. Better to attach it to the top of each resistor - though, as this will allow you to have a free hand to make the adjustment w/o worrying about slipping off and shorting something out.

First off, MOSFETs cannot be tested the same way that one would test - let's say - a 2sc1969 bipolar transistor could be tested. Because MOSFETs need a turn of voltage threshold to even operate at. Apples and oranges.

Next, you used the service manual to set up the radio. For the most part, it will work. But when it comes to the final transistors and they are MOSFETs; then they are not adjusted the same way. Probably why you snapped them. The gate voltage must be adjusted for each; 3.65v for the driver and 3.9v for the final. This is done with the mic gain set at 0, any mode, and must be tested when keying the mic at the gate pin of each device.
**EDIT** with the biasing board left in the radio!!!!

I agree with Sonoma about replacing the AM Regulator or its companion part. If one hasn't failed then the other might have.

Hey Robb .......
In one of your earlier posts a year or so before this one, you had wrote to set the driver bias voltage at 3.6 to 3.7 volts and the final voltage to 3.5v.
In this post I see you recommend 3.65v for the driver and 3.9v for the final.
I am now into my DX959B and wondering which it should be ????
I value your assistance on the forum with the 959 MOSFET radio.

 

[Robb]

Hey Robb .......
In one of your earlier posts a year or so before this one, you had wrote to set the driver bias voltage at 3.6 to 3.7 volts and the final voltage to 3.5v.
In this post I see you recommend 3.65v for the driver and 3.9v for the final.
I am now into my DX959B and wondering which it should be ????
I value your assistance on the forum with the 959 MOSFET radio.

The 'earlier' numbers were the numbers that the Galaxy techs recommended via an email I sent requesting that info - because that info was otherwise not available on CBTricks.

Running a mosfet at/above 4v is the point that isn't recommended; but you can run them less than that. Output varies a little; but longevity of the part changes with the amount of voltage you feed them.

Use the Galaxy recommended voltages - IMO . . .

 

[Ranch55]

The 'earlier' numbers were the numbers that the Galaxy techs recommended via an email I sent requesting that info - because that info was otherwise not available on CBTricks.

Running a mosfet at/above 4v is the point that isn't recommended; but you can run them less than that. Output varies a little; but longevity of the part changes with the amount of voltage you feed them.

Okay, but I am asking what I need to set the bias voltages at because I am in my 959 at this moment. I had to replace both driver and final MOSFETs IRF520. I want to get the driver and final bias set. So which should have the higher voltage?

 

[Robb]

Split the baby and set them both to 3.75v; should be OK.