Hi Andrew,
Do you have an example of a similar power supply (CLC or CRC) showing a connection scheme that is not flawed? I'm having trouble visualizing the returns for the second stage filter caps.
Thanks!
There are two sets of caps. The ground for the first set of caps should connect to the rectifier and secondary as necessary. The ground for the second set of caps should connect to the amp circuit. Then connect these two grounds together; this connection carries some ripple current but no charging pulses. This keeps charging pulses away from the amp. The second, improved, version given in the example does not do this.eastinontario said:
Maybe it just semantics but wouldn't class A have 1/2 the supply current variation per rail, as half the load current current comes from an increase in the high side and the other half from a decrease of the low side idle current.
Thanks
-Antonio
Something along those lines, depending on topology, I never stated that ClassA has no supply rail current variation, just less than ClassAB.
vac
Do you mean the PSU caps input ground (rectifier etc) and output ground (amp circuit)? They can be very close together, as long as they are not coincident. The idea is simply that the PSU output cap ground has one and only one connection to the charging current loop, so no charging current escapes into the signal circuitry.eastinontario said:
The charging current loop should be as small as possible to reduce magnetic induction - use twisted wires or close parallel PCB tracks where possible.
Hello,
I remember reading that using a choke will reduce inrush current.
If i am using a 300 va 30volts ac transformer with one single feeding, a bridge made from 4 mur 860 diodes, an 100millihenry, 1.3 A 1.3ohm input choke folowed by 27000uf with a 100ohm bleeder across the first cap , then another choke and finally a combination of 47000uf and 27000uf in the amp itself would i need a current limiter on the primary side? The amp will be switched on by its own switch once the outboard power supply is ''charged'.
I have some alumium 25 resistors tp make a 60ohm for the primary side. Don't wanna use a relais just an extra switch if the inrush current wouldn't be temepoered by the chokes.
Thanks in advance. Kind greetings, Edward
I remember reading that using a choke will reduce inrush current.
If i am using a 300 va 30volts ac transformer with one single feeding, a bridge made from 4 mur 860 diodes, an 100millihenry, 1.3 A 1.3ohm input choke folowed by 27000uf with a 100ohm bleeder across the first cap , then another choke and finally a combination of 47000uf and 27000uf in the amp itself would i need a current limiter on the primary side? The amp will be switched on by its own switch once the outboard power supply is ''charged'.
I have some alumium 25 resistors tp make a 60ohm for the primary side. Don't wanna use a relais just an extra switch if the inrush current wouldn't be temepoered by the chokes.
Thanks in advance. Kind greetings, Edward
Hello,
The chokes were a gift from a friend. They look nice and they are from a reputed manufacturer. Maybe later i will try to find bigger ones with more millihenry and less dc resistance.
My main concern was the current during inrush because the mur 860 are only rated 8A. Today i did receive the caps so i plan to start soldering this weekend.
I can use a variac and see what happens if i reduce the tension offered to the primary side of the transformer.
I remember with my hiraga Le classe A with a 500va 21 volts ei transformer, a 25A bridge, then 68.000uf, then a 200 millihenry, 2,5 A and 0,8 ohm and finally 2*330.000uf the lights did dimm but the fuse didn't break.
Also remember that a resistor in primary side will have a hard life. Once i did have a 25 watt explode. With the alumium housing the internal winding has only 2 ways out left or right.
Total current from the circuit and the bleeder will be about 1A. With my tube amps the high tension will rise slowly because of the much higher induction used. I think if it takes a few seconds for the tension to arrive at its working point the caps will have an easier life.
Thanksalot, sincere greetings, Edward
The chokes were a gift from a friend. They look nice and they are from a reputed manufacturer. Maybe later i will try to find bigger ones with more millihenry and less dc resistance.
My main concern was the current during inrush because the mur 860 are only rated 8A. Today i did receive the caps so i plan to start soldering this weekend.
I can use a variac and see what happens if i reduce the tension offered to the primary side of the transformer.
I remember with my hiraga Le classe A with a 500va 21 volts ei transformer, a 25A bridge, then 68.000uf, then a 200 millihenry, 2,5 A and 0,8 ohm and finally 2*330.000uf the lights did dimm but the fuse didn't break.
