Rectifier efficiency?

[tonestack]

EZ81 ~= 1.1 x secondary AC voltage

[dehughes]

I put together this one some yrs ago and have found it helpful as a starting place to consider designing a build.

With respect, 10thtx

Right on! Thanks! I'll put that information into the Numbers spreadsheet.

Now, I'm seeing discrepancies in some of this information. One says the rectified voltage increase of the EZ81 is 1.3, the other 1.1. Also, that chart says the voltage drop of a 5AR4 is 30v, elsewhere I'm finding 17v as the figure. Same with the 5Y3 info...some say it drops 60v, the chart says 20.

Should I just split the difference of all the conflicting information, or is there a definitive number somewhere?

[Sonny ReVerb]

The voltage drop varies with the load. See the 5U4GB datasheet in my earlier post. The rectifier becomes slightly more efficient at higher voltages, I believe.

[tonestack]

Now, I'm seeing discrepancies in some of this information. One says the rectified voltage increase of the EZ81 is 1.3, the other 1.1. Also, that chart says the voltage drop of a 5AR4 is 30v, elsewhere I'm finding 17v as the figure. Same with the 5Y3 info...some say it drops 60v, the chart says 20.

Should I just split the difference of all the conflicting information, or is there a definitive number somewhere?

Unlike Si diodes, tube rectifiers do not have fixed forward voltage drops. The equations that I gave you are approximations for what are, in fact, voltage and current dependent curves. For that very reason, I used the approximately-equal-to symbol instead of the equals symbol (i.e., "~=" instead of "=").

For example, here is a link to a set of data sheets for the EZ81: http://tdsl.duncanamps.com/link.php?target=632560BA

If you take a minute to view the 250-0-250 curve on page 4, you will notice that the rectified voltage starts off at ~1.4 x secondary AC voltage with zero current drawn. By the time we hit 100ma, the multiplier drops to ~1.08. As the EZ81 is commonly used in circuits that draw between 80 and 100ma, one can use 1.1 x secondary AC voltage as a quick and dirty starting point. After all, a tube amp is like a nuclear weapon--one only needs to get close to the target for it to work as designed!

[dehughes]

Thanks tonestack. That helps. I've always known that I should learn to "read" the curves on the data sheets, but have procrastinated in doing so.

So then, one could say that a 5V4 and 5R4 have voltage drop values of 23 and 45v, respectively, at their "average" current output?

As well, (to add to the spreadsheet), what is the maximum capacitor size you can place after a 1N4007 diode?

Look over this revision when you get the chance and let me know what corrections need to be made...

[tonestack]

Thanks tonestack. That helps. I've always known that I should learn to "read" the curves on the data sheets, but have procrastinated in doing so.

So then, one could say that a 5V4 and 5R4 have voltage drop values of 23 and 45v, respectively, at their "average" current output?

No, you are trying to force a constant forward voltage drop into what is actually a voltage and current dependent curve. The method that I suggested is a linear approximation of this curve.

Do you remember solving linear equations in high school? If so, then the values 1.1, 1.2, 1.3, and 1.4 are the "m" coefficient in the equation y = mx + b. The secondary AC voltage is the variable "x." The constant "b" is zero, and the variable "y" represents the rectified voltage. Therefore, when dealing with tube rectifiers, we need to find dy/dx (the slope of the line, which is the variable "m" in the equation shown above).

For example, using the quick and dirty method that I outlined in previous postings, we would get the following DC rectified voltages for a 5Y3.

1.1 x 200 VAC = 220 VDC
1.1 x 250 VAC = 275 VDC
1.1 x 300 VAC = 330 VDC

An Si diode has a fixed forward voltage drop of 0.6 volts, which for all intents and purposes is zero when dealing with tube voltages; therefore, the DC voltage is the square root of 2 x secondary AC voltage (the square root of 2 is ~1.4).

1.4 x 200 VAC = 280 VDC
1.4 x 250 VAC = 350 VDC
1.4 x 300 VAC = 420 VDC

Now, let's calculate the delta (difference) between the 5Y3 values and the Si values.

280 - 220 = 60
350 - 275 = 75
420 - 330 = 90

As one can see, the delta between the optimal rectified voltages (Si diode) and the 5Y3-rectified voltages is not constant, which tells us that there is a rate of change (a.k.a. slope) variable in the equation.

As well, (to add to the spreadsheet), what is the maximum capacitor size you can place after a 1N4007 diode?

Silicon diodes do not have the same constraints as tube rectifiers when used in capacitor-input circuits. For example, one will not kill a silicon diode with 200uf of capacitance.

Look over this revision when you get the chance and let me know what corrections need to be made...

Plug my coefficients into the conversion factors section and you will be a lot closer to real world numbers.

[dehughes]

Plug my coefficients into the conversion factors section and you will be a lot closer to real world numbers.

Okay....what are your coefficients again? (Sorry...I'm mathematically challenged...)

[tonestack]

Okay....what are your coefficients again? (Sorry...I'm mathematically challenged...)

Si diode 1.4
5AR4 1.3
5U4GB 1.2
5Y3 1.1

[M Fowler]

This is the same as I posted?

[dehughes]

Okay....what are your coefficients again? (Sorry...I'm mathematically challenged...)

Si diode 1.4
5AR4 1.3
5U4GB 1.2
5Y3 1.1

Right, got those...but I'm still trying to get the figures for the 5V4, 5R4, and EZ81, as well as the max capacitor value allowed after a 1N4007 diode...

[tonestack]

Right, got those...but I'm still trying to get the figures for the 5V4, 5R4, and EZ81, as well as the max capacitor value allowed after a 1N4007 diode...

I do not have values for the 5V4 or 5R4 because I have never found a need for either of these tubes. The coefficient for the EZ81 is 1.1.

As far as to a max capacitance value for the 1N4007: Si diodes work on a different principle than tube rectifiers. They can handle thousands of microfarads of capacitance. With an Si diode, one needs to be mindful of the current and peak inverse voltage (PIV) ratings. The PIV rating should be at least double the B+ voltage to prevent reverse breakdown. A 1N4007 has a 1A current rating and a 1000 PIV rating.

[M Fowler]

2Amp 5V4 voltage drop is 25v
2Amp 5Y3 voltage drop is 60v
5R4 not found ?5AR4 same as GZ34
EZ81
5U4G or GB 3Amp 44v to 54v drop depending on tube conducting
5AR4/GZ34 1.9Amps

[dehughes]

Great. Thanks. I like both the 5R4 and 5V4, especially the latter.

I suppose all I need now is the average voltage drop of the EZ81, and then the maximum filter capacitance ratings for the 5V4, EZ81, and 5R4 tubes. I'll have to scour the data sheets and see if I can find some which give that info (many do not have max. cap ratings on them).

[tonestack]

2Amp 5V4 voltage drop is 25v
2Amp 5Y3 voltage drop is 60v
5R4 not found ?5AR4 same as GZ34
EZ81
5U4G or GB 3Amp 44v to 54v drop depending on tube conducting
5AR4/GZ34 1.9Amps

With respect to what voltage? The current values listed above exceed the continuous ratings for these tubes.

[M Fowler]

I was using my old 1963 12th Ed. Sylvania Tecnical Manual I've had for so many years I can't remember.

So I also am posting this wedsite so that you can figure out the data you are looking for.

http://tdsl.duncanamps.com/show.php?des=5V4
http://tdsl.duncanamps.com/show.php?des=5Y3