this is what i have so far for the psu -
* the transformer above
* a Gz32 tube rectifier
* 2x Hammond 157g chokes ( 30H , 40ma , 595 ohm , 400v )
I'm stuck on what capacitors to buy for the PSU - can anyone recommend what value capacitors ?
i have pretty much decided on 6cg7 input and 6922 on the output ... is that a good combo ?
im aiming at around 300v b+ , is that correct for those tubes ?
thanks everyone!
* the transformer above
* a Gz32 tube rectifier
* 2x Hammond 157g chokes ( 30H , 40ma , 595 ohm , 400v )
I'm stuck on what capacitors to buy for the PSU - can anyone recommend what value capacitors ?
i have pretty much decided on 6cg7 input and 6922 on the output ... is that a good combo ?
im aiming at around 300v b+ , is that correct for those tubes ?
thanks everyone!
ok after doing even more reserch , ive finally settled on this design of the power , hopefully it will be ok ...
opinions ?
opinions ?
An externally hosted image should be here but it was not working when we last tested it.
(Just FWIW) You might find this page interesting. http://www.vt52.com/
You'll need to navigate to [DIY Corner]-[Tips & Tricks]-[PS design]
You'll need to navigate to [DIY Corner]-[Tips & Tricks]-[PS design]
thanks for that ... ive revised my ps design a bit .
just 1 more question i think i have all values sorted out except r15.
but one more question re R15=
is it R15 = the mu of the [input tube ( mu-2) / the output tube (mu+2)] ?
or are they talking about the same tube ?
in all the examples they talk about one tube , i can only see one r15 value for both tubes.
just 1 more question i think i have all values sorted out except r15.
but one more question re R15=
is it R15 = the mu of the [input tube ( mu-2) / the output tube (mu+2)] ?
or are they talking about the same tube ?
in all the examples they talk about one tube , i can only see one r15 value for both tubes.
If you use schottky diodes on your filament supply, you can add a section and have a supply that should have less diode noise and lower output ripple. I used 1N5821's (anything close in spec. would be fine) and 10mF/1R/10mF/1R/10mF. Less than 30mV ripple. You can, of course, adjust the series resistance if the voltage is too high.
Sheldon
Sheldon
I;m going to try it with ac heaters first , and if it hums i will go DC ..
i think i have my chassis worked out ...
i;m slowly now doing the wiring ... should be ready to switch on by next weekend.
i think i have my chassis worked out ...
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
i;m slowly now doing the wiring ... should be ready to switch on by next weekend.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
look , im expecting the worst ... eg hum , problems... but at least i am giving it a go...
hey , im still trying to comprehend what referencing(elevating) the heater voltage means ?
the centre tap of the heater , goes to b+ .... hmmm how can this be right ?
ive seen a few schematics that show that ...
hey , im still trying to comprehend what referencing(elevating) the heater voltage means ?
the centre tap of the heater , goes to b+ .... hmmm how can this be right ?
ive seen a few schematics that show that ...
In the spec sheets you'll find that the voltage differential between the heater and cathode has limits (positive and negative). The heater is not connected to the cathode for DC but its insulation is limited. Heaters are 'normally' grounded as cathodes are 'normally' near ground, and leaving the heaters not tied to anything usually brings noise.
A heater is not meant to emit electrons but it will, possibly introducing hum. Raising the heater's DC above the cathode should prevent this. A common method is to run a 220k resistor from the B+ in series with another smaller value to ground.
This creates a simple voltage divider. Choose your lower resistor to give around 40V at their junction. Connect to the heater to 'float' it there, and put a capacitor across the lower resistor to shunt noise to ground.
There are other more elegant ways to do this including using an active voltage divider.
A heater is not meant to emit electrons but it will, possibly introducing hum. Raising the heater's DC above the cathode should prevent this. A common method is to run a 220k resistor from the B+ in series with another smaller value to ground.
