• 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.

Beta Matching Mosfets????

Onelasttime

Sr. Member
Aug 3, 2011
1,185
755
123
Ok so I decided to give in to the dark side and play with some mosfets. So if I am going to use mosfets in a push-pull configuration do I have to match them like you do with high power rf bjt's? I was think you might just match the gates but did not know. If you do need to match the hfe does one use something like an Atlas DCA?


I am not aware of any company offering matched mosfets at all other than Palomar. So it has me wondering? I can tell you I have never matched mosfets in an audio circuit or in a power supply but I have a lot to learn with electronics!

Thanks!
 

A MOSFET doesn't have a "beta" rating. Beta, or "hfe" is ratio of currents, used to describe the DC-current gain of a bipolar transistor. The gate of a MOSFET is an insulator, so there is no steady DC input current to measure.

We match the turn-on threshold gate voltage, or "Vth". Chineseum testers on Ebay that can measure this are pretty cheap.

This one costs more since it has a rechargeable battery and a case https://www.ebay.com/itm/LCR-TC1-Tr...054307?hash=item2394d83b63:g:e4oAAOSwPAxaK4yG

This one is half the price, but it has no case and runs from external power https://pulsar.ebay.com/plsr/clk/0/SADS/9?pld={"mecs":"162942418061dca6023798344e66a305ce1e37466d42","enc":"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"}

Tried both. I prefer the one with the case.

And no, I have never discovered how a parameter a textbook calls "DC beta" gets abbreviated to "hfe".

73
 
  • Like
Reactions: LeapFrog
What you want to match is the RDSon and Threshold voltage. They have little Arduino computer that reads all of those functions. I don't think it has a Gain function for FET's but I know it does for Bipolar transistors. Since beta is a ratio of currents.
Be sure to use the "Enhanced mode" mode FET's because they allow a small amount of DC power to flow continuously to magnetize the inductors/transformers. The ERF7530 is an enhanced mode.
FET's are voltage controlled and bipolar transistors are current controlled.
This is the one I bought: https://www.ebay.com/itm/2017-New-T...513005?hash=item281e357a2d:g:xJcAAOSwzXxaBXG0
I have repaired numerous Palomar and Black Cat amps with other fets and it worked fine.
I wanted to try these but still have stock of others on hand. https://www.allelectronics.com/item/irfp064v/power-mosfet-60v-130a-irfp064v/1.html
 
Last edited:
And no, I have never discovered how a parameter a textbook calls "DC beta" gets abbreviated to "hfe".

73

Probably equally confusing as to how so many words ends up as three letters but here it is. hFE is an abbreviation, and it stands for "Hybrid parameter forward current gain, common emitter", and is a measure of the DC gain of a junction transistor.
 
Ok so I decided to give in to the dark side and play with some mosfets. So if I am going to use mosfets in a push-pull configuration do I have to match them like you do with high power rf bjt's? I was think you might just match the gates but did not know. If you do need to match the hfe does one use something like an Atlas DCA?


I am not aware of any company offering matched mosfets at all other than Palomar. So it has me wondering? I can tell you I have never matched mosfets in an audio circuit or in a power supply but I have a lot to learn with electronics!

Thanks!

Do you know the man at telestar electronics? I believe he uses these mosfets. You might message him on YouTube or Facebook.
 
Nomadradio that is what I thought but I was not sure. I had never heard of anyone matching them before Palomar started selling their re-branded Chinessium transistors and it sounded crazy to me because it did not match what I knew from car audio and modern home audio or power supplies and motor controls. I am aware of matching gate turn on voltage that was what I meant by "matching gates". I have not built anything for RF using them and have not done the conversions to mosfets as finals in radio's either.

With all of the above said I am old enough now that I have no fear of asking because right when you think you know something those pesky exception always pop up and bite you in the rear! LOL

What became of IGBT's for RF amplifiers did they lose out to Mosfets and LDMOS one taking the low road and the other taking the high road? Years ago I thought they where going to really take off????
 
There are "Modules" that can with stand extreme SWR and severe impedance mismatches, input or output will not destroy them. Fairly expensive too.
Very simple designs yield big results.
 
