-
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.
How many watts from a 2879
- Thread starter 543_Dallas
- Start date
How many watts from a Toshiba 2sc2879(a) ? For short duration @ 15-18VDC with a few hundred amps available...a lot. Always amazes me, so much from something so small.
They have caused me to contract a serious laziness condition called Solidstateitus-Easyium as all I have to do is flip a switch ON instead of waiting for something to warm up and second guessing about tuning this and that.
They have caused me to contract a serious laziness condition called Solidstateitus-Easyium as all I have to do is flip a switch ON instead of waiting for something to warm up and second guessing about tuning this and that.
Here's something to think about. Did anyone else notice that a single ended 2879 amp can produce 250 watts at a 25 amp draw and those are the same numbers we can expect from a "two pill"? That's right, a one pill amp makes the same output and draws the same maximum current as any two pill you've seen. The main advantages of the two pill are both transistors share the work and some harmonics are cancelled in the typical circuit.
Even though many of us understand that a push-pull amplifier only has one device at a time turned on, many of us think that pair of transistors can still make twice as much output as one, when that is not the case. For decades I never understood why the competition builders insisted on combining multiple push-pull stages to make larger amps when half of the dissipation of transistors was essentially being wasted.
Why didn't combining multiple single ended stages become more popular to double the available dissipation especially considering how this application was pushing that specification to the max? Once you get all of the transistors to conduct simultaneously, you'll be able to get twice the output or have twice the headroom at the same output if you're looking for a cleaner signal. Be prepared to feed and cool it as a two pill amp could now draw 50 amps and dissipate twice the heat!
Then there are some interesting things you can do with the input splitter and output combiner. Why is the two port device so common? The typical 8 pill uses 6 of them and each has loss. That's why the 6 pill and 3 port combiners tend to form a more efficient amplifier, only using one on the input and one on the output. Medical grade RF generators have been combining 8 transistors using one splitter and one combiner for many years. We can too. Just get your phasing correct on the right core material.
Even though many of us understand that a push-pull amplifier only has one device at a time turned on, many of us think that pair of transistors can still make twice as much output as one, when that is not the case. For decades I never understood why the competition builders insisted on combining multiple push-pull stages to make larger amps when half of the dissipation of transistors was essentially being wasted.
Why didn't combining multiple single ended stages become more popular to double the available dissipation especially considering how this application was pushing that specification to the max? Once you get all of the transistors to conduct simultaneously, you'll be able to get twice the output or have twice the headroom at the same output if you're looking for a cleaner signal. Be prepared to feed and cool it as a two pill amp could now draw 50 amps and dissipate twice the heat!
Then there are some interesting things you can do with the input splitter and output combiner. Why is the two port device so common? The typical 8 pill uses 6 of them and each has loss. That's why the 6 pill and 3 port combiners tend to form a more efficient amplifier, only using one on the input and one on the output. Medical grade RF generators have been combining 8 transistors using one splitter and one combiner for many years. We can too. Just get your phasing correct on the right core material.
first, again...
https://forums.qrz.com/index.php?threads/2sc2879-producing-250-watts-hahaha.248732/
Ignore the idiots, misfits, morons, maladjusted personalities, egos, and brain dead impersonations of human ( I use the term loosely) intelligence.
Read what Tom Rauch (W8JI) has to say.
Second , push-pull.
http://www.circuitstoday.com/push-pull-amplifier
Next , back to basics
https://www.worldwidedx.com/threads/pill-factory.230557/#post-628379
or
So if each transistor is using the same drive, dissipation, and current draw to amplify just one half of the waveform than the output will be x 2 .
A fine point to be sure but ...
https://forums.qrz.com/index.php?threads/2sc2879-producing-250-watts-hahaha.248732/
Ignore the idiots, misfits, morons, maladjusted personalities, egos, and brain dead impersonations of human ( I use the term loosely) intelligence.
Read what Tom Rauch (W8JI) has to say.
