Ok... I need to have 150v for my pre-amp at 40mA. Planning to use 1 or 2 VR150 reg tubes.
should I use one at 20mA or parallel 2 for 40mA..
so if I understand... I need at least 180v for strike... say 200v.. so 200-150 = 50v / 80mA (40mA for VR tubes and 40mA for pre-amp) is 625 ohm dropping resistor. will this work?
how do I equalize the current in the 2 tubes? and should I parallel them with a RC network to shunt the inductance ( I have heard 200 ohm and 0.3uf per VR tube).
power supply will be (per channe) 83 tube rect, 20H low DCR choke, 40uf oil cap, 625 resistor, 2 VR tubes, the RC filter network then my pre-amp.
my pre-amp is a u-stage so more PSRR there.
am i heading the "right" way???
should I use one at 20mA or parallel 2 for 40mA..
so if I understand... I need at least 180v for strike... say 200v.. so 200-150 = 50v / 80mA (40mA for VR tubes and 40mA for pre-amp) is 625 ohm dropping resistor. will this work?
how do I equalize the current in the 2 tubes? and should I parallel them with a RC network to shunt the inductance ( I have heard 200 ohm and 0.3uf per VR tube).
power supply will be (per channe) 83 tube rect, 20H low DCR choke, 40uf oil cap, 625 resistor, 2 VR tubes, the RC filter network then my pre-amp.
my pre-amp is a u-stage so more PSRR there.
am i heading the "right" way???
Hi,
Never // VR tubes, this doesn't work.
Do you need to regulate a tube drawing 40mA?
If so you'll need more than one VR tube in series set at about half their max operating current.
So a single OA2 (VR150) won't cut it, 2 VR75s in series might but check the datasheet for Imax.
Other alternatives are series regs or shuntregs, the latter not using VR tubes but a penthode or triode in shunt with the load.
Never use USED VR tubes unless you know exactly how they were used and intend to use them in that exact same way.
Unless of course you know how to condtion/recondition these tubes for a new service.
Cheers,
Never // VR tubes, this doesn't work.
Do you need to regulate a tube drawing 40mA?
If so you'll need more than one VR tube in series set at about half their max operating current.
So a single OA2 (VR150) won't cut it, 2 VR75s in series might but check the datasheet for Imax.
Other alternatives are series regs or shuntregs, the latter not using VR tubes but a penthode or triode in shunt with the load.
Never use USED VR tubes unless you know exactly how they were used and intend to use them in that exact same way.
Unless of course you know how to condtion/recondition these tubes for a new service.
Cheers,
Where did you got that from?, current trough the VR tube should be more then specified min, (usually 5mA) and less then max, (usually 30 or 40mA depending on type). Regulation is almost the same inside the specified current range.
I dont see any problem in your application. Your preamp draw 40mA continous current so if the voltage drop resistor is calculated for 60mA current the VR tube will draw 20mA and your preamp the rest.
An example:
Raw DC 200V, dropping resistor value = (200-150)/0.06 = 833 ohm, 820 ohm is OK.
Problem with using VR tubes comes when the load is changing like for instance in a class AB amp or when using them for screen voltage regulation, in class A circuits they are easy to use.
Regards Hans
can you tell us more please.
Hi,
Pretty much the same way one regenerates CRTs or big kW broadcast valves actually.
Just like capacitors have a memory effect so do valves; they can be trained for a certain service which can be quite beyond believe.
That's the "jesuit" answer...
Whether it's worth the time and energy put into it depends on what you want and how desperate you are but most of the time it works.
Cheers,
Pretty much the same way one regenerates CRTs or big kW broadcast valves actually.
Just like capacitors have a memory effect so do valves; they can be trained for a certain service which can be quite beyond believe.
That's the "jesuit" answer...
Whether it's worth the time and energy put into it depends on what you want and how desperate you are but most of the time it works.
Cheers,
Hi,
That will always do the job provided the B+ delivers the required voltage, i.e. more than 150V (180 something or thereabouts)
However if a non-regulated B+ is used the overvoltage at cold start can be used to ignite the VR as well, this is however a little trickier to get right.
