God, I just heard his voice in my head and his laugh... I'm not trying to push that man over the edge, I like Mike. I like Mike and Buddy... I'd damn sure take their word on anything over some of the other people...
-
You can now help support WorldwideDX when you shop on Amazon at no additional cost to you! Simply follow this Shop on Amazon link first and a portion of any purchase is sent to WorldwideDX to help with site costs.
Blown Up Anytone AT-6666
- Thread starter TheRATT
- Start date
Matching any transistors will never be the same unless you have a witch that will tap her broom on them lol , just move on with the subject.
Well, matching MOSFETS for RF use is a fairly complicated beast.
This is why they stick to much simpler means of matching like I said before about using a curve tracer. That will probably get you close enough.
When you start getting into RF then load-pull is one test to perform.
None of these yo-yos are doing any of this and the reality is that the DC parameters are good enough. You can use one of those peak component testers and probably be close enough and I bet that's all they do.
So, while I was writing this post I went and looked at the description of the ERF2030+ for sale on Amazon, and this is what it says:
Uh, yep. Vgs within 1/100th of a volt.... if you say so.
And, there are two variants of the 13N10... the FQP13N10 and the FQP13N10L
And they are compatible with each other *IF* you make the appropriate circuit alterations and ensure the bias can be set properly because these are not "drop in" compatible. If you swap in an L you will probably have magic smoke.
Similar for the IRF520 vs IRF520N but other parameters are different that may cause a release of magic smoke. I doubt the non-N variant is in use in CB radios.
But, yeah. I am sure the Chinese designed a MOSFET for this guy using some "new wafer design" so he can sell them CHEAPER than semiconductors from legitimate manufacturers, some with their own foundry too.
Yeah, what could POSSIBLY be wrong with this picture.
This is why they stick to much simpler means of matching like I said before about using a curve tracer. That will probably get you close enough.
When you start getting into RF then load-pull is one test to perform.
None of these yo-yos are doing any of this and the reality is that the DC parameters are good enough. You can use one of those peak component testers and probably be close enough and I bet that's all they do.
So, while I was writing this post I went and looked at the description of the ERF2030+ for sale on Amazon, and this is what it says:
Uh, yep. Vgs within 1/100th of a volt.... if you say so.
And, there are two variants of the 13N10... the FQP13N10 and the FQP13N10L
And they are compatible with each other *IF* you make the appropriate circuit alterations and ensure the bias can be set properly because these are not "drop in" compatible. If you swap in an L you will probably have magic smoke.
Similar for the IRF520 vs IRF520N but other parameters are different that may cause a release of magic smoke. I doubt the non-N variant is in use in CB radios.
But, yeah. I am sure the Chinese designed a MOSFET for this guy using some "new wafer design" so he can sell them CHEAPER than semiconductors from legitimate manufacturers, some with their own foundry too.
Yeah, what could POSSIBLY be wrong with this picture.
Well I’ve got 25 of the 2030+ pieces, ordered from my wholesaler, maybe they are fakes? Maybe it’s simply because of the bias? Possibly the same companion part for the original 2030’s isn’t correct for the new design? I do not know. I got agitated with them and gave up on them. If it’s simply a bias issue, I’d sure love to know what’s the correct part to use/build.
I’d like to get something back out the investment I put into so many of those things but I’m not above tossing them all in the trash; write it off as a bad experience.
I don’t see where they’re that much cheaper anyway, not by enough to worry myself with them. I was mostly curious to see the power level they’d put out. But I don’t agree with the statement made that 13N10’s are junk.
I’d like to get something back out the investment I put into so many of those things but I’m not above tossing them all in the trash; write it off as a bad experience.
I don’t see where they’re that much cheaper anyway, not by enough to worry myself with them. I was mostly curious to see the power level they’d put out. But I don’t agree with the statement made that 13N10’s are junk.
What are your expectations with these parts? 30w is all any of them are good for, including the 13n10's.
If the companion part has been changed; then Palomar would have issued and made new ones to sling right along with them. Any mosfet upgrade will require re-peaking the tx coils; did you do that?
Yes I did. The best I got was maybe 20w peak.
Example : Took a Cobra 29 and did a mosfet conversion, used the IRF520 with the 369 companion part. Ended up with a 4w key, 10w avg swing, 40w pep.
Ok, yea I’m aware of distortion and all that other stuff and how it’s possibly not a “true” 40 watt peak. This is just my example from experience in doing nothing except the conversion.
Swapped nothing but the 2030+ and nothing I could do would get me past the 20w peak. And maybe it’s cleaner and meaner, maybe that’s all it’ll do and the IRF piece is just trashy and nasty and full of “fake news”. Idk. But (while it could be lies) I’ve read where others got more, so..
It’s rolling with the 520’s so why try to reinvent the wheel? I gave up.
But I’m headed out to the shop right now and I have many, many radios that are terrified of me taking an iron to them. One will fall victim tonight and I’ll try the 2030+ again. I shall report back with everything I do.
One last time. After this, if my results are satisfactory to me, you can have the remaining 24.
Example : Took a Cobra 29 and did a mosfet conversion, used the IRF520 with the 369 companion part. Ended up with a 4w key, 10w avg swing, 40w pep.