Second , push-pull.
http://www.circuitstoday.com/push-pull-amplifier
Next , back to basics
https://www.worldwidedx.com/threads/pill-factory.230557/#post-628379
Consider me one of the "many" with one reservation. Simply the lossy , broad banded nature of the splitter/combiner adds losses. Maybe it's easier to think of the push and pull as "area within the curve" described as one each adjacent positive and negative peak from zero (DC) to peak.
or
So if each transistor is using the same drive, dissipation, and current draw to amplify just one half of the waveform than the output will be x 2 .
A fine point to be sure but ...
If you have reservations, then just look at it from an input power standpoint. One single ended 2879 amp can draw 25 amps safely and two 2879's in push-pull can also, only draw 25 amps maximum before you've exceeded their specs. Simply using a different circuit to extract the power of two transistors from one, sounds too good to be true but easily confirmed on the test bench with just a few measurements.
If you drive the push-pull pair towards 30 amps, one will open soon. That's the same current one can draw in single ended circuits. Two transistors can only double the output if they both conduct simultaneously. In push-pull, each transistor turns on alternately and amplifies half of the waveform. At any given time, only one is working. With a pair of single ended circuits, both transistors turn on simultaneously during one half cycle while the other half of the waveform "rings out" through the resonant output circuit.
If you drive the push-pull pair towards 30 amps, one will open soon. That's the same current one can draw in single ended circuits. Two transistors can only double the output if they both conduct simultaneously. In push-pull, each transistor turns on alternately and amplifies half of the waveform. At any given time, only one is working. With a pair of single ended circuits, both transistors turn on simultaneously during one half cycle while the other half of the waveform "rings out" through the resonant output circuit.
There seems to be some confusion here.
2 x sd1446 (12a devices ) rm italy amp can and do pull more than 20a on FM every day over here.
Mrf455 version can pull at least as much in regular use & survive,
Both devices can pull more than the stock 2 x 12a fuses can handle if you get long winded.
4 x 1446 kl500 can pull more than 40a on FM.
More than the stock 3 x 12a fuses or fuse holders can handle.
That's why locals ask me to remove the holders & fit inline fuses on new & used amps.
To stop fuses blowing & holders melting as often is the case in their cb amps RM italy used 4 x 12a fuses in their 4 x 1446 hla300, ms1051 versions still blow the fuses when driven hard on FM
On stock volts a 2 x 2879 dave style amp pulls more current than a 4 x 1446 kl500 on FM.
measured with fluke 87's & 500a 0.25% manganin shunt.
se amps can be optimally matched for load impedance at any voltage below actual breakdown voltage ( not the 18v that the conservative datasheet says ),
& don't have lossy transformers & combiners until you start combining stages,
pp amps can only be matched optimally for one voltage & drive level due to fixed interger turns ratios,
unless you include an output matching network like the dave style comp amps & understand how to tune them for comp drive & voltage levels.
2 x sd1446 (12a devices ) rm italy amp can and do pull more than 20a on FM every day over here.
Mrf455 version can pull at least as much in regular use & survive,
Both devices can pull more than the stock 2 x 12a fuses can handle if you get long winded.
4 x 1446 kl500 can pull more than 40a on FM.
More than the stock 3 x 12a fuses or fuse holders can handle.
That's why locals ask me to remove the holders & fit inline fuses on new & used amps.
To stop fuses blowing & holders melting as often is the case in their cb amps RM italy used 4 x 12a fuses in their 4 x 1446 hla300, ms1051 versions still blow the fuses when driven hard on FM
On stock volts a 2 x 2879 dave style amp pulls more current than a 4 x 1446 kl500 on FM.
measured with fluke 87's & 500a 0.25% manganin shunt.
se amps can be optimally matched for load impedance at any voltage below actual breakdown voltage ( not the 18v that the conservative datasheet says ),
& don't have lossy transformers & combiners until you start combining stages,
pp amps can only be matched optimally for one voltage & drive level due to fixed interger turns ratios,
unless you include an output matching network like the dave style comp amps & understand how to tune them for comp drive & voltage levels.