This creates a simple voltage divider. Choose your lower resistor to give around 40V at their junction. Connect to the heater to 'float' it there, and put a capacitor across the lower resistor to shunt noise to ground.
There are other more elegant ways to do this including using an active voltage divider.
Referencing the heater voltage simply means connecting the center tap of the heater supply secondary to some point of your circuit. This point can generally be ground; there ARE exceptions to this of course.
Here's why: There is a capacitance between the cathode and the heater of each tube. If the heater circuit is left "floating", unreferenced, unconnected to some stable point of voltage, then the heater circuit via its capacitance to the cathodes can transfer signals from one tube to the other. This creates feedback paths that are not on the schematic. To visualize a simpler version of this draw a single lone wire on your schematic... then, from each cathode in your circuit, connect a small capacitor to the wire. Not we wanted... right? All the cathodes can talk to each other.
When you ground the heater circuit, all the capacitors are still there, but now the currents all coupled to ground. The cathodes do not talk to each other.
Now, connecting the heater to B+???? I cannot imagine ANY circuit where this could be correct... crazy.
There is a spec listed on most tube data sheets: Anode to heater voltage or vice-versa. There is an insulating oxide on the heater filament and sometimes some other goodies to prevent any DC current flow from the anode to the heater. This insulation can only stand so much voltage though... hence the spec.
In circuits where all the anodes are relatively close to ground, say within 50 Volts or so, grounding the heater supply would be typical because most tubes list the max heater to cathode voltage as a 100 volts or so... so all is well.
However, in the case of cathode followers and tubes acting as plate loads for other tubes, it is possible to have anodes that are much more than 100 Volts above ground, in these cases, the voltage specs would be violated and bad thing could be expected. Breakdown of the heater insulation.
If the range of anode voltages is extreme enough it is sometimes neccesary to use separate supplies for the different heaters. It is often the case though, that some intermediate voltage can be used to reference the heater circuit.
Let's say you had one cathode at 5 volts and another had a cathode at 100 Volts. Both tubes have a plus or minus K to H voltage spec. of 100 Volts. If we reference the heater to a 50 Volt point, we would stay well within the spec's. Where do we get the 50 Volts? A common practice is to use a simple voltage divider connected between B+ and ground.
Some things to watch for:
** The max rating for cathode to heater voltage is not always symmetrical. Plus or minus 100 Volts is common. But, there many tubes/specs that are unequal, such as: the heater must not be more than 100 Volts higher than the cathode, and must not be more than 200 Volts below the cathode. You must check out the data for your tubes.
** Some would insist, and with good reason, that the heater voltage should always be negative with respect to cathode. Let's say the heater had some minor flaw in the insulation, electrons from the cathode are attracted to this just as they are to the plate... this could set up a current flow you don't want. Is this practice always followed? No, but something to consider...
Have you posted your schematic? I could take peek and recommend a reference scheme for your heaters.
Here's why: There is a capacitance between the cathode and the heater of each tube. If the heater circuit is left "floating", unreferenced, unconnected to some stable point of voltage, then the heater circuit via its capacitance to the cathodes can transfer signals from one tube to the other. This creates feedback paths that are not on the schematic. To visualize a simpler version of this draw a single lone wire on your schematic... then, from each cathode in your circuit, connect a small capacitor to the wire. Not we wanted... right? All the cathodes can talk to each other.
When you ground the heater circuit, all the capacitors are still there, but now the currents all coupled to ground. The cathodes do not talk to each other.
Now, connecting the heater to B+???? I cannot imagine ANY circuit where this could be correct... crazy.
There is a spec listed on most tube data sheets: Anode to heater voltage or vice-versa. There is an insulating oxide on the heater filament and sometimes some other goodies to prevent any DC current flow from the anode to the heater. This insulation can only stand so much voltage though... hence the spec.
In circuits where all the anodes are relatively close to ground, say within 50 Volts or so, grounding the heater supply would be typical because most tubes list the max heater to cathode voltage as a 100 volts or so... so all is well.
