Starting to like the MOSFETS - am I wrong?

[CTStallion]

I've done dozens of Cobra 29's over the years, and was always partial to 2SC1969 finals. However, I recently got the opportunity to work on the following FACTORY MOSFET'ed Galaxy transceivers:

- Galaxy DX-98VHP which has three (3) IRF-520's (1 as a driver and twin finals) driving an amplifier section consisting of eight (8) IRF-520's for a total of 11 of these amazing lil puppies.... and a

- Galaxy DX-2547 CB Radio AM/SSB Base Station with a single IRF-520 final transistor (I think the driver may have been a MOSFET as well, but I can't recall)

Anyway... the DX-98VHP exhibited super performance on both AM and SSB. It also incorporated an internal cooling fan in the amplifier section which will no doubt help keep the 8X520's cool under load. If I had my 'druthers', I'd snag up one of these in a heartbeat. What's not to love?

The DX-2547 was also impressive. The factory .15 uV SSB receive sensitivity gives this thing some EARS! Can't remember the RX transistor number but it was NOT the typical 2SC1674.... it was quieter... was NOT a 2SC2999, but was an equivalent with even better gain. Also, and here's what impressed me about the MOSFET final, when I adjusted the ALC for SSB output power, the radio would do over 30 watts SSB (factory Galaxy stock mic) on the internal AC power supply off of a single IRF-520. Not knowing the current output rating of the AC supply, I turned the rig down to 22W SSB PEP.

30W SSB with the turn of a VR POT off of a single IRF520 seemed pretty impressive to me; and from what I've read about the IRF520, it looks like it can handle every bit of it. (Again, my only fear was the factory power supply. Anybody know the current handling capacity of the internal AC to DC pwr supply in the DX-2547?)

So, WHY am I starting to like MOSFETS? Well, over the years I've either repaired, modified, converted or enhanced dozens & dozens (perhaps hundreds) of radios, mostly Cobra 29's. I've always been partial to the 2SC1969 final, but in the past few 10-12 radios I've done, I've begun to incorporate the IRF-520's in an increasing number of Cobra 29's. Here's what I've found:

- Older Cobra 29 LTD's (especially Philippines), GTL's and XLR's: seem to prefer the 2SC1969 for higher output power. I've gotten over 45 watts out of these radios.

- Newer 29 LTD's (and all SoundTrackers, Nightwatches, WX band radios): seem to respond BEST to the IRF520 mod (with the companion part designed for the ERF-2030). With these (especially SoundTrackers) it seems they respond best to maintaining a deadkey in the range of 2-4 watts and allowing 'some' swing into the 8-12 RMS range. Typical yield is about 25-35 watts PEP on modulation.

- 1990's and new Made in CHINA Versions: these are truly the WILDCARDS. Some like the 2SC1969 mod, some seem to like the IRF520. Since the 520's are plentiful, less expensive, and easier to install, I think I'm gonna stick to this arrangement unless any of you can provide reasons for sticking with an NPN bipoloar final.

So, I guess I'm really starting to like the MOSFETs especially for the following reasons:

1. They are becoming more and more prevalent from the FACTORY (especially in the Galaxy chassis rigs, and in the Export radios)

2. Many radios respond well to them (even the Cybernet chassis as seen on YouTube)

3. They are (certainly) less expensive then 2SC1969 or equivalents

4. They are 'easier' to install (relative) in a Cobra 29 chassis

5. They have a higher power dissipation rating than a 1969

6. A single finaled MOSFET can develop higher output power than a single finaled 2SC1969 in the Galaxy chassis with just the turn of a pot(s), whereas you'd have to make some circuit modifications to get the 2SC1969's to 'come-to-life' in an older radio

7. 2SC1969's are no longer being manufactured. The closest we can get from the 'factory' is the 20W Eleflow version.

8. In a pinch, the IRF510 (15W version) is readily available from our local Radio Shack. (Try finding a genuine Mitsubishi 2SC1969 from Radio Shack!, lol

9. I've gotten "burned" in the past with counterfeit 2SC2166 and 2SC1969's either NOT working at all, or failing to provided desired gain or power dissipation requirements. So far, (knock wood) all the 510's and 520's I've tried have worked flawlessly.