All it took to demonstrate this was finding we could get over 250 watts out of one single ended 2879 as Tom did, at a maximum of 30 amps on the collector (5 more than rated). It's not even remotely possible to pull 60 amps on any 2879 push-pull 2 pill. I've compared the maximum input power capability to remove the variable of efficiency but both circuits were adjusted for maximum output and minimum input VSWR. Find the breakdown current that opens any pair of RF transistors in push-pull and it will match that of the same device in a single ended circuit. I learned this building current limited protection circuits for these amps and finding that the same circuit used to protect all of the 2 pill push-pull stages turned out to be the same values that safely protected its one pill driver stage.
Call it whatever you want but it's still not possible to have both transistors conduct simultaneously in push-pull as they do with a pair of single ended circuits (not counting bias current). Therefore you can't expect a push-pull amp to develop more output power than one single ended circuit because you can't apply any more input power than a single device can handle.
In an effort to reduce skepticism, I offer this challenge. Can anyone honestly demonstrate a push-pull pair of 2879's drawing significantly more than 30 amps of collector current? The engineer over at Ameritron has already confirmed 30 amps is about the maximum one transistor can handle and it made more output than most expected. Two of those single transistor circuits combined in phase can draw 60 amps.
I've never seen a push-pull pair of 2879's draw much over 30 amps and would be very interested to see anyone reach close to double. The transistor with the higher gain will always flash open before you can hit 35 amps. If the output stage cannot support the combined collector current of both transistors (50 amp rating, 60 amps maximum test), it cannot produce the combined output of both devices.
All of the points made earlier about efficiency, voltage and matching are very important when it comes to achieving the maximum output power. But, how much of that really matters in this example once we find out that the push-pull stage has no ability to support the combined collector current of both output devices?
In this case, the first goal was getting the stage to handle the anticipated input power of the two transistors combined. Then you can work out the details with efficiency, voltage and impedance matching for maximum output.
I've never seen a push-pull pair of 2879's draw much over 30 amps and would be very interested to see anyone reach close to double. The transistor with the higher gain will always flash open before you can hit 35 amps. If the output stage cannot support the combined collector current of both transistors (50 amp rating, 60 amps maximum test), it cannot produce the combined output of both devices.
All of the points made earlier about efficiency, voltage and matching are very important when it comes to achieving the maximum output power. But, how much of that really matters in this example once we find out that the push-pull stage has no ability to support the combined collector current of both output devices?
In this case, the first goal was getting the stage to handle the anticipated input power of the two transistors combined. Then you can work out the details with efficiency, voltage and impedance matching for maximum output.
I thought the point of a push-pull design was to have each transistor do it's maximum conducting on opposite peaks, reducing the duty cycle for both devices.
Logically, the amp should never draw more current than a single device can handle, since the other one should be off or close to it at the point in the waveform.
Unless you have short, and then you'll find out what your unintended fusible links are.
Logically, the amp should never draw more current than a single device can handle, since the other one should be off or close to it at the point in the waveform.
Unless you have short, and then you'll find out what your unintended fusible links are.
TM86, you are exactly correct. Now think about a pair of 6146 tubes in the output stage of an older HF rig. Both tubes are combined in phase which means they are both conducting at the same time, but only on one half of the waveform. It provides half wave pulses to the tank coil which fills in the missing half cycle through the "flywheel" effect when the field collapses back into the coil.
This is almost identical to what I'm doing with a pair of 2879's. The only difference is since we are working with a much lower impedance using transistors, I've combined the transistors before the input circuit and after the output circuit on each transistor so that we are working with 50 ohms and not less than 1. The transistors also use coils in their matching circuits so the flywheel effect works here too with both transistors working at the same time for twice the capacity of push-pull.
This is almost identical to what I'm doing with a pair of 2879's. The only difference is since we are working with a much lower impedance using transistors, I've combined the transistors before the input circuit and after the output circuit on each transistor so that we are working with 50 ohms and not less than 1. The transistors also use coils in their matching circuits so the flywheel effect works here too with both transistors working at the same time for twice the capacity of push-pull.
OK, this could get the instantaneous current up to twice the current draw of a single device. The average should still be only be equal to a single device, since they're both off or close to it for half the cycle.
And had I bothered to read the rest of the thread, I would have already known that's already been accounted for.