Last edited:
These are the Palomar "matched pair" ERF2030+ I bought off eBay, sadly only two pair out of five actually matched according to the provided stickers.. o_O


Here is what the text reads off of one "pair":

Vgs(on)=3.854 V @ iD=5.01mA & ig=0uA
Vgs(off)=3.053 V @ iD=5.2 uA
gm= 34.8mA per Volt @ iD = 3-5 mA
 

Attachments

  • 20180331_215112 (1).jpg
    20180331_215112 (1).jpg
    2.9 MB · Views: 191
  • 08_163_4_JPL_Scheick.pdf
    306.8 KB · Views: 550
Last edited:
They just don't make IGBTs fast enough for 27 MHz RF. They're okay in Tesla coils at a tenth or so of that frequency.

The IRFP064V is sexy enough. Doesn't need a mica insulator. But a gate capacitance of 6700pf? Gotta learn to translate the way these switchmode parts are described to predict RF performance. Datasheets say nothing about bandwidth or signal gain. Does tell you all you need about making it switch on and off efficiently.

Figured somebody would have by now. No such luck.

Hmmm. The '064 shows a rise time of 200 ns and fall time 150 ns. Sounds to me like an upper limit around 5 or 6 MHz, more or less. The IRF520 shows rise time and fall time of 23 ns. Between six and 9 times as fast. Apparently fast enough for 27 MHz.

Still would like to know which of the switchmode MOSFETs is being rebranded "ERF9530". Apparently the 7530 is dead forever, but the 9530 is the replacement for it.

Don't have a setup to accurately measure rise time. Reverse-engineering the 9530's secret identity with a capacitance meter is probably too crude by itself.

Sure wish somebody would.

73
 
  • Like
Reactions: LeapFrog
Yes, I tried to fight the good fight and just say no to motor control and power supply mosfets. The price for BJT's has just gotten too out of sight to not play around with the cheap mosfets at this point. Not really switching my opinions just yet but willing to play around with them at this point. I would never stake my family name on them or a professional reputation but for my own use, it is fine to play around with. Sadly even with Palomars markup they are still dirt cheap when looking at price vs performance. The only downside is durability any mismatch at all at any phase angle spells certain death for all of these non-rf-specific fets!

If I understand these things right the only real way to determine gain is in the circuit based on pout.

Yes as expensive as bjt's are getting and it wont be long before the HG and such are up there in price too! They are quickly pricing themselves out of a market if you ask me. It is just going to make the LDMOS transistors look that much more attractive!
 
It would likely be faster to ask a semiconductor company if they had a transistor in house that could be used in place and give them the 7530's older specifications and that it was going to be used off-label for RF production and see what they say. Hit the major players up and sit back and see what they say. I have tracked down harder things to find like trying to find the machine shop in Taiwan that made a particular part. I have down it from photographs and calling around on more than one occasion to find a supplier that could match the quality and production demands I needed at a price point that allowed me to remain competitive. It is never easy but in this case, all you need to do is email or call semiconductor companies with a specification and see what they say. For the record, I have done that in the automotive trade, musical instrument construction and aviation parts. So I know a thing or two about doing it. I just do not have a dog in the fight when it comes to electronics since I only work on my own gear.

I know this much Palomar is not having anyone spin up a custom part for them you know it is off the shelf. So at the very least, we have that working in our favor. That said it is still like looking for a needle in a haystack. The only other thing we know for sure is that it is some Chinese made part by a company without a solid reputation otherwise they would use that companies name.

I just had to send back some clutch parts from one two different USA vendors. One sold me an Exedy Clutch kit but it did not contain Exedy Parts they were easy to spot Chinese fakes with no country of origin, no part numbers, no name brands on any of the parts. The company claimed that all of their parts were made in Japan or America. If a part is made in Germany, Japan, USA you always want that on the part because you command a premium for the parts and they sell themselves. The other company did similar they showed a Seco made clutch disc Made in Korea with part numbers etc laser engraved on the part and rebranded Valeo on the part as well. What I got though was a completely different part than listed. Different design, color, friction material, sanitized part again with no country of origin, no part numbers no brand name etc.....The only reason to have a sanitized part is if you would not want to be associated with the OEM because of the country of origin, quality control or price point.