I'm guessing but I think what was meant here is that the VR tubes regulate best when set to half their max current capability.
As you put it yourself it's almost the same within the operating window, however for Classes outside Class A this aspect should be taken into consideration to keep the VR tube well within it operating region.
Cheers,
That will always do the job provided the B+ delivers the required voltage, i.e. more than 150V (180 something or thereabouts)
However if a non-regulated B+ is used the overvoltage at cold start can be used to ignite the VR as well, this is however a little trickier to get right.
I'm guessing but I think what was meant here is that the VR tubes regulate best when set to half their max current capability.
As you put it yourself it's almost the same within the operating window, however for Classes outside Class A this aspect should be taken into consideration to keep the VR tube well within it operating region.
Cheers,
You can use 0A2 for your application. 0A2 runs at 150V from 5-30mA. But, you must have the following conditions:
1. The VR tube can only be paralleled by a 100n (0.1u) cap, maximum. No big electrolytics or you'll get relaxation oscillations.
2. If you want regulation, no filter stages can follow this stage.
3. Because the last PS cap is so small, all variations in class A current draw are seen by the VR tube and must be sourced/sunk by it.
4. You can idle the preamp at 40mA and the VR tube at ~17ma and it will regulate as long as your class A current swings do not exceed 12mA peak.
So, of the 40mA idle current, how much variation do you get at full output? If it's less than 12mA peak, you're in.
Other VR tubes will have different operating characteristics. But, so far as I know, they all have the small parallel cap limitation.
1. The VR tube can only be paralleled by a 100n (0.1u) cap, maximum. No big electrolytics or you'll get relaxation oscillations.
2. If you want regulation, no filter stages can follow this stage.
3. Because the last PS cap is so small, all variations in class A current draw are seen by the VR tube and must be sourced/sunk by it.
4. You can idle the preamp at 40mA and the VR tube at ~17ma and it will regulate as long as your class A current swings do not exceed 12mA peak.
So, of the 40mA idle current, how much variation do you get at full output? If it's less than 12mA peak, you're in.
Other VR tubes will have different operating characteristics. But, so far as I know, they all have the small parallel cap limitation.
. . .
First of all, YOU CANNOT HEAR A POWER SUPPLY. YOU ARE IMAGINING THINGS. Even if you COULD, you must be using a TERRIBLE circuit because a good designer goes for PSRR. Frank is spouting BS to the best and worst of my knowledge. Either a VR tube works or it doesn't.
Second, we need to know your expected current requirement and range. Typically a regulator is used where you need a voltage source which is stable over a range of currents. For instance, it's wasteful to use one on class A circuits because they are inherently constant-current and can be ran from R-C filters. Something like a class AB amp's screen supply is a good use for regulators.
But I know you probably don't care about this so I'll get on with operating a VR tube.
If your load has a current draw of 60 to 70mA for various conditions (say, zero signal vs. full power vs. overdriven, etc...) at 250V, you could use a 0C3 and 0D3 in series to regulate it. Here, 80mA total would be useful: thus when the circuit draws 60mA, the tubes compensate by drawing the remaining 20mA; when the circuit draws 70mA, the tubes draw 10mA. (The minimum is 5mA, below which it loses a bit of regulation; if load increases too much, voltage will drop, the tubes will extinguish and regulation will be lost until they ignite again.) At turn-on, with a rectifier which heats up before the other tubes (esp. SS which "heats up" instantly
), yes the tubes will carry the entire 80mA. This is no worry as these types are rated for a significant bit of current (100-150mA) for short periods - just this reason in fact.
You could just as well run them hot, which gains the advantage of better regulation (because for a given load variance, in the above case, 10mA, VR current changes less), but you run the risk of toasting them on turn-on, or more likely, if someone forgot to install a tube and they are ran at startup current continuously.
