Other techinques...
Quarter phasing:
Have you thought of using a technique sometimes called "quarter phasing"? This involves producing a quadrature (or similar) waveform so that there are more cycles to rectify and hence less smoothing required.
100Hz / 120Hz AC:
After full wave rectification, use AC coupling into a resonant filter to produce the double frequency AC. This will have far less thermal pulsing.#
Combination:
Use "Quarter phasing" rectification to produce 150Hz / 180Hz.
Just some mad ideas...
Cheers,
Quarter phasing:
Have you thought of using a technique sometimes called "quarter phasing"? This involves producing a quadrature (or similar) waveform so that there are more cycles to rectify and hence less smoothing required.
100Hz / 120Hz AC:
After full wave rectification, use AC coupling into a resonant filter to produce the double frequency AC. This will have far less thermal pulsing.#
Combination:
Use "Quarter phasing" rectification to produce 150Hz / 180Hz.
Just some mad ideas...
Cheers,
... to 'ear is human...
Only trouble with pushing the fundamental up to 150 /180 Hz is that the ear is more sensistive to the resisdual hum than at 50/60 Hz or 100/120 Hz again...
I don't know that the easier filtering gives you a gain over the higher sensistivity (it not just extra hearing gain but it is starting to interfere with the musical spectrum much more than the 50Hz does...)
In my book there is no substitute for throwing extra iron at it (as long as it doesn't saturate!) and pretending it is still ac - just with all the ripples smoothed out...
ciao
James
Only trouble with pushing the fundamental up to 150 /180 Hz is that the ear is more sensistive to the resisdual hum than at 50/60 Hz or 100/120 Hz again...
I don't know that the easier filtering gives you a gain over the higher sensistivity (it not just extra hearing gain but it is starting to interfere with the musical spectrum much more than the 50Hz does...)
In my book there is no substitute for throwing extra iron at it (as long as it doesn't saturate!) and pretending it is still ac - just with all the ripples smoothed out...
ciao
James
The more I look at it...
....The cleverer the idea of the indirectly heated cathode becomes.
I will be relying on CMRR (push-pull amplifier), so increasing the ripple frequency is probably a bad idea. Interesting thoughts, though, and well worthy of further investigation, if not for this particular application.
DC heaters for 2A3 using linear regulators would be wildly inefficient, and wear the cathode unevenly, even if periodically inverted.
If it wasn't for the linearity, I'd really hate this valve.
....The cleverer the idea of the indirectly heated cathode becomes.
I will be relying on CMRR (push-pull amplifier), so increasing the ripple frequency is probably a bad idea. Interesting thoughts, though, and well worthy of further investigation, if not for this particular application.
DC heaters for 2A3 using linear regulators would be wildly inefficient, and wear the cathode unevenly, even if periodically inverted.
If it wasn't for the linearity, I'd really hate this valve.
Fundamental orifi..
James,
Refreshing:
My thought was that a good deal of the hum we hear from DHT filaments is due to their low mass, and hence their short thermal memory. It's a matter of refreshing it more often.
Harmonious:
Another reason we can't seem to null it out completely, is that it contains harmonics. You can these see null at a slightly different point.
You may have hit the nail on the head with your "more iron" suggestion. By extra filtering, such as a tuned circuit, the spectrum would be much purer, and hence the AC balance pot, more effective at nulling.
Cheers,
James,
Refreshing:
My thought was that a good deal of the hum we hear from DHT filaments is due to their low mass, and hence their short thermal memory. It's a matter of refreshing it more often.
Harmonious:
Another reason we can't seem to null it out completely, is that it contains harmonics. You can these see null at a slightly different point.
You may have hit the nail on the head with your "more iron" suggestion. By extra filtering, such as a tuned circuit, the spectrum would be much purer, and hence the AC balance pot, more effective at nulling.
Cheers,
Re: Fundamental orifi..