However, in the case of cathode followers and tubes acting as plate loads for other tubes, it is possible to have anodes that are much more than 100 Volts above ground, in these cases, the voltage specs would be violated and bad thing could be expected. Breakdown of the heater insulation.
If the range of anode voltages is extreme enough it is sometimes neccesary to use separate supplies for the different heaters. It is often the case though, that some intermediate voltage can be used to reference the heater circuit.
Let's say you had one cathode at 5 volts and another had a cathode at 100 Volts. Both tubes have a plus or minus K to H voltage spec. of 100 Volts. If we reference the heater to a 50 Volt point, we would stay well within the spec's. Where do we get the 50 Volts? A common practice is to use a simple voltage divider connected between B+ and ground.
Some things to watch for:
** The max rating for cathode to heater voltage is not always symmetrical. Plus or minus 100 Volts is common. But, there many tubes/specs that are unequal, such as: the heater must not be more than 100 Volts higher than the cathode, and must not be more than 200 Volts below the cathode. You must check out the data for your tubes.
** Some would insist, and with good reason, that the heater voltage should always be negative with respect to cathode. Let's say the heater had some minor flaw in the insulation, electrons from the cathode are attracted to this just as they are to the plate... this could set up a current flow you don't want. Is this practice always followed? No, but something to consider...
Have you posted your schematic? I could take peek and recommend a reference scheme for your heaters.
cm_ls1 said:
No reason not to try AC heating. It may work fine.
Also, don't assume that hum is necessarily due to AC on the filaments. I had built some SE amps that were pretty quiet - your head had to be right in front of the 100dB+ horn to hear it. But I wanted to use one to drive headphones too. In that case, the hum was enough to be audible during quiet passages. I switched one amp to DC on the driver (6SN7) and the hum was reduced but only by half or maybe not that much. The leftover hum was loudest on the channel next to the power transformer - suggesting inductive coupling. I also noticed that the hum was even increased from before, when I first turned on the amp - also consistent with inductive coupling, as the current draw would increase while the filament supply caps were charging. So I played with shielding between the power transformer and OPT's. No hum at all now, just the faintest tube rush.
Compromised URL removed by Moderation
poobah said:OK... that's the power supply. Now show us the amp.
I think this is the pre in question: http://www.diyaudio.com/forums/showthread.php?postid=960937#post960937
Hey Brian,
Straighten me out here as Sheldon pointed out much the same and I am a tube dumbass.
I had read that electrons emit from both the inner and outer surfaces of the cathode... less so on the inside because of an absence of the special coatings. And that any breach in the heater insulation would soak these electrons up were the heater positive with respect to the cathode.
Made sense at the time. And there seems to be at least some tubes where the heater is allowed to go somewhat positive and much more so negative... from the specs anyway.
I failed to realize that a breached heater insulation could be a thermionic emitter in its own right.
DHT tubes types aside, which is it?
Or is it both? A tube with a leaky heater should go into EC8010's tube disintegrater?
SORRY cm_ls1... scratch that paragraph until Mr. Beck clears this up.
Straighten me out here as Sheldon pointed out much the same and I am a tube dumbass.
I had read that electrons emit from both the inner and outer surfaces of the cathode... less so on the inside because of an absence of the special coatings. And that any breach in the heater insulation would soak these electrons up were the heater positive with respect to the cathode.
Made sense at the time. And there seems to be at least some tubes where the heater is allowed to go somewhat positive and much more so negative... from the specs anyway.
I failed to realize that a breached heater insulation could be a thermionic emitter in its own right.
DHT tubes types aside, which is it?
Or is it both? A tube with a leaky heater should go into EC8010's tube disintegrater?
SORRY cm_ls1... scratch that paragraph until Mr. Beck clears this up.
this is what i did ... for a first attempt -
hope it works -
hope it works -
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.