-
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.
Six pill High Drive Fat Boy
- Thread starter byron81ward
- Start date
in a way, yes they are.
while cool, their amps are in ab1 but after they heat up they start to lose bias and are no longer ab1
Do you know how they 'lose bias'?
If so, maybe you would like to share that information in this thread.
I am sure some of us would find that interesting and maybe something to learn from.
The 2x4 X-Force amp I have was originally biased Class "B" from X-Force. I found that the biasing was done very poorly (by design) and I removed their biasing, and re-biased everything myself. The transistors were biased at about .35 volts at the bases. Nowhere near close enough for AB1 operation.
I re-biased them using all Radio Scrap components, and they now run at about .62V at keydown. That brings it into AB1 territory. For info on the biasing, go into the homebrew section, and you'll see my thread on this conversion.
http://www.worldwidedx.com/home-brew/138726-x-force-amp-finished.html
The 600W was measured on a Bird 43A Thru-line wattmeter using a Cobra 2000 GTL doing 18W on SSB, to a Hy-Gain SP-500 with a flat SWR. I use Belden RG-8U solid center for coax.
Keep in mind that the input to the driver and final sections both are "padded". Note the resistors on the input transformer on the driver stage, and note the resistors in between the driver and final stages. Those resistors are "limiting" input wattage to a degree to reduce overdriving.
~Cheers~
I re-biased them using all Radio Scrap components, and they now run at about .62V at keydown. That brings it into AB1 territory. For info on the biasing, go into the homebrew section, and you'll see my thread on this conversion.
http://www.worldwidedx.com/home-brew/138726-x-force-amp-finished.html
The 600W was measured on a Bird 43A Thru-line wattmeter using a Cobra 2000 GTL doing 18W on SSB, to a Hy-Gain SP-500 with a flat SWR. I use Belden RG-8U solid center for coax.
Keep in mind that the input to the driver and final sections both are "padded". Note the resistors on the input transformer on the driver stage, and note the resistors in between the driver and final stages. Those resistors are "limiting" input wattage to a degree to reduce overdriving.
~Cheers~
i say "lose bias" but a more correct wording would have been "bias changes".
when you bias an amp you both lower efficiency and create extra heat.
heat changes tolerances.
tolerance changes cause voltage changes to the bias circuit.
most biasing circuits aren't regulated and as amp pulls harder, voltage drops causing a decrease in bias voltage.
and one thing that always makes me wonder... as far as i know and what i have always understood. the number "1" in ab1 makes the class ab refer to tubes not transistors.
"1" indicates that the tube draws zero grid current.
"2" indicates that the output stage grid voltage is being pushed above the 0 volt mark and into a positive grid voltage, causing the grid to draw some current from the driver stage.
so are texas stars biased to cutoff?
my understanding is that texas star amps are ab class
class A conducts current 100% of the time
class B has bias point to roughly 50% conducting, this was the old "push-pull" where two stages were used. one stage conducted the 1st half of the waveform and the second stage picked up just as the first was turning off.
class AB is somewhere between class A and class B.
where is up to the imagination... 30-40mils? 70-80 mils? a 2sc2879 calls for 100mils of bias current but you could go as far as 200mils per device as long as you don't overdrive.
so class AB has a "small amount" of bias current is flowing. In the push/pull output stage, there is a little overlapping conduction of each stage.
when you bias an amp you both lower efficiency and create extra heat.
heat changes tolerances.
tolerance changes cause voltage changes to the bias circuit.
most biasing circuits aren't regulated and as amp pulls harder, voltage drops causing a decrease in bias voltage.
and one thing that always makes me wonder... as far as i know and what i have always understood. the number "1" in ab1 makes the class ab refer to tubes not transistors.
"1" indicates that the tube draws zero grid current.
"2" indicates that the output stage grid voltage is being pushed above the 0 volt mark and into a positive grid voltage, causing the grid to draw some current from the driver stage.
so are texas stars biased to cutoff?
my understanding is that texas star amps are ab class
class A conducts current 100% of the time
class B has bias point to roughly 50% conducting, this was the old "push-pull" where two stages were used. one stage conducted the 1st half of the waveform and the second stage picked up just as the first was turning off.
class AB is somewhere between class A and class B.
where is up to the imagination... 30-40mils? 70-80 mils? a 2sc2879 calls for 100mils of bias current but you could go as far as 200mils per device as long as you don't overdrive.
so class AB has a "small amount" of bias current is flowing. In the push/pull output stage, there is a little overlapping conduction of each stage.
I could believe that. Maybe, just maybe ... that's why they sell a 4 fan accessory add-on.
you're correct but out of all the texas star users that you have seen, how many of them had fans mounted to cool the inside so that the diodes stayed at a steady temperature which would allow the bias to remain constant?
the fans are normally drawing air across the heatsink and that would take extra time for the cooling effect to make its way to the top side of the transistors where the diodes are, especially after a good longwinded keydown
[Edit] you would need a regulated bias circuit to counter thermal tracking for best results
and on that note, where is bob85? he knows way more about this subject than i do lol
[Edit 2] a version of a regulated bias circuit although I think bob85s design is better since it is switchable
the fans are normally drawing air across the heatsink and that would take extra time for the cooling effect to make its way to the top side of the transistors where the diodes are, especially after a good longwinded keydown
[Edit] you would need a regulated bias circuit to counter thermal tracking for best results
and on that note, where is bob85? he knows way more about this subject than i do lol
[Edit 2] a version of a regulated bias circuit although I think bob85s design is better since it is switchable
Last edited:
Ok, here are some pics I just took. These pics are taken of my Bird meter just now when I tested the X-Force amplifier. I used a different Cobra 2000 GTL in this test than what I used in the thread in my previous post, as this one does 16W SSB as shown.