There are two aspects to this. Thermally, a higher supply frequency allows the thermal inertia of the filament to smooth more effectively. Electronically, Vgk is varied by the heater supply frequency, and causes common-mode hum that is cancelled by push-pull action. CMRR tends to fall with increasing frequency.
dhaen said:
There are two aspects to this. Thermally, a higher supply frequency allows the thermal inertia of the filament to smooth more effectively. Electronically, Vgk is varied by the heater supply frequency, and causes common-mode hum that is cancelled by push-pull action. CMRR tends to fall with increasing frequency.
Now this is really silly...
An audio zero crossing detector that inverts the filament polarity each crossing. For a smooth ramping, it'll have to be predictive, hence a long audio delay line will be necessary. That will allow you to make a cup of tea, between dropping the stylus, and sound appearing...
An audio zero crossing detector that inverts the filament polarity each crossing. For a smooth ramping, it'll have to be predictive, hence a long audio delay line will be necessary. That will allow you to make a cup of tea, between dropping the stylus, and sound appearing...
The shame, the shame...
You'd be surprised, looking at the curves, how much of a difference 2.5v makes...
Yes, the 2A3's will be used push-pull, allowing hum cancellation (with a fair wind blowing). It's just that I'm fussy and want to make this amplifier to be as good as I can. I'll need to do some tests on whether hum is really a problem, or whether it's a figment of my imagination. (I broke a constant-volume gas thermometer when I was a kid - huge amounts of mercury exposed to the air.)
You'd be surprised, looking at the curves, how much of a difference 2.5v makes...
Yes, the 2A3's will be used push-pull, allowing hum cancellation (with a fair wind blowing). It's just that I'm fussy and want to make this amplifier to be as good as I can. I'll need to do some tests on whether hum is really a problem, or whether it's a figment of my imagination. (I broke a constant-volume gas thermometer when I was a kid - huge amounts of mercury exposed to the air.)
DC HEATERS....
Hi,
Seems most of you have already forgotten the dual rail supply I suggested a while a go.
Theoretically at least it would provide well regulated DC to the DHT and would prevent any cold spots on the cathode.
Use a center tapped balancing pot for nulling out any misalignment of the cathode structure and you're done.
Another thought for heater filtering iso of caps and chokes is to use a low voltage valve as a ripple killer.
Not fit for higher current levels, I admit.
Cheers,
Hi,
Seems most of you have already forgotten the dual rail supply I suggested a while a go.
Theoretically at least it would provide well regulated DC to the DHT and would prevent any cold spots on the cathode.
Use a center tapped balancing pot for nulling out any misalignment of the cathode structure and you're done.
Another thought for heater filtering iso of caps and chokes is to use a low voltage valve as a ripple killer.
Not fit for higher current levels, I admit.
Cheers,
Re: DC HEATERS....
I remember your posts now. Unfortunately i did not understand, and failed to ask for your clarification. I think there was a lot of background noise on that thread
So, if you've the patience - please reveal all
Cheers,
Frank,fdegrove said:Hi,
Seems most of you have already forgotten the dual rail supply I suggested a while a go.
Theoretically at least it would provide well regulated DC to the DHT and would prevent any cold spots on the cathode.
snip...
Cheers,
I remember your posts now. Unfortunately i did not understand, and failed to ask for your clarification. I think there was a lot of background noise on that thread
So, if you've the patience - please reveal all
Cheers,
No, not forgotten
I don't even dare calculate the efficiency of a pair of 1.25V 2.5A linear regulators!
Unfortunately, because anode current is theoretically dependent on the three-halves power law (Child's law), it isn't possible to balance out AC heater hum. In practice, emission is more linear than predicted, and better cancellation is possible, but it's still not quite perfect.
I don't even dare calculate the efficiency of a pair of 1.25V 2.5A linear regulators!
Unfortunately, because anode current is theoretically dependent on the three-halves power law (Child's law), it isn't possible to balance out AC heater hum. In practice, emission is more linear than predicted, and better cancellation is possible, but it's still not quite perfect.
Hi,
Well, you can't have you're cake and eat it too.
Hi John,
The advantage of AC is that it feeds the cathode evenly every 50/60 cycles once balanced out using the virtual midpoint of the humbucking pot.
AC being AC it pulsates hence the hum...
Why not try to do the same but this time using DC?