10. MOSFETs are the 'future'... even though I constantly refer to myself as an "analog guy in a digital world", I either need to get with the program or be left behind.

11. Just about every late-model Ham transceiver coming out of Japan have been using MOSFETs for quite some time. They develop good power: CHEAPLY (I suppose.) I remember the "old days" when blowing the finals in a ham transceiver could cost "big bucks" for the transistors. Recently, I needed to change the HF finals only in my Icom 706MKIIG (MOSFETS) and I think the pair cost me less than $60 from RF Parts. (I probably could've done better price-wise if I had shopped around, or sourced them from overseas).

Seems to me like Galaxy, RCI - Ranger, RF Limited, and the 'Exports' have all already jumped on the MOSFET bandwagon. In fact, the ONLY NEW radios I'm seeing still coming from the factory with NPN 2SC2078 final transistors are Cobra, Midland, and Uniden. And, IF you want to significantly increase the output power of the 2078 equipped transceivers, the choice is clear: MOSFET POWER.

What's NOT to love? What do you guys think???

 

[Carl]

most power transitors have went out of production in the past few years. that's why so many company's are switching to mosfets.

 

[mackmobile43]

You're well ahead of the game by introducing yourself to the mosfet as the older version of the transistor is on it's way out the door kinda like the internal combustion engine.

 

[NorthStar]

Yep bipolars are toast for the most part.

 

[Onelasttime]

Wow!!! I heard of a runner's hi but never a mosfet hi! LOL Just sit the fet down and step away! LOL

Ok so we have had some time now to get used to non-rf rated fet's in rf production. What are your thoughts now?

Other than the insane cost difference are you still hi as a kite on mosfets??? The only thing I can see is better profit margins from the top to the bottom of the food chain.

The pep power is sometimes better than the bjt parts but not always. The rms power is seldom any better at all. Often we are talking a insignificant improvement. By insignificant I mean it will not change our ability to communicate over using the same setup but with bjt parts from the past.

So far the only real argument for them is as follows: 1) The older bjt transistors are no longer made and expensive! 2) They produce more power but not enough to be meaningful in the real world. Not exactly wowing me with those as Pro's!!!

1)As I see it. You can still get the older parts you just have to pay more for them.

2)The non-rf specific fet's have terrible vswr mismatch survivability!!!

3) They lack durability almost as a design feature and even with decent swr's are failure prone when used in rf production when not designed for it. No one is talking about using actual rf-specific fet's in CB radio's or export radio's. Why? Because they are not cheap!

4) 99% of fet's do not make much power or have much gain at the voltages we operate at. Modifying the radio to step up the voltage to the fet's to 28V, 48V, 100V would defeat the cost savings so no one is doing that.

5) Did I mention the lack of GAIN at 12V-16V for these fet's??? If you do not understand how important gain is hfe to optimum 2 way radio operation I suggest you read up on it. The Bird 43 watt meter is not the end all be all tool for rf communication validation!

When I see a NXP or Freescale Fet in a CB or Export radio I will change my mind! LOL Seriously though we should at the very least demand rf-specific fet's in our rf-radio's. At the very least we should have a gain of at least 10 minimum for AM/FM/SSB/CW radio's and even higher for AM only which traditionally had hfe's of 50-60 minimum for the driver and final. They need to be rated for rf in linear mode with gain at the voltages we operate the mostly mobile gear at. Watt's with out gain is like horsepower without torque looks good on a dearlership brochure might even work on race track that is set up for high sustained speeds but in the real world will be almost useless! We do not drive around on a dyno and we do not talk to a Bird 43 meter.


I am thinking about trying some RD16HHF1's I can get real ones fairly cheap much cheaper than the MP 2SC1969's and 2166's I just got. I should be able to get 10 of them for the same price as a single MP or the two parts above. It should produce about the same power as a 2SC1969 and even more gain at out voltages!

 

[543_Dallas]

Manufacturers do a good job of making a radio produce a clean signal. Why not leave that alone and amplify that with some sort of device that can do so in a linear fashion?

