Rectifiers

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
It's a little more complicated than that. What type of load do you have on the rectifier? You'll have a different multiplier based on whether you have a resistive load, capacitor input load or a choke input load. Your multiplier of 1.41 assumes a full wave/half wave capacitor input load with solid state rectifier. In addition, you'll get higher peak output from solid state rectifiers than from tube rectifiers in general, due to the plate resistance of a tube diode.

BT
 
80hinhiding said:
I just tried a 6x4 tube rectifier.  Put 269VAC on each plate, took a reading from the transformer high voltage center tap and the cathode, and got a reading of 244VDC.

I had been expecting to see around 358VDC, thinking it'd be (VAC * 1.414) -22 on the secondary.

Is something wrong or do tube rectifiers not increase the voltage?

Adam
Did you put a capacitor? The 1.414 factor works only if the voltage is filtered. In case there's no capacitor, the rectified signal is not DC, the measurement is absolute value, identical to AC.
 
"I just tried a 6x4 tube rectifier.  Put 269VAC on each plate, took a reading from the transformer high voltage center tap and the cathode, and got a reading of 244VDC".

Greetings ,First post.
I thought the 6x4 was a full wave rectifier, I think you need to ground the secondary HV center tap.
Then with tube full wave rects its like 1.1 -1.3  x half the secondary. Depends on tube used.
A solid state bridge would be 1.414 x full winding. no center tap.

GJ

 
Gregg Johns said:
"I just tried a 6x4 tube rectifier.  Put 269VAC on each plate, took a reading from the transformer high voltage center tap and the cathode, and got a reading of 244VDC".

Greetings ,First post.
I thought the 6x4 was a full wave rectifier, I think you need to ground the secondary HV center tap.
That's what the OP did in measuring the voltage between cathode and center tap.

A solid state bridge would be 1.414 x full winding. no center tap.
Actually, the formula used by the OP is more specific.
 
80hinhiding said:
Okay perfect guys thanks as always.  I just added a high voltage 220uF 450V cap and a resistor divider and it now reads 341VDC, and drains pretty quickly after the power is turned off.  :)

Adam
220uF with a 6X4? Wow! You're gonna fry this poor valve.
 
abbey road d enfer said:
220uF with a 6X4? Wow! You're gonna fry this poor valve.

Yup.  Download one of the RCA Receiving Tube Manuals online.  RC-30 shows a "typical" filter cap for the 6X4 as 10uF.  Also has lots of neat info for choke input filters, voltage vs load graphs, etc.
 
There is no "maximum" capacitance. It depends on how much resistance is in the transformer, or added in the circuit.

The tube-book "typicals" are ball-park for typical transformer resistances.
 
80hinhiding said:
It was cool seeing all the old designs and seeing references to picture tubes, etc.
Those schematics in the old RCA manuals are a wealth of information, with good circuit descriptions. There is much to be learned from good ol' fashioned paper and ink.

I had a copyright '73 since '73, it is falling apart from overuse, found a '63 at a hamfest, and then a '40, the '40 is much smaller. In checking the copyright date on the '73 to write this, there was a piece of sandpaper as a bookmark, opened to 6CA7/EL34. What a surprise. ::)

Gene

 
do you like doing complicated Fourier Analysis?


that's what the propeller heads say it is going to take to calculate DC voltage from transformer AC.


so lazy engineers use a graph and then maybe check it in the lab.


in this graph, R-S is your transformer secondary resistance added to the resistance of your vacuum tube.


R-L is your load, which will be determined by what kind of circuit you are powering.


what the graph says is that if you are dragging a lot of current thru your power supply, no matter how big a cap you use, you will never get near your peak AC voltage provided by the transformer.


but if your load is light, like 300 volts / 10 ma = 300 K ohms, then the more capacitance you add, the higher your DC voltage will be. But now we have a regulation problem. Your DC voltage will vary depending on the load, maybe some musical transients will drag that voltage down a bit, so you want to have a big enough cap so that you are to the right of the "knee" in this graph.


so if you have a load of 300K and your transformer resistance and rect tube resistance is 300 ohms, your percentage of R-S to R-L will be small, like 300/300 K = 0.1 percent. Therefore your  curve will look like the one on the top of this graph. So you will need a large cap to get max DC voltage from your pwr supply, then, to improve regulation and ripple, you can add some RC pi sections .

