The total VA rating divided by the total output voltage gives the maximum continuous AC current. If all the windings have the same current rating.
So for your example the total output voltage is 240Vac (120+120Vac).
The specified VA rating is 2000VA, therefore the maximum AC current is 2000/240 = 8.33Aac.
You cannot draw a continuous current any greater than 8.33Aac.
If that transformer feeds a capacitor input filter, then you must de-rate according to the manufacturer's information.
This usually results in the maximum continuous DC current from the capacitor input filter as being roughly half the maximum AC current.
i.e. for your 2000VA 240Vac transformer the maximum continuous DC current is roughly 4.17Adc
So for your example the total output voltage is 240Vac (120+120Vac).
The specified VA rating is 2000VA, therefore the maximum AC current is 2000/240 = 8.33Aac.
You cannot draw a continuous current any greater than 8.33Aac.
If that transformer feeds a capacitor input filter, then you must de-rate according to the manufacturer's information.
This usually results in the maximum continuous DC current from the capacitor input filter as being roughly half the maximum AC current.
i.e. for your 2000VA 240Vac transformer the maximum continuous DC current is roughly 4.17Adc
A typical guide for a ClassA amplifier's transformer is roughly 6times to 10times the maximum output power.
If you use a 400VA transformer, then that guide tells you that the maximum output power will be somewhere in the range 40W to 67W
That's roughly double the power of an F5 and will require a 2pair output stage instead of the F5's 1pair output stage.
Your 25+25Vac transformer will get you to roughly that ballpark, i.e. ~50W into 8ohms.
Last edited:
OK definitely have run that one into the ground. So from earlier post in this thread i can try changing values of R5/R6 to something around 2.2k to get bias set higher. What do I need to worry about if I do that. I am running fets in cascade and have about 45 volt rails. I did notice that I was clipping on the scope at +/- 32 volts. With 45 volt rails isn't that rather low. BTW that is without a load connected. Does that mean something is amiss? Thanks for all the help. Be patient with me.
Hmm I think there is a little misunderstanding here regarding your transformer.
Do I have this right?
You have a 2000VA transformer that has an input of 415Vac.
It has 2 windings that allows a series wiring that output 240Vac or you can parallel it to get 120Vac.
Since you are running it on 120Vac from the wall, you end up getting only around 35Vac output when wired in parallel.
Your question is, since it is a 2000VA transformer at 415Vac input do I get that full current rating with a 120Vac input and output of 35Vac (parallel wiring) ?
Is this correct?
Is this some sort of data centre power distribution transformer?
Do I have this right?
You have a 2000VA transformer that has an input of 415Vac.
It has 2 windings that allows a series wiring that output 240Vac or you can parallel it to get 120Vac.
Since you are running it on 120Vac from the wall, you end up getting only around 35Vac output when wired in parallel.
Your question is, since it is a 2000VA transformer at 415Vac input do I get that full current rating with a 120Vac input and output of 35Vac (parallel wiring) ?
Is this correct?
Is this some sort of data centre power distribution transformer?
Last edited:
Dazed you have it exactly right. I did not completely understand the ratings of the xfmr. This transformer is an industrial CPT used with international voltage levels.
Andrew has it correct it uses 2 phases of 3 phase on the primary. I do believe it is conservatively rated. It also carries a rating of 6kVA burst currents. Anyway I thought that I would be able to use this for a stereo amp. Maybe just for a monoblock now.
Anyway this is DIY and it is all about learning new things. I very grateful for all the help along the way.
Andrew has it correct it uses 2 phases of 3 phase on the primary. I do believe it is conservatively rated. It also carries a rating of 6kVA burst currents. Anyway I thought that I would be able to use this for a stereo amp. Maybe just for a monoblock now.
Anyway this is DIY and it is all about learning new things. I very grateful for all the help along the way.