 

[Tallman]

I've done dozens of Cobra 29's over the years, and was always partial to 2SC1969 finals. However, I recently got the opportunity to work on the following FACTORY MOSFET'ed Galaxy transceivers:

- Galaxy DX-98VHP which has three (3) IRF-520's (1 as a driver and twin finals) driving an amplifier section consisting of eight (8) IRF-520's for a total of 11 of these amazing lil puppies.... and a

- Galaxy DX-2547 CB Radio AM/SSB Base Station with a single IRF-520 final transistor (I think the driver may have been a MOSFET as well, but I can't recall)

Anyway... the DX-98VHP exhibited super performance on both AM and SSB. It also incorporated an internal cooling fan in the amplifier section which will no doubt help keep the 8X520's cool under load. If I had my 'druthers', I'd snag up one of these in a heartbeat. What's not to love?

The DX-2547 was also impressive. The factory .15 uV SSB receive sensitivity gives this thing some EARS! Can't remember the RX transistor number but it was NOT the typical 2SC1674.... it was quieter... was NOT a 2SC2999, but was an equivalent with even better gain. Also, and here's what impressed me about the MOSFET final, when I adjusted the ALC for SSB output power, the radio would do over 30 watts SSB (factory Galaxy stock mic) on the internal AC power supply off of a single IRF-520. Not knowing the current output rating of the AC supply, I turned the rig down to 22W SSB PEP.

30W SSB with the turn of a VR POT off of a single IRF520 seemed pretty impressive to me; and from what I've read about the IRF520, it looks like it can handle every bit of it. (Again, my only fear was the factory power supply. Anybody know the current handling capacity of the internal AC to DC pwr supply in the DX-2547?)

So, WHY am I starting to like MOSFETS? Well, over the years I've either repaired, modified, converted or enhanced dozens & dozens (perhaps hundreds) of radios, mostly Cobra 29's. I've always been partial to the 2SC1969 final, but in the past few 10-12 radios I've done, I've begun to incorporate the IRF-520's in an increasing number of Cobra 29's. Here's what I've found:

- Older Cobra 29 LTD's (especially Philippines), GTL's and XLR's: seem to prefer the 2SC1969 for higher output power. I've gotten over 45 watts out of these radios.

- Newer 29 LTD's (and all SoundTrackers, Nightwatches, WX band radios): seem to respond BEST to the IRF520 mod (with the companion part designed for the ERF-2030). With these (especially SoundTrackers) it seems they respond best to maintaining a deadkey in the range of 2-4 watts and allowing 'some' swing into the 8-12 RMS range. Typical yield is about 25-35 watts PEP on modulation.

- 1990's and new Made in CHINA Versions: these are truly the WILDCARDS. Some like the 2SC1969 mod, some seem to like the IRF520. Since the 520's are plentiful, less expensive, and easier to install, I think I'm gonna stick to this arrangement unless any of you can provide reasons for sticking with an NPN bipoloar final.

So, I guess I'm really starting to like the MOSFETs especially for the following reasons:

1. They are becoming more and more prevalent from the FACTORY (especially in the Galaxy chassis rigs, and in the Export radios)

2. Many radios respond well to them (even the Cybernet chassis as seen on YouTube)

3. They are (certainly) less expensive then 2SC1969 or equivalents

4. They are 'easier' to install (relative) in a Cobra 29 chassis

5. They have a higher power dissipation rating than a 1969

6. A single finaled MOSFET can develop higher output power than a single finaled 2SC1969 in the Galaxy chassis with just the turn of a pot(s), whereas you'd have to make some circuit modifications to get the 2SC1969's to 'come-to-life' in an older radio

7. 2SC1969's are no longer being manufactured. The closest we can get from the 'factory' is the 20W Eleflow version.

8. In a pinch, the IRF510 (15W version) is readily available from our local Radio Shack. (Try finding a genuine Mitsubishi 2SC1969 from Radio Shack!, lol

9. I've gotten "burned" in the past with counterfeit 2SC2166 and 2SC1969's either NOT working at all, or failing to provided desired gain or power dissipation requirements. So far, (knock wood) all the 510's and 520's I've tried have worked flawlessly.

10. MOSFETs are the 'future'... even though I constantly refer to myself as an "analog guy in a digital world", I either need to get with the program or be left behind.