No pulsing, no him, I figured...
When using a fullwave bridge we have the possibility to make a dual rail, regulate those and attach + and - at either side of the cathode.
By using the same kind of pot we can now make sure the DHT is heated evenly so no cold (islands) spots should be present.
I think that was the basic idea but I can't for the life of me remember in what thread I posted it...
Quite likely it was my response to your concern about unenly heated cathodes though.
Keep in mind, I've never tried this...it was just an
that popped up when someone ( maybe yourself) expressed the concern.
Cheers,
Well, you can't have you're cake and eat it too.
Hi John,
The advantage of AC is that it feeds the cathode evenly every 50/60 cycles once balanced out using the virtual midpoint of the humbucking pot.
AC being AC it pulsates hence the hum...
Why not try to do the same but this time using DC?
No pulsing, no him, I figured...
When using a fullwave bridge we have the possibility to make a dual rail, regulate those and attach + and - at either side of the cathode.
By using the same kind of pot we can now make sure the DHT is heated evenly so no cold (islands) spots should be present.
I think that was the basic idea but I can't for the life of me remember in what thread I posted it...
Quite likely it was my response to your concern about unenly heated cathodes though.
Keep in mind, I've never tried this...it was just an
that popped up when someone ( maybe yourself) expressed the concern.Cheers,
Frank,
As far as I can see, any DC scheme will produce an "emissive" gradient along the length of the heater. The higher the heater voltage, and the greater the mu, the worse the effect will be.
I cannot see the difference between running the heater between 0v and 2.5v, or -1.25v and +1.25v. (All voltages relative to ground). The gradient remains the same, the bias point changes by 1.25 volts.
The pot you mention would be good at nulling any residual ripple of the DC supply, but unless decoupled with a lrge value cap, becomes an entry point for common mode noise.
Have I missed something?
Cheers,
As far as I can see, any DC scheme will produce an "emissive" gradient along the length of the heater. The higher the heater voltage, and the greater the mu, the worse the effect will be.
I cannot see the difference between running the heater between 0v and 2.5v, or -1.25v and +1.25v. (All voltages relative to ground). The gradient remains the same, the bias point changes by 1.25 volts.
The pot you mention would be good at nulling any residual ripple of the DC supply, but unless decoupled with a lrge value cap, becomes an entry point for common mode noise.
Have I missed something?
Cheers,
Hi,
Sure enough.
The idea however is to split the heater into two parts which hopefully would reduce the emissive gradient and by the same token give us more even emission of the cathode.
From my manager days I always find it useful to split "big" problems into smaller ones...by the time I'm done the problem usually has solved itself...
I was hoping, no more than that really, that by the divide et impera technique, I could somehow make the cathode believe it had equal I and V coming from both ends and behave accordingly.
I.e. heat the cathode element in a more even manner...probably just my wishfull thinking.
Good idea...it's indeed best to decouple the pot to avoid CMN breakthrough, thanks.
Cheers,
Sure enough.
The idea however is to split the heater into two parts which hopefully would reduce the emissive gradient and by the same token give us more even emission of the cathode.
From my manager days I always find it useful to split "big" problems into smaller ones...by the time I'm done the problem usually has solved itself...
I was hoping, no more than that really, that by the divide et impera technique, I could somehow make the cathode believe it had equal I and V coming from both ends and behave accordingly.
I.e. heat the cathode element in a more even manner...probably just my wishfull thinking.
Good idea...it's indeed best to decouple the pot to avoid CMN breakthrough, thanks.
Cheers,
Nuts
I've often wondered if it was possible to heat 2.5V filaments with a large pair of 1.25V lead-acid batteries in series , deriving the centre tap from between the batteries in an output stage running fixed bias with 1ohm resistor to 0V with floating charging circuit . Would this work or am I nuts ?
316a
I've often wondered if it was possible to heat 2.5V filaments with a large pair of 1.25V lead-acid batteries in series , deriving the centre tap from between the batteries in an output stage running fixed bias with 1ohm resistor to 0V with floating charging circuit . Would this work or am I nuts ?
316a
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.