Also remember that a resistor in primary side will have a hard life. Once i did have a 25 watt explode. With the alumium housing the internal winding has only 2 ways out left or right.
Total current from the circuit and the bleeder will be about 1A. With my tube amps the high tension will rise slowly because of the much higher induction used. I think if it takes a few seconds for the tension to arrive at its working point the caps will have an easier life.
Thanksalot, sincere greetings, Edward
The Pangea design (out board ps for Musical Fidelity boxes) is a disaster, but the one thing they do demonstrate well is that a little inductance goes a long way in this era of wall-wart SMPS. they put the choke after the linear regulator which just bolixes things up badly. had they put it in front of the regulators the effect would have been beneficial.
I don't understand at all why one would put a choke in the ground leg of a 3-terminal regulator. It's almost comical -- and they get $150 for the product.
I don't understand at all why one would put a choke in the ground leg of a 3-terminal regulator. It's almost comical -- and they get $150 for the product.
The resistor in the primary feed of the transformer is intended as a soft start, i.e. it limits the start up current of the transformer and allows a close rated fuse to be used. This resistor must be bypassed very quickly, other wise the now operating transformer will draw sufficient current that the resistor will very rapidly overheat.
I suggest the relay bypass across the soft start resistor be timed to operate at between 100ms and 300ms after initial start switch ON.
If you want to limit the charging current of the smoothing capacitors then investigate slow charging of capacitors. These are normally NTC resistors that are fitted in the secondary feed to the rectifiers. Again a relay bypass should take the NTCs out of circuit, but for this duty the time delay should be much longer. Expect 5s to 20s as a reasonable time for the capacitor to substantially charge using the limited current available through the NTC.
I suggest the relay bypass across the soft start resistor be timed to operate at between 100ms and 300ms after initial start switch ON.
If you want to limit the charging current of the smoothing capacitors then investigate slow charging of capacitors. These are normally NTC resistors that are fitted in the secondary feed to the rectifiers. Again a relay bypass should take the NTCs out of circuit, but for this duty the time delay should be much longer. Expect 5s to 20s as a reasonable time for the capacitor to substantially charge using the limited current available through the NTC.
Hello Andrew and all the others,
I did do some test a week ago with only 10000uf, a 25A bridge, a choke input with 100 millihenry with a bleeder that draw enough current to make it work as a choke input. Also did do a test with a resistor representing the actuyalk load from the amp just to see if the choke wouldn't buzz ( could only hear it buzz with my stethoscope ).
Now i will use it with the caps that i will be using later, a second choke and a bridge made from 4 mur 860.
With the 10.000 uf i did use a variac because the cap wasn't very young. Later did not use the variac anymore.
Like i said i can always fire it up slowly with the variac . My main concern are the diodes. The amp ( virtue sensation m901) has a sofstart function to allow it to function with big battery or regulated power supply as well )
From the past i remember that a big transformer did cause a big surce at switching on. My caps are a lot smaller than my old hiraga which wasn't choke input. My future amp will have a second 1.3 ohm choke between the 2 caps so i guess the charging currents will be kind of tempered.
Because of the layout from the outboard power supply a resistor in series with the first choke and a switch to bypass that resistor would be the easiest way to go . I have a few 25 aluminium 15 ohm 25 watt.
Thanksalot, Edward
I did do some test a week ago with only 10000uf, a 25A bridge, a choke input with 100 millihenry with a bleeder that draw enough current to make it work as a choke input. Also did do a test with a resistor representing the actuyalk load from the amp just to see if the choke wouldn't buzz ( could only hear it buzz with my stethoscope ).
Now i will use it with the caps that i will be using later, a second choke and a bridge made from 4 mur 860.
With the 10.000 uf i did use a variac because the cap wasn't very young. Later did not use the variac anymore.
Like i said i can always fire it up slowly with the variac . My main concern are the diodes. The amp ( virtue sensation m901) has a sofstart function to allow it to function with big battery or regulated power supply as well )
From the past i remember that a big transformer did cause a big surce at switching on. My caps are a lot smaller than my old hiraga which wasn't choke input. My future amp will have a second 1.3 ohm choke between the 2 caps so i guess the charging currents will be kind of tempered.