As long as your ripple current is low, you should be able to use a large cap as the peaks will not fry your tube. but using a large cap on a circuit that draws a heavy load would not be a good idea.


you could go choke input to provide a steady current to the rect tube, but you would lose a lot of voltage. regulation would be better but nobody uses that circuit anymore so scratch that.


see that in this graph, once you get past the knee, adding more capacitance does not really help, so the trick is to find the value that gets you a bit past the knee and use that so you keep your ripple peaks down.

also see that once you get a big enuff cap, a variance in R-L (your load) will not change the DC voltage very much, (good regulation)

once note of interest is it appears that for a given load resistance, decreasing either the secondary resistance or the rectifier resistance will give you more voltage, but your regulation will decrease.
seems counter intuitive,  :eek:

"w" on the horizontal axis represents 2 pi f, so that will be a constant, so your variables are C and your load, R-L.


i think this is all correct, if not i can apologize later.  ;D

 

Attachments

  • pwr supply.JPG
    pwr supply.JPG
    265.9 KB · Views: 16
CJ said:
so lazy engineers use a graph and then maybe check it in the lab.
As far as I can see, this graph ignores the effects of the transformer's inductance, which is generally adequate in the context of vacuum rectification, where the dominant factor is the added resistance of the valve and the DCR of the xfmr.
In the case of SS rectification, this can be no more neglected, though; as you mentioned earlier: "you could go choke input to provide a steady current to the rect tube, but you would lose a lot of voltage". It's a challenge xfmr winders have to face in linear high-po audio amplifiers, making the leakage inductance low enough.

 
here is a figure showing how a choke, which resists a change in current, can provide a steady current flow to the rectifier tube, I1 and I2 , but resulting in bigger voltage peaks, which means more capacitance or more pi sections needed to flatten that out,

and a cap input pic showing a smother ripple voltage but increased stress on the tube in the form of higher peak current values,

you would thing that the tube would cool off during peaks and thus see the same stress as the choke input due to the reduced duty cycle, but apparently this is not the case, and seeing first hand rectifier arc displays in C input pwr  supplies seems to bear this out,
 

Attachments

  • choke.JPG
    choke.JPG
    140 KB · Views: 20
CJ said:
here is a figure showing how a choke, which resists a change in current, can provide a steady current flow to the rectifier tube, I1 and I2 , but resulting in bigger voltage peaks, which means more capacitance or more pi sections needed to flatten that out,

and a cap input pic showing a smother ripple voltage but increased stress on the tube in the form of higher peak current values,
I don't really understand this graph, because the rectified voltage is not smoothed; in audio applications, there's always a smoothing cap. It looks to me like the rectification used with high-po DC motors.
 
80hinhiding said:
So, does this mean using an inductor should yield constant current, more headroom in the dynamics, but cause a shorter life span for the rectifier tube?
You're inferring too much; constant current would not result in more headroom. Headroom is essentially voltage dependant. Sag is counter-productive for headroom, and although choke input and cap input differ in that respect, there is no practical difference in the context of a preamp. Power amps are another issue.

And you're misunderstanding what's been written. Input choke is better for life span of rectifier (smoothes current peaks).

BTW a choke does not yield constant current really; as any good inductor, it resists changes in current, but only up to a certain point. No way a choke can yield constant-current when the load goes from short-circuit to open-circuit.
 
I think very few would argue against the benefits the slowly ramping up the ht .Applying ht before the cathodes of the tubes have had a chance to get up to temp can have a drastic effect on tube life. I suspect the gradual rise of ht makes for better longevity of PSU caps also . A standby switch for silicon diode rectified units seems very worthwhile to me , I ve seen lots of modern tube gear  that applys full unloaded ht volts to the tubes from the instant the power is turned on . Ive modified some smaller amps in my own collection so that as soon as the unit is plugged in the heaters warm ,and then modified the power switch into a standby , there is a potential downside if the unit is left in standby mode for a very long period of time ,but same is true of any tube equipment with a standby switch.

Can anyone give me any insight into using foil caps after a tube rectifier , I only need 5uf to reach the required ht voltage from the transformer I have , Im looking at using the LCR brand motor run cap followed by two stages of LC with big electrolytics. I remember seeing some designs that used 50uf foil reservoir and it didnt take long for the rectifier to die catastrophically .
 
Tubetec said:
Applying ht before the cathodes of the tubes have had a chance to get up to temp can have a drastic effect on tube life.

That is often said on internet forums but I don't think it's true for receiving tubes.
 
Hi Heikki,
Yeah Ive read the arguments for and against ,but almost any  amp with silicon diodes Ive seen  that was without a standby was a tube muncher,and an extra toggle switch costs a fraction of what a new set of tubes cost. So for me its a very worthwhile exercise to always include the standby .
 
> for a given load resistance, decreasing either the secondary resistance or the rectifier resistance will give you more voltage, but your regulation will decrease.

"For a given load resistance", regulation is not in the picture.

Do you mean off-load to given load? Still, lower series resistance is less sag. And if "regulation" is the amount of sag, lower is better.
 
> this graph ignores the effects of the transformer's inductance, which is generally adequate in the context of vacuum rectification

Compute the reactance of the leakage inductance and add it to the resistance. This is not quite right, but will never be far wrong, and will be conservative. The R drops the top, the L rounds the corner, but "R"=R+xL won't be far wrong.

Also depends on transformer size, as it generally does in power systems. When I figure the lamp-dim in my house, including L makes a minor difference; but in large distribution networks the L dominates to the point that they can sometimes neglect R.
 
Back
Top