In your case, 40mA for the load, which I presume is class A, you can run the tube relatively cool, around 10mA (rather than the middle of their range, which is unneccesary because you only expect one level of current). Thus total consumption is 50mA, and from a 200V supply, (200-150)/.05 = 1kohm, (200-150)*.05 = 2.5W minimum. A 5 or 7W sand resistor would work nicely.
Or you could design the preamp to run off 200V with low PSRR. Whatever.
Tim
First of all, YOU CANNOT HEAR A POWER SUPPLY. YOU ARE IMAGINING THINGS. Even if you COULD, you must be using a TERRIBLE circuit because a good designer goes for PSRR. Frank is spouting BS to the best and worst of my knowledge. Either a VR tube works or it doesn't.
Second, we need to know your expected current requirement and range. Typically a regulator is used where you need a voltage source which is stable over a range of currents. For instance, it's wasteful to use one on class A circuits because they are inherently constant-current and can be ran from R-C filters. Something like a class AB amp's screen supply is a good use for regulators.
But I know you probably don't care about this so I'll get on with operating a VR tube.
If your load has a current draw of 60 to 70mA for various conditions (say, zero signal vs. full power vs. overdriven, etc...) at 250V, you could use a 0C3 and 0D3 in series to regulate it. Here, 80mA total would be useful: thus when the circuit draws 60mA, the tubes compensate by drawing the remaining 20mA; when the circuit draws 70mA, the tubes draw 10mA. (The minimum is 5mA, below which it loses a bit of regulation; if load increases too much, voltage will drop, the tubes will extinguish and regulation will be lost until they ignite again.) At turn-on, with a rectifier which heats up before the other tubes (esp. SS which "heats up" instantly
), yes the tubes will carry the entire 80mA. This is no worry as these types are rated for a significant bit of current (100-150mA) for short periods - just this reason in fact.You could just as well run them hot, which gains the advantage of better regulation (because for a given load variance, in the above case, 10mA, VR current changes less), but you run the risk of toasting them on turn-on, or more likely, if someone forgot to install a tube and they are ran at startup current continuously.
In your case, 40mA for the load, which I presume is class A, you can run the tube relatively cool, around 10mA (rather than the middle of their range, which is unneccesary because you only expect one level of current). Thus total consumption is 50mA, and from a 200V supply, (200-150)/.05 = 1kohm, (200-150)*.05 = 2.5W minimum. A 5 or 7W sand resistor would work nicely.
Or you could design the preamp to run off 200V with low PSRR. Whatever.
Tim
Sch3mat1c said:
That's true for push-pull class A, but not for a typical pre-amplifier, which is single-ended. Remember that a single-ended circuit produces its voltage swing by changing current through a load resistance. This means that the power supply's output impedance is indistinguishable from the designer's intentional load resistance. Since RC filters inevitably have an output impedance that rises at low frequencies, a regulator does make sense for a typical pre-amplifier.
Hi,
Maybe you're unaware of it but most amplifiers do have at least part of their PS in the signal path, but what do I know...
That only goes to show what you know...
May I suggest you either read again or get some experience? Or both?
Better still talk to someone who's actually worked at a tube factory. You'd be surprised at how much isn't actually in the public engineering books.
Cheers,
Maybe you're unaware of it but most amplifiers do have at least part of their PS in the signal path, but what do I know...
That only goes to show what you know...
May I suggest you either read again or get some experience? Or both?
Better still talk to someone who's actually worked at a tube factory. You'd be surprised at how much isn't actually in the public engineering books.
Cheers,
Well folks, I believe that my post above is correct (leaving out the possible contribution from the last dropping resistor).
sch3matic, class A devices are not constant current. The are only constant current when you average over a cycle (however you can define that with an audio signal). They are not instantaneously constant current. That means that there are current changes all the time in respose to the signal that only average out over time to zero.
Therefore, the PS must be able to respond to the instantaneous current demands. When there is a big cap at the last filter stage, this cap charges and discharges to follow the needs of the circuit and is continuously charged by the preceding PS stages. It shows a charactersitice impedance that is frequency dependent and does affect the signal.
