Fellow experts😀 I have searched the forum, and ''googled''.......is it correct that when using a transformer with a 200ma capacity, a voltage doubler circuit in place of a normal SS rectifier configuration, will only provide HALF of that available current(100ma max)?
Any help would be appreciated.😀
Any help would be appreciated.😀
yep!
combine Ohms Law with the Conservation of Energy principle and there you are. In fact, there you are, less a bit of energy losses...
combine Ohms Law with the Conservation of Energy principle and there you are. In fact, there you are, less a bit of energy losses...
Ziggy said:
Recently, SY expounded on the 3 Laws of Thermodynamics. I submit his explanation reduces to a form of Murphy's Law. "You can't win, you can't break even, and you can't quit the game."
Seriously, in cap. I/P filter PSUs (including voltage multipliers), approx. 1/2 of the rectifier winding's AC RMS rating is available as DC. In a multiplier, that available current is distributed across each of the steps. So, in a doubler PSU plan on 1/4 of the power trafo RMS rating as DC current.
Actually, it's less than half, because the peak-to-average ratio is very high. Besides, you'll get poor performance if you use a doubler that close to the winding's rating. Doublers require low impedance to obtain reasonable regulation.
Tim
Tim
Yes, that probably explains why the Cit II power tranny high voltage secondary winding has a low impedance of only 11 ohms or so.
Yet, I once saw a Japanese valve amplifier circuit book with dozens and dozens of schematics - MOST were voltage doublers
.......so what gives if there are so many downfalls😕
.......so what gives if there are so many downfalls😕
ray_moth said:
Yes, but do remember that the voltage is only half that of a normal bridge topology, thus also half the winding resistance (and even less because thicker wire can be used - more space).
I have tried both, and am undecided. At the time it suited the available transformers. Ripple is somewhat worse with doublers. The post about the capacitor voltage ratings is valid - but then one could also simply use two capacitors in series with a normal bridge (with necessary resistor voltage dividers).
Not quite (depending on the circuit); one still gets two 'bumps' per mains cycle (one either side of the dc), but the total capacity must remain the same, thus twice for each capacitor, yes, since they are effectively in series.
Hi Johan,
Don't forget you are limited to around 1/2 the current. Your capacitance required will also be less. The ripple is easy to regulate out, or cancel in push-pull output stages.
I would tend to go for a full wave or bridge design. It's the sand in me. 😉
-Chris
Don't forget you are limited to around 1/2 the current. Your capacitance required will also be less. The ripple is easy to regulate out, or cancel in push-pull output stages.
I would tend to go for a full wave or bridge design. It's the sand in me. 😉
-Chris
Eli Duttman said:
Recently, SY expounded on the 3 Laws of Thermodynamics. I submit his explanation reduces to a form of Murphy's Law. "You can't win, you can't break even, and you can't quit the game."
Seriously, in cap. I/P filter PSUs (including voltage multipliers), approx. 1/2 of the rectifier winding's AC RMS rating is available as DC. In a multiplier, that available current is distributed across each of the steps. So, in a doubler PSU plan on 1/4 of the power trafo RMS rating as DC current.
Sch3mat1c said:
This is right on
In sizing a transformer for voltage doubler service I would choose
an RMS rating at 4X to 6X the DC output current. The peak current
the transformer designer assumes is 1.414X the RMS current if it's
a control transformer or other non-rectifier service transformer.
As you lower the DCR and ESR in the rectifier loop, the regulation
improves immensely, and at the same time the peak charging current
increases immensely. It's not uncommon for the peak charging
current to be 20X the DC output as in this PSUD example I just
happen to be working on.
This power supply is designed to deliver 1200V at 700mA, and uses
a 2KVA transformer with a 480V 4A winding. The low DCR of this
size transformer is nice for regulation.
This works out to about 2.5:1 ratio of KVA to delivered DC watts
and a current ratio of almost 6:1 RMS current to DC current out.
This is for a class AB amp so the regulation is important for good
performance.
Also, some transformers specified for rectifier service like the
Hammond 300, 700, etc. are specified in terms of full wave CT
choke input filter delivered DC current. For bridge and doubler
service I would use multipliers of 2X and 4X respectively to
scale the load current up to the transformer rating, and then
add a factor on top of that for cap input and doubler peak current.
Some transformers are rated for Continuous Commercial Service by
load and temperature rise. Here I like to build the safety factor in by
operating at a comfortably lower temperature than the max.
The most common mistake I made in my early electronics designs
was to undersize the power supply. No more ;-)
Michael
PS the diodes are actually going to be 2X series 1000V 16A HEXFREDs
PPS Peak currents that are too high seem to cause mechanical
buzzing in some transformers.
Attachments
...adding more of capacitance. Right. 😉anatech said:
...when idle, i.e. no sound. 😉
But sound will be intermodulated with ripples.
Wavebourn said:
I thought I was seeing 120 Hz ripple with the voltage doubler I just
built. It's a "scale model" of the big one I'm modeling in PSUD with
less capacitance but more inductance and one filter stage.
Two charging pulses per AC cycle, one on the positive peak and one
on the negative peak? I think of two 1/2 wave rectifiers with C input
filters, one positive and one negative, connected in series...
Michael
Hi Anatoliy,
-Chris
Nope. Not my style and adding capacitance can have it's own troubles.
That depends on how the gain stages are set up. To avoid "motorboating" you would need to cancel the ripple or use a regulator. Of course a classic R-C network in the B+ might be enough isolation, the ripple might still be high then. I would do something along these lines. An AC balance control may help here also.
-Chris
You probably did. The Greinacher, AKA "full wave", doubler produces ripple at 2X the mains freq. There is some justification for using full wave verbiage, even though the circuit is a pair of 1/2 wave rectifiers wired "back to back". Along with the ripple freq., the rectifier winding conducts on both halves of the waveform.
JMO, doubler B+ PSUs should be SS rectified. That allows the stack capacitance to be HUGE. Deal with the "hash" by using a low DCR choke and a reservoir capacitor.
Michael Koster said:
Sure. 60 Hz pulses charge each individual capacitor, but the load sees 120 Hz ripples.
anatech said:
Gain stages may be battery powered, but as soon as ripples present on B+ of a push-pull output stage it is called "balanced modulator" that null out ripples only and only when shoulders are ideally balanced, but the musical signal outbalance them by definition.
Another side effect of the voltage-doubler B+ supply is increased risk of more noise to other windings (heaters) off the same PT due to the floating secondary.
The voltage doubler PS is quite common in many of the cheaper Chinese tube amps, often with a slightly underdimensioned PT. A hum-balancing pots is usually used to reduce the effect of the noisy heater supply. The noise is not the typical clean hum, more like a "buzzzz" sound. this was discussed in a thread here
Many of these amps sound quite good though 🙂
SveinB
The voltage doubler PS is quite common in many of the cheaper Chinese tube amps, often with a slightly underdimensioned PT. A hum-balancing pots is usually used to reduce the effect of the noisy heater supply. The noise is not the typical clean hum, more like a "buzzzz" sound. this was discussed in a thread here
Many of these amps sound quite good though 🙂
SveinB
- Status
- Not open for further replies.