11. Just about every late-model Ham transceiver coming out of Japan have been using MOSFETs for quite some time. They develop good power: CHEAPLY (I suppose.) I remember the "old days" when blowing the finals in a ham transceiver could cost "big bucks" for the transistors. Recently, I needed to change the HF finals only in my Icom 706MKIIG (MOSFETS) and I think the pair cost me less than $60 from RF Parts. (I probably could've done better price-wise if I had shopped around, or sourced them from overseas).

Seems to me like Galaxy, RCI - Ranger, RF Limited, and the 'Exports' have all already jumped on the MOSFET bandwagon. In fact, the ONLY NEW radios I'm seeing still coming from the factory with NPN 2SC2078 final transistors are Cobra, Midland, and Uniden. And, IF you want to significantly increase the output power of the 2078 equipped transceivers, the choice is clear: MOSFET POWER.

What's NOT to love? What do you guys think???

I've done dozens of Cobra 29's over the years, and was always partial to 2SC1969 finals. However, I recently got the opportunity to work on the following FACTORY MOSFET'ed Galaxy transceivers:

- Galaxy DX-98VHP which has three (3) IRF-520's (1 as a driver and twin finals) driving an amplifier section consisting of eight (8) IRF-520's for a total of 11 of these amazing lil puppies.... and a

- Galaxy DX-2547 CB Radio AM/SSB Base Station with a single IRF-520 final transistor (I think the driver may have been a MOSFET as well, but I can't recall)

Anyway... the DX-98VHP exhibited super performance on both AM and SSB. It also incorporated an internal cooling fan in the amplifier section which will no doubt help keep the 8X520's cool under load. If I had my 'druthers', I'd snag up one of these in a heartbeat. What's not to love?

The DX-2547 was also impressive. The factory .15 uV SSB receive sensitivity gives this thing some EARS! Can't remember the RX transistor number but it was NOT the typical 2SC1674.... it was quieter... was NOT a 2SC2999, but was an equivalent with even better gain. Also, and here's what impressed me about the MOSFET final, when I adjusted the ALC for SSB output power, the radio would do over 30 watts SSB (factory Galaxy stock mic) on the internal AC power supply off of a single IRF-520. Not knowing the current output rating of the AC supply, I turned the rig down to 22W SSB PEP.

30W SSB with the turn of a VR POT off of a single IRF520 seemed pretty impressive to me; and from what I've read about the IRF520, it looks like it can handle every bit of it. (Again, my only fear was the factory power supply. Anybody know the current handling capacity of the internal AC to DC pwr supply in the DX-2547?)

So, WHY am I starting to like MOSFETS? Well, over the years I've either repaired, modified, converted or enhanced dozens & dozens (perhaps hundreds) of radios, mostly Cobra 29's. I've always been partial to the 2SC1969 final, but in the past few 10-12 radios I've done, I've begun to incorporate the IRF-520's in an increasing number of Cobra 29's. Here's what I've found:

- Older Cobra 29 LTD's (especially Philippines), GTL's and XLR's: seem to prefer the 2SC1969 for higher output power. I've gotten over 45 watts out of these radios.

- Newer 29 LTD's (and all SoundTrackers, Nightwatches, WX band radios): seem to respond BEST to the IRF520 mod (with the companion part designed for the ERF-2030). With these (especially SoundTrackers) it seems they respond best to maintaining a deadkey in the range of 2-4 watts and allowing 'some' swing into the 8-12 RMS range. Typical yield is about 25-35 watts PEP on modulation.

- 1990's and new Made in CHINA Versions: these are truly the WILDCARDS. Some like the 2SC1969 mod, some seem to like the IRF520. Since the 520's are plentiful, less expensive, and easier to install, I think I'm gonna stick to this arrangement unless any of you can provide reasons for sticking with an NPN bipoloar final.

So, I guess I'm really starting to like the MOSFETs especially for the following reasons:

1. They are becoming more and more prevalent from the FACTORY (especially in the Galaxy chassis rigs, and in the Export radios)

2. Many radios respond well to them (even the Cybernet chassis as seen on YouTube)

3. They are (certainly) less expensive then 2SC1969 or equivalents

4. They are 'easier' to install (relative) in a Cobra 29 chassis

5. They have a higher power dissipation rating than a 1969

6. A single finaled MOSFET can develop higher output power than a single finaled 2SC1969 in the Galaxy chassis with just the turn of a pot(s), whereas you'd have to make some circuit modifications to get the 2SC1969's to 'come-to-life' in an older radio

7. 2SC1969's are no longer being manufactured. The closest we can get from the 'factory' is the 20W Eleflow version.