The simplest experiment to prove this is to continuously make this capacitor smaller in value and measure the frequency response of the amp. As the cap gets smaller, the low frequency response of the amplifier will get poorer because the impedance of the PS starts getting higher and higher at low freqs. Period. You will be able to hear this.
When the last stage is a VR tube with a 100n cap, there is no big cap reservoir and the VR tube must source/sink the instantaneous current changes caused by the circuit.
Depending on the size of the last dropping resistor, it may not have to source and sink 100% of it, but usually it's close to that number.
So, you can't assume that setting the DC operating conditions to bias idle currents is sufficient. You also have to know how much current swing (in both directions) the circuit will ask of the PS. This maximum current swing must be within the regulating limits of the VR tube.
Now in a simple preamp, the current changes may be miniscule, say less than 1mA for maximum output. In which case, the DC idle conditions are essentially the AC conditions. But, on the headwize site we've been building WCF (class A topology) headphone drivers with 20mA current swings, regulated by VR tubes. And this is stretching the VR tubes outside of their regulating range, but is still working. Builders of this design are actually able to see the current changes in the VR tubes as their light output pulses with the audio signal. Meaning that they are actually regulating over a range of currents.
sch3matic, class A devices are not constant current. The are only constant current when you average over a cycle (however you can define that with an audio signal). They are not instantaneously constant current. That means that there are current changes all the time in respose to the signal that only average out over time to zero.
Therefore, the PS must be able to respond to the instantaneous current demands. When there is a big cap at the last filter stage, this cap charges and discharges to follow the needs of the circuit and is continuously charged by the preceding PS stages. It shows a charactersitice impedance that is frequency dependent and does affect the signal.
The simplest experiment to prove this is to continuously make this capacitor smaller in value and measure the frequency response of the amp. As the cap gets smaller, the low frequency response of the amplifier will get poorer because the impedance of the PS starts getting higher and higher at low freqs. Period. You will be able to hear this.
When the last stage is a VR tube with a 100n cap, there is no big cap reservoir and the VR tube must source/sink the instantaneous current changes caused by the circuit.
Depending on the size of the last dropping resistor, it may not have to source and sink 100% of it, but usually it's close to that number.
So, you can't assume that setting the DC operating conditions to bias idle currents is sufficient. You also have to know how much current swing (in both directions) the circuit will ask of the PS. This maximum current swing must be within the regulating limits of the VR tube.
Now in a simple preamp, the current changes may be miniscule, say less than 1mA for maximum output. In which case, the DC idle conditions are essentially the AC conditions. But, on the headwize site we've been building WCF (class A topology) headphone drivers with 20mA current swings, regulated by VR tubes. And this is stretching the VR tubes outside of their regulating range, but is still working. Builders of this design are actually able to see the current changes in the VR tubes as their light output pulses with the audio signal. Meaning that they are actually regulating over a range of currents.
Thankyou!
Hey guys thanks for all the help!!!
my actual pre-amp circuit is a 01a tube loaded by a mosfet source follower... in a u-follower.
The mosfet runs at 40mA... the 01a at around 2.5mA... look at the alan kimmel stuff for more info.
it is a pre-amp... so probably going to pull a max of 45mA to 35mA..
the source follower is loaded by a 2.6K resistor and and sits at 100v..
so.. say 10v output / 2600 = 3.8mA
I think just 1 VR tube would work... hey are there different types of VR tubes that glow different colors??? I have some right now that glow white....
Hey guys thanks for all the help!!!
my actual pre-amp circuit is a 01a tube loaded by a mosfet source follower... in a u-follower.
The mosfet runs at 40mA... the 01a at around 2.5mA... look at the alan kimmel stuff for more info.
it is a pre-amp... so probably going to pull a max of 45mA to 35mA..
the source follower is loaded by a 2.6K resistor and and sits at 100v..
so.. say 10v output / 2600 = 3.8mA
I think just 1 VR tube would work... hey are there different types of VR tubes that glow different colors??? I have some right now that glow white....
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.