Sadly people always want to maximize their profits and often that means secrets and lies! I never feel comfortable staking my reputation on any part that is a generic jelly bean parts. I always want to know who is making the parts I am using. Like I said in the above people will often pay a premium for parts amde in Japan, Germany or the USA or made by a company you trust to have quality control even if it is made in China so a Sony, Fujitsu, Toshiba, Motorola made part that is made in China, Israel, Korea, Mexico etc...Why? Because we know that even if it is being made in China or India that those companies are still demanding good quality control standards.
 
What you want to match is the RDSon and Threshold voltage. They have little Arduino computer that reads all of those functions. I don't think it has a Gain function for FET's but I know it does for Bipolar transistors. Since beta is a ratio of currents.
Be sure to use the "Enhanced mode" mode FET's because they allow a small amount of DC power to flow continuously to magnetize the inductors/transformers. The ERF7530 is an enhanced mode.
FET's are voltage controlled and bipolar transistors are current controlled.
This is the one I bought: https://www.ebay.com/itm/2017-New-T...513005?hash=item281e357a2d:g:xJcAAOSwzXxaBXG0
I have repaired numerous Palomar and Black Cat amps with other fets and it worked fine.
I wanted to try these but still have stock of others on hand. https://www.allelectronics.com/item/irfp064v/power-mosfet-60v-130a-irfp064v/1.html
I think if you buy a 10 pack they are $3 and some change but if you buy 2 or 4 matched they are something like $9 each which is just silly. I understand some premium for labor involved in testing them but not 2X or 3X more.
 
I think if you buy a 10 pack they are $3 and some change but if you buy 2 or 4 matched they are something like $9 each which is just silly. I understand some premium for labor involved in testing them but not 2X or 3X more.
That's why I bought the little tester. for a little over $20.00 I can test 20 to 30 fets and recover my cost which I have a long time ago. Also nice to find out what the gate capacitance is between the different models of fets.
 
  • Like
Reactions: LeapFrog
Do you know the man at telestar electronics? I believe he uses these mosfets. You might message him on YouTube or Facebook.
I do not know him in real life. I do own 1 of his kits, one finished amp and about 16 pcb's from him and one of his loud mouth audio preamp's. .

The finished amp I bought from him was his single MRF454 with tuned input no transformer coupling. It is the most power single bjt 12V amp not using anything exotic that I have ever seen! The kit I purchased was his 2 transistor amp and you got everything in a box except the heat sink and enclosure and that is the very best 2 transistor amp I have ever used in my almost 45 years of life. Build went flawless. The PCB's I own are a mix of 2 transistor, 1 transistor and 4 transistor boards. In the past when revises a board design and spins up another board design he sells the older revision boards on ebay for people that want to experiment and build their own gear so I always snatch them up when ever they come up for sale on ebay.

He made a compression module called Voice Max that I wished I had purchased but by the time I got back into radio he had discontinued them. So I have 2 of his loudmoth modules. They basically use a low noise op amp and provide phantom power, amplification and filtering so you can add an electret to any radio and get seriously good audio. It has a low pass filter function as well. A few cap changes and that combined with an electret mic and it is golden. No need for blocking cap, resistors or any of that in the audio line of the mic since it is built into the module. It is what people think they are getting with a power mic that they never actually get with a power mic! Not talking about direct inject or MMM since this is still going through the stock audio chain. For someone in a car though or that does not want to have to fool with direct inject or MMM but wants much better audio it was a good an option. He stopped making those. I think it passes 8V up the pin not 48V like true phantom power in an audio setting would have. The only down side is that the phantom power is not user adjustable and all the parts on the board are smd.
 

dxChat
Help Users
  • No one is chatting at the moment.
  • @ Wildcat27:
    Hello I have a old school 2950 receives great on all modes and transmits great on AM but no transmit on SSB. Does anyone have any idea?
  • @ ButtFuzz:
    Good evening from Sunny Salem! What’s shaking?
  • dxBot:
    63Sprint has left the room.