8. In a pinch, the IRF510 (15W version) is readily available from our local Radio Shack. (Try finding a genuine Mitsubishi 2SC1969 from Radio Shack!, lol

9. I've gotten "burned" in the past with counterfeit 2SC2166 and 2SC1969's either NOT working at all, or failing to provided desired gain or power dissipation requirements. So far, (knock wood) all the 510's and 520's I've tried have worked flawlessly.

10. MOSFETs are the 'future'... even though I constantly refer to myself as an "analog guy in a digital world", I either need to get with the program or be left behind.

11. Just about every late-model Ham transceiver coming out of Japan have been using MOSFETs for quite some time. They develop good power: CHEAPLY (I suppose.) I remember the "old days" when blowing the finals in a ham transceiver could cost "big bucks" for the transistors. Recently, I needed to change the HF finals only in my Icom 706MKIIG (MOSFETS) and I think the pair cost me less than $60 from RF Parts. (I probably could've done better price-wise if I had shopped around, or sourced them from overseas).

Seems to me like Galaxy, RCI - Ranger, RF Limited, and the 'Exports' have all already jumped on the MOSFET bandwagon. In fact, the ONLY NEW radios I'm seeing still coming from the factory with NPN 2SC2078 final transistors are Cobra, Midland, and Uniden. And, IF you want to significantly increase the output power of the 2078 equipped transceivers, the choice is clear: MOSFET POWER.

What's NOT to love? What do you guys think???

One of the best things about FET's is that they have a positive Temperature Co-efficient. The hotter they get the less current they conduct.
Bi-polar transistors the hotter they get the more power they will conduct and end up in thermal runaway. That's the negative temperature Co-efficient. That's why BETA MATCHING is so important for amplifiers.

 

[space cowboy]

Seriously though we should at the very least demand rf-specific fet's in our rf-radio's.

beggars can't be choosers.

 

[543_Dallas]

These little fets are the sweep tubes of today.

 

[BBB]

MOSFETS are for motor control and blow up regularly is what I have experianced. No failures have I had with bipolars.

 

[sonoma]

I have not had a mosfet go out as of yet and have converted about 50 so far and no problems.I even use the 13n10 fets some say are no good

 

[Robb]

The only mosfet I had that failed was done by myself to see what would happen to it. It shattered with a mild snap. Other than that, every radio that I have that has a mosfet in it are all still chugging away without any issue . . .

 

[Captain Kilowatt]

MOSFETS are for motor control and blow up regularly is what I have experianced. No failures have I had with bipolars.

I guess it depends on what device you are using and how you are using it. MOSFETs have proven reliability in commercial gear since the early 80's when commercial broadcast transmitters were being made with them. Today they run 50Kw AM transmitters with them and virtually all commercial RF gear uses them. I have a 1.2 Kw amplifier for 6m that uses four MRF151G's which ar3 actually old technology. The amp was originally us3d in a television transmitter where reliability was foremost. No......MOSFET's ARE reliable but you need to use the proper devices and know how they should be run.

 

[rabbiporkchop]

They are just fine for a 50 ohm load. How many cb'ers can tell you their feedpoint impedance? Most don't have a clue, hence the problem.

 

[543_Dallas]

They are just fine for a 50 ohm load. How many cb'ers can tell you their feedpoint impedance? Most don't have a clue, hence the problem.

Just because something works is a bad reason to settle for an inferior part. Why wouldn't the radio manufacturer use a better device? Higher profit margins is why. The same reason cb shops replaced the 1969 upgrade with the mosfet upgrade.

You make a good point about the antenna system. The manufacturer knows what their customers are capable of. Why not use a better device that will tolerate a less than perfect antenna. Why not have the transmitter fold back the power to save the finals when things are really bad?

This would have been a lot to ask 10 or 20 years ago but I think it's a reasonable expectation today.