Ok, yea I’m aware of distortion and all that other stuff and how it’s possibly not a “true” 40 watt peak. This is just my example from experience in doing nothing except the conversion.
Swapped nothing but the 2030+ and nothing I could do would get me past the 20w peak. And maybe it’s cleaner and meaner, maybe that’s all it’ll do and the IRF piece is just trashy and nasty and full of “fake news”. Idk. But (while it could be lies) I’ve read where others got more, so..
It’s rolling with the 520’s so why try to reinvent the wheel? I gave up.
But I’m headed out to the shop right now and I have many, many radios that are terrified of me taking an iron to them. One will fall victim tonight and I’ll try the 2030+ again. I shall report back with everything I do.
One last time. After this, if my results are satisfactory to me, you can have the remaining 24.
Guys, pay attention to the values you use to drive these units.
A typical 2078 you're replacing - has and requires some inductive input to offset it's own internal reactance.
MOSFETs' don't have the reach, or touch for that matter, to their power-side junction substrate to really affect any admittance issues like you'd have with a bipolars' linear operation. All you can do is provide some power and hope like h*ll the Gate can provide the rest of the "Conversion method" that offsets their high-impedance input.
The Gate drive voltage is what gets a lot into trouble - for some MOSFET types, they are not a consistent value that you can use a set amount and works every time. It's a gamble that sometimes pays off, other times - it's an epic fail.
Even with platforms for the same maker. What works for one, cannot always be considered workable for another.
If you can't use those ERF2030+'s - PM me - we'll talk....
A typical 2078 you're replacing - has and requires some inductive input to offset it's own internal reactance.
MOSFETs' don't have the reach, or touch for that matter, to their power-side junction substrate to really affect any admittance issues like you'd have with a bipolars' linear operation. All you can do is provide some power and hope like h*ll the Gate can provide the rest of the "Conversion method" that offsets their high-impedance input.
The Gate drive voltage is what gets a lot into trouble - for some MOSFET types, they are not a consistent value that you can use a set amount and works every time. It's a gamble that sometimes pays off, other times - it's an epic fail.
Even with platforms for the same maker. What works for one, cannot always be considered workable for another.
If you can't use those ERF2030+'s - PM me - we'll talk....
I have the SS9900 it looks just like that.
has anyone found a service manual for it yet?
Last edited:
I read something interesting somewhere last week. We cant find service manuals to lots of smt technology items not just cb radios but most smt built electronics.
Chicom gets schmatics n they found a loophole and build exacting electronics dirt cheep and sell right beside name brand electronics. Uniden n many others are starting to safeguard themselves. Seems like i read it in an email about a new shortwave am fm radio and that was in the report.i will look n see if i still got thatvemail n post it,might deleted it
Chicom gets schmatics n they found a loophole and build exacting electronics dirt cheep and sell right beside name brand electronics. Uniden n many others are starting to safeguard themselves. Seems like i read it in an email about a new shortwave am fm radio and that was in the report.i will look n see if i still got thatvemail n post it,might deleted it
FET's do not have a gain figure. Gain is the ratio of Base Current to Collector current.
It is a current operated device.
FET's on the other hand are voltage operated devices and as such will not have a gain value since the is no current going in to the gate. The gate is capacitive and blocks DC current.
FET's are matched with Vgs and RDS on and gate capacitance. I don't think a standard curve tracer will work for matching those features. I have a small computerized transistor test that does measure those parameters and is also good for matching Bipolar transistors for gain.
Biploar transistors gain
https://www.brighthubengineering.co...17851-forward-current-gain-in-bjts-explained/
Field effect transistor
https://www.electronics-tutorials.ws/transistor/tran_5.html
It is a current operated device.
FET's on the other hand are voltage operated devices and as such will not have a gain value since the is no current going in to the gate. The gate is capacitive and blocks DC current.
FET's are matched with Vgs and RDS on and gate capacitance. I don't think a standard curve tracer will work for matching those features. I have a small computerized transistor test that does measure those parameters and is also good for matching Bipolar transistors for gain.
Biploar transistors gain
https://www.brighthubengineering.co...17851-forward-current-gain-in-bjts-explained/
Field effect transistor
https://www.electronics-tutorials.ws/transistor/tran_5.html
Last edited:
Not the gate capacitance, no.
But it will for Vgs.
The people matching these and selling them are only interested in Vgs.
They don't care about gate capacitance. It's a gimmick, remember?
When we were looking at ERF2030s they looked suspiciously identical to the FQP13N10s we were trying at the same time.
We were debating trying an experiment by which we put an FQP10N10 and an ERF2030 together in push-pull (like the Anytone) and see what happens.
The speculation is, that it will just work.
But it will for Vgs.
The people matching these and selling them are only interested in Vgs.
They don't care about gate capacitance. It's a gimmick, remember?
When we were looking at ERF2030s they looked suspiciously identical to the FQP13N10s we were trying at the same time.
We were debating trying an experiment by which we put an FQP10N10 and an ERF2030 together in push-pull (like the Anytone) and see what happens.
The speculation is, that it will just work.