Hi Andrew,
This amp runs Class A all the time and draws 5As minimum at idle, 2.5As each voltage rail through the 40A capable bridge rectifiers. You would expect the rectifiers to got hot, but they run fairly cool, could be their rating verses the current draw, don't know. The output transistors on the other hand idle at 50C at the sinks and about 54C at their TO-3 cases, but then they are attached to a large mass of aluminum to spread the heat to the sinks. I am using a laser thermometer to measure things, I will have to try a more primative thermometer on the rectifier to cross check but the touch test says cool.
This amp runs Class A all the time and draws 5As minimum at idle, 2.5As each voltage rail through the 40A capable bridge rectifiers. You would expect the rectifiers to got hot, but they run fairly cool, could be their rating verses the current draw, don't know. The output transistors on the other hand idle at 50C at the sinks and about 54C at their TO-3 cases, but then they are attached to a large mass of aluminum to spread the heat to the sinks. I am using a laser thermometer to measure things, I will have to try a more primative thermometer on the rectifier to cross check but the touch test says cool.
You have this wrong.
If the output bias is 2.5A then the supply rail quiescent current is also 2.5A, not 5A
If you have two or more channels running from the same supply rails, then the total supply rail current is the bias current times the number of channels sharing that supply.
You have this wrong as well.
A ClassA amplifier has an output bias.
add on a little bit for the front end stages.
The supply rail current is then 2.5A + 0.01A, ~ 2.51A
When the amplifier is delivering power to a load the supply rail currents change.
as one current goes up the other one goes down.
When the ClassA amplifier is delivering it's maximum ClassA current one rail is supplying twice the bias current plus a little bit, about 5.01A
The other rail in the meantime is delivering zero load current plus a little bit.
i.e. the minimum supply rail current for F5 and similar Push Pull ClassA amplifier is ~0.01A, not 5A
The load current is the DIFFERENCE between the supply rail currents, i.e 5.01A - 0.01A = 5A
Your rectifier and the smoothing caps will see some, or most, of this changing current demand.
Last edited:
i do have some problem believing that those "tiny" sinks can handle 500W dissipation and maintaine 25-30c above ambient.
Last edited:
Re the bridges, each diode in the bridge drops 0.7V and passing average 5A. Actually it is not so simple because this is 50Hz AC and not continuous, but rectifiers do run pretty hot even in a regular F5 let alone a turbo. If you have cool rectifiers and a cool sink of that size, you're not pushing 500W.
Hi Andrew,
The output bias is 5A, front end is very low. 40 output transistors, each drawing .125A, .125v across their 1 ohm emitter resistor, half on the positive voltage rail and half on the negative, or 2.5 Amps per rail. Each rail has a 40 amp bridge for the dual output toroid. These amps are monoblocks.
Hi AudioSan,
The heat sinks are not "tiny". There are 8 individual sinks 8" long x 4" wide with 1.3875" fins. Behind that is 3" wide by 14" long by 1/2" thick aluminum heat spreader with the transistors bolted to a 6" wide by 14" long by 1/4" aluminum plate attached to the heat spreader, all with thermal grease between, which adds internal heat radiating as well as the external sink. It will maintain the 25C rise in temp, and another 5C more under load, and has since 1989 when first built.
Hi sangram,
5 amps * 115v = 575 W. Each rectifier in the bridge is handling 1/4 of that or 143 watts, the rectifiers are rated at 40 amps which at 115v would be 4600 watts. I don't know why the bridge rectifiers are not running hotter but they are not. I will check them again after run music for 2 hours and see what their temp is again.
Have a good one,
John
The output bias is 5A, front end is very low. 40 output transistors, each drawing .125A, .125v across their 1 ohm emitter resistor, half on the positive voltage rail and half on the negative, or 2.5 Amps per rail. Each rail has a 40 amp bridge for the dual output toroid. These amps are monoblocks.
Hi AudioSan,
The heat sinks are not "tiny". There are 8 individual sinks 8" long x 4" wide with 1.3875" fins. Behind that is 3" wide by 14" long by 1/2" thick aluminum heat spreader with the transistors bolted to a 6" wide by 14" long by 1/4" aluminum plate attached to the heat spreader, all with thermal grease between, which adds internal heat radiating as well as the external sink. It will maintain the 25C rise in temp, and another 5C more under load, and has since 1989 when first built.
Hi sangram,
5 amps * 115v = 575 W. Each rectifier in the bridge is handling 1/4 of that or 143 watts, the rectifiers are rated at 40 amps which at 115v would be 4600 watts. I don't know why the bridge rectifiers are not running hotter but they are not. I will check them again after run music for 2 hours and see what their temp is again.