Using a 100W slug, 16W.
The second pic shows 580W on the Bird meter. I used my B&K 1040 Servicemaster to generate audio for the test, two tone audio.
Using a 1000W slug, 580W out. 18W in gets me a little over 600W out, about 610 or so. Hope these pics put it into perspective of what I was trying to do.
~Cheers~
Using a 100W slug, 16W.
The second pic shows 580W on the Bird meter. I used my B&K 1040 Servicemaster to generate audio for the test, two tone audio.
Using a 1000W slug, 580W out. 18W in gets me a little over 600W out, about 610 or so. Hope these pics put it into perspective of what I was trying to do.
~Cheers~
biggest thing that you did was add more linearity to the amp by bringing it out of class c.
it's not as efficient as it was... power in vs power out... not drive power vs output power.
its using more working current now and basically doing the same output wattage with less drive but its cleaner audio
it's not as efficient as it was... power in vs power out... not drive power vs output power.
its using more working current now and basically doing the same output wattage with less drive but its cleaner audio
Yes!
Since I run SSB 99.9% of the time... I wanted to be able to run something that sounded good. I knew I had to modify this amp to run AB if I wanted to keep clean output, and not splatter the band. I try not to overdrive things, hence the 16W radio. The previous owner claimed over 1200W PEP... I get 600W. I bet I know where the other 600W went.
~Cheers~
Since I run SSB 99.9% of the time... I wanted to be able to run something that sounded good. I knew I had to modify this amp to run AB if I wanted to keep clean output, and not splatter the band. I try not to overdrive things, hence the 16W radio. The previous owner claimed over 1200W PEP... I get 600W. I bet I know where the other 600W went.
~Cheers~
Well I guess the only real way to settle the issue of loss of AB biasing is to test the bias voltage at the tracking diode after the amp has warmed up.
One would think that a loss of AB biasing to class C biasing would be noticed. Class C biasing has no biasing at low levels and effectively shuts off current to the transistor so it doesn't transmit the weaker, more subtle audio (or RF in this case) signals below the switching current level of the transistors. The result is a gravel-like sound when transmitting. An AB class biased amp will draw more current because it is providing more power at lower signal levels because the transistors are transmitting with all of the signal present; not just the signals that exceed the current threshold of class C biasing..
If it is no longer in the AB bias mode because of heat; then the thermal tracking diode is getting too much heat due to too much input power.
As I understand it, the whole point of the AB biasing is to get the transistors to track the signal more closely to what is being inputted. The goal is to get the transistor to be as linear as possible and keep out distortion. I know the transistors on the data sheets use 12.5v for optimal linearity and lowest possible distortion . . .
One would think that a loss of AB biasing to class C biasing would be noticed. Class C biasing has no biasing at low levels and effectively shuts off current to the transistor so it doesn't transmit the weaker, more subtle audio (or RF in this case) signals below the switching current level of the transistors. The result is a gravel-like sound when transmitting. An AB class biased amp will draw more current because it is providing more power at lower signal levels because the transistors are transmitting with all of the signal present; not just the signals that exceed the current threshold of class C biasing..
If it is no longer in the AB bias mode because of heat; then the thermal tracking diode is getting too much heat due to too much input power.
As I understand it, the whole point of the AB biasing is to get the transistors to track the signal more closely to what is being inputted. The goal is to get the transistor to be as linear as possible and keep out distortion. I know the transistors on the data sheets use 12.5v for optimal linearity and lowest possible distortion . . .
Last edited:
Here's a pic of the board of the amp. I have made a couple changes since I first completed the project, but I left the biasing alone. The main change was adding surge capacitors to the output transformers in each stage. That's about it.
~Cheers~
~Cheers~
Texas Star DX350 is biased from the RF output and the DX 1200/1600 are biased from the supply voltage.
In either case, since the bias circuit is not regulated, the amplifier goes from a state of less bias to more bias as the transistors gain more heat.
Even if the biasing resistors increase their resistance (PTC) with heat and causes less current to flow for the purpose of biasing the transistors the transistors hfe (gain) increases from heat and the amplifier moves from a state of less bias to a state or more bias.
Which is just the opposite of what has been stated previously in this thread.
In either case, since the bias circuit is not regulated, the amplifier goes from a state of less bias to more bias as the transistors gain more heat.
Even if the biasing resistors increase their resistance (PTC) with heat and causes less current to flow for the purpose of biasing the transistors the transistors hfe (gain) increases from heat and the amplifier moves from a state of less bias to a state or more bias.
Which is just the opposite of what has been stated previously in this thread.
- No one is chatting at the moment.