I have been running with one channel on the bench. With 47 volt rails and 1 amp bias per output transistors. A total of 4 transistors, 2pr. I am using large heatsinks with forced air cooling. I realize that the transistor dissipation is high but well under their rating of 200watts each. Does the transistor dissipation matter that much as long as you can keep them cool? I am taking actual heatsink temps and with the fans they are running between 90-95deg F. I can hold my finger on the transistor case for 10 seconds or more. I have run a channel under these conditions for over 3 hours without issues. Can an amp be run safely under these conditions?
1 Amp per device at 47V?
IRFP240 datasheet says .83°C/W Rtjc. So keep in mind that the actual junction temperature of the device at your conditions (~50W dissipation) is always about 40°C above the temperature of the case.
150W rating is under ideal conditions, with case temperature of 25°C which would mean 150W*0.83°C/W + 25°C case = ~150°C = Tmax.
I can't speak out of experience, but I wouldn't want my devices to run at or above 100°C. In your example take 40°C heat sink, 0.5°C/W thermal resistance between device and heat sink, you'd be at 105°C junction.
Let the experts confirm, but I would not recommend this configuration. Add a third pair of output transistors so that they'll run at .6 Amps each. That way the temperature of each device would go down by about 15°C. Or use lower supply voltages, 47V sounds like a lot of class AB power.
IRFP240 datasheet says .83°C/W Rtjc. So keep in mind that the actual junction temperature of the device at your conditions (~50W dissipation) is always about 40°C above the temperature of the case.
150W rating is under ideal conditions, with case temperature of 25°C which would mean 150W*0.83°C/W + 25°C case = ~150°C = Tmax.
I can't speak out of experience, but I wouldn't want my devices to run at or above 100°C. In your example take 40°C heat sink, 0.5°C/W thermal resistance between device and heat sink, you'd be at 105°C junction.
Let the experts confirm, but I would not recommend this configuration. Add a third pair of output transistors so that they'll run at .6 Amps each. That way the temperature of each device would go down by about 15°C. Or use lower supply voltages, 47V sounds like a lot of class AB power.
The device must be de-rated for Tc
The device endures stress when passing signal to a resistive load.
The device endures more stress when passing signal to a reactive load.
After temperature de-rating there must be enough SOA margin for the load driving stresses.
Read David Eather.
Read ESP.
The device endures stress when passing signal to a resistive load.
The device endures more stress when passing signal to a reactive load.
After temperature de-rating there must be enough SOA margin for the load driving stresses.
Read David Eather.
Read ESP.
Hi All
I am planning to build two F5 Turbo monoblocks with Toshiba 2SK1530/2SJ201 outputs (four output pairs per monoblock). Rail voltage will be around +/-38 VDC. I believe I need to change some resistor values relative to the original as published by Nelson Pass. Which resistors should be changed, how, and why?
Thanks in advance for any hints!
I am planning to build two F5 Turbo monoblocks with Toshiba 2SK1530/2SJ201 outputs (four output pairs per monoblock). Rail voltage will be around +/-38 VDC. I believe I need to change some resistor values relative to the original as published by Nelson Pass. Which resistors should be changed, how, and why?
Thanks in advance for any hints!
The turn on Vgs of the Toshibas is around 2V while the ones in the schematic for the F5Tv3 are higher, so change R11/R12 to 1.21K and R5/R6 to 680R. This assumes your Idss of the Jfets is around 7ma. If much lower then increase R5/R6.
Nash
These mode (advise) can be used at F5Tv1?
My rail voltage will be 28V also use Toshiba fet but the smaller brother 2SK1529 & 2SJ200 3 pair per channel.. I do have some of those Toshibas I want to put then in good use.
Greetings
If you are using 3pair, then you are using semiconductors in parallel.
You MUST match paralleled semiconductors.
If you match them at your expected Id, then you will measure the Vgs. You can't match without measuring !
That gives you your starting point for selecting the Drain load resistors and jFET Id.
You MUST match paralleled semiconductors.
If you match them at your expected Id, then you will measure the Vgs. You can't match without measuring !
That gives you your starting point for selecting the Drain load resistors and jFET Id.