Have a good one,
John
that is tiny sinks for 500W idle dissipation. very very tiny.
its 2.5A bias pr rail? and rail voltage is? it must be +/-100V to dissipate 500W.
its 2.5A bias pr rail? and rail voltage is? it must be +/-100V to dissipate 500W.
Last edited:
The rail voltage is +/-48.5v or 97 volts at 5A = 485 w.
Maybe each rail is contributing 5 A per rail, I am thinking backwards, half the total Voltage on each rail not current. Duh. Oh well, it has been a long week with the new puppy waking me up at night.
Maybe each rail is contributing 5 A per rail, I am thinking backwards, half the total Voltage on each rail not current. Duh. Oh well, it has been a long week with the new puppy waking me up at night.
I'm going to have to hook up an amp meter to be sure rather than calculated numbers. I will try it tonight.
OK, last night I measured the current on each supply rail and it is indeed 2.56 amps each, or about 5 amps total for about 250 watts total at idle. Also after 3 hours of running the amp and all temps stable I read 40C at the bridge rectifiers and further back on the aluminum angle bracket at the capacitor screw terminals the temp was 35C. While only about 10C over ambient it is still an excellent suggestion to move the bridge rectifiers to the bottom case when I put the new bank of caps in.
Thanks,
John
Thanks,
John
A rule of thumb is that for every 10 degrees C cooler, the capacitors last twice as long. Or vice versa.
Hi,
I have just finished my F5 Turbo V2, and I have a question. My PCB comes from ebay, and transistors and diodes are placed in one line. The question is can transistors and diodes be placed on the same heat sink ?
I have just finished my F5 Turbo V2, and I have a question. My PCB comes from ebay, and transistors and diodes are placed in one line. The question is can transistors and diodes be placed on the same heat sink ?
My toroid started up with the diyaudio softstart circuit!
I want to build a three box amp with transformer in one box and amps in a separate box. The transformer is center tapped. I have the white PSU boards from Tea Bag.
The options are:
1) One diode bridge feeding one cap bank for both channels in the transformer box. ( DC via umbilical cord)
2) One diode bridge, feeding two cap banks (one per channel) in the transformer box.( DC via umbilical cord)
3) One diode bridge, feeding two cap banks (one per channel) in the amp box.( Rectified AC via umbilical cord)
4) two diode bridges, feeding two cap banks (one per channel) in the amp box. ( Rectified AC via umbilical cord)
5) two diode bridges, feeding two cap banks (one per channel) in the transformer box. (DC via umbilical cord)
6) two diode bridges, feeding four caps (one per rail) in the transformer box, and cap bank in the amp box via umbilical cord.
Of all solutions, it appears that 1) would have the least ripple voltage, since it is the largest capacitor bank. However, there is a benefit in transient response with the caps inside the amps.
Options 1-5 use only the parts that came with the kit. Option 6 uses 4 extra caps. Not sure if there is anything to gain with option 6 over option 4...
Any comments are greatly appreciated. THANKS!
I want to build a three box amp with transformer in one box and amps in a separate box. The transformer is center tapped. I have the white PSU boards from Tea Bag.
The options are:
1) One diode bridge feeding one cap bank for both channels in the transformer box. ( DC via umbilical cord)
2) One diode bridge, feeding two cap banks (one per channel) in the transformer box.( DC via umbilical cord)
3) One diode bridge, feeding two cap banks (one per channel) in the amp box.( Rectified AC via umbilical cord)
4) two diode bridges, feeding two cap banks (one per channel) in the amp box. ( Rectified AC via umbilical cord)
5) two diode bridges, feeding two cap banks (one per channel) in the transformer box. (DC via umbilical cord)
6) two diode bridges, feeding four caps (one per rail) in the transformer box, and cap bank in the amp box via umbilical cord.
Of all solutions, it appears that 1) would have the least ripple voltage, since it is the largest capacitor bank. However, there is a benefit in transient response with the caps inside the amps.
Options 1-5 use only the parts that came with the kit. Option 6 uses 4 extra caps. Not sure if there is anything to gain with option 6 over option 4...
Any comments are greatly appreciated. THANKS!
Hi BigE,
I like two bridges, one for each transformer winding (just lessen the load on each bridge) one bank of caps in the PS box, and if you have the room in the amp box then use two caps, one for each rail, and some film caps to bypass for higher freq close to the load. Or just go with #1, simply is often better.
Have a good one,
John
I like two bridges, one for each transformer winding (just lessen the load on each bridge) one bank of caps in the PS box, and if you have the room in the amp box then use two caps, one for each rail, and some film caps to bypass for higher freq close to the load. Or just go with #1, simply is often better.
Have a good one,
John