Because of the layout from the outboard power supply a resistor in series with the first choke and a switch to bypass that resistor would be the easiest way to go . I have a few 25 aluminium 15 ohm 25 watt.
Thanksalot, Edward
Aluminium cased resistors are not designed to handle short-term surges, as you discovered. This is partly because the resistance element is quite small so has little thermal capacity. Other power resistors can do better; reading datasheets can be helpful. Basically you need a resistor which is fairly large for its continuous power handling if you want good surge performance.
I offered 20r 5W wirewounds a while back. No takers, even though I thought the price was very attractive to UK Members.
I use these and 10r as soft start resistors. I assemble a stack of 3 to 7 side by side for 30r to 140r bypassable soft start. I also use these for high power dummy loads (45off 20r in series parallel for a 4r0 200W).
These assemblies have quite high thermal capacity and can stand high overload in the short term. Very different from metal film which do not tolerate overload at all well.
I use these and 10r as soft start resistors. I assemble a stack of 3 to 7 side by side for 30r to 140r bypassable soft start. I also use these for high power dummy loads (45off 20r in series parallel for a 4r0 200W).
These assemblies have quite high thermal capacity and can stand high overload in the short term. Very different from metal film which do not tolerate overload at all well.
Hello,
I have the big ceramic wirewound resistors as well. They have the dimensions of a good cigar. However don't have the space to put them inside the cabinet. One needs space to get rid of the heat.
Today did make some more metalwork so i think i will be able to fire it up this weekend.
Sincere greetings, Edward
I have the big ceramic wirewound resistors as well. They have the dimensions of a good cigar. However don't have the space to put them inside the cabinet. One needs space to get rid of the heat.
Today did make some more metalwork so i think i will be able to fire it up this weekend.
Sincere greetings, Edward
If you are using wirewounds for soft start duty, then don't develop/deissipate much heat. They are (or should be) in circuit for at most 300ms.
Yes, the peak dissipation is very high, but that is for such a short time that the resistors don't even get warm on the outside. I suspect the wire inside would tell us a very different story. That is precisely why one needs high thermal capacity and why one needs high overload tolerance for these soft start resistors.
A bank of radial 5W wirewounds do not take up "too much" space/volume.
Yes, the peak dissipation is very high, but that is for such a short time that the resistors don't even get warm on the outside. I suspect the wire inside would tell us a very different story. That is precisely why one needs high thermal capacity and why one needs high overload tolerance for these soft start resistors.
A bank of radial 5W wirewounds do not take up "too much" space/volume.
Hello,
Today i did do the test with 4 mur860 and just the bleeder to draw 250ma to make it a choke input. I did use a variac but i think i did have a short circuit in the hardwiring of the 4 mur860. I didn't even had the time to turn the variac up. I will make a new bridge with more space.
With an old fashioned kmpc1001 bridge i could fire it up even with a big 40 ohm bleeder on the output cap to simulate the actual load. If i leave the variac in '' high voltage position '' there is some noise at switching on but i don't know if it is the variac or the power supply transformer. The variac is fused. I haven't fused the outboard supply yet. After doing that i can fire it up directly.
The power supply will be on all the time so don't need slow inrush to make the caps last longer.
Maybe i will look for an audiophile bridge instead of the 8A mur860. A lot easier then using perforated board, individual diodes with their individual heatsinks.
Thanksalot so far, Edward
Today i did do the test with 4 mur860 and just the bleeder to draw 250ma to make it a choke input. I did use a variac but i think i did have a short circuit in the hardwiring of the 4 mur860. I didn't even had the time to turn the variac up. I will make a new bridge with more space.
With an old fashioned kmpc1001 bridge i could fire it up even with a big 40 ohm bleeder on the output cap to simulate the actual load. If i leave the variac in '' high voltage position '' there is some noise at switching on but i don't know if it is the variac or the power supply transformer. The variac is fused. I haven't fused the outboard supply yet. After doing that i can fire it up directly.
The power supply will be on all the time so don't need slow inrush to make the caps last longer.
Maybe i will look for an audiophile bridge instead of the 8A mur860. A lot easier then using perforated board, individual diodes with their individual heatsinks.
Thanksalot so far, Edward
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.