• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

tube rectifier

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Suppose you have a preamp with solid-state rectification (the Audio Electronic Supply AE-3). Dennis Had, the designer of the preamp (and also Cary Audio owner/designer) offers one particular upgrade for this preamp- tube rectification. The purpose of rectification is to correct osciliations in voltage, correct? What would be the advantage of having him upgrade the preamp to tube rectification, and would it be worth it? I have heard that tube rectification is someone less reliable, are there any other negatives? Thanks!

bonsai
 
analog_sa said:
It is not less reliable. A rectifier tube in a preamp will give you thousands of hours of service before replacement.

Provided they are treated with respect.

I have heard that some (commercial) manufacturers use something like 1000µF behind their valve rectifiers with no apparent regard to the charging current those caps will cause.

Valve rectifiers will most certainly not be reliable when mistreated.
 
1000µF behind their valve rectifiers
Ouch!! Just a thought though, you know in PSU designer it tells you when you exceed the current rating of a given rectifier, couldnt you go a little bigger on the first cap as the rectifier turns on slowly? Not instantly as it is simulated. Also are the current ratings of rectifiers absolute, or an average current rating, i.e. more like a max dissipation rating?
Cheers,
Steve
 
baggystevo82 said:

...when you exceed the current rating of a given rectifier, couldnt you go a little bigger on the first cap as the rectifier turns on slowly?...
Cheers,
Steve

No,

Most (all?, I haven't checked) rectifier tubes have info in the data sheet regarding the max size capacitor you can use immediately after the tube in a circuit. For instance the 5U4-G can have a max of 40uF in a cap input circuit.
 
bonsai171 said:
What would be the advantage of having him upgrade the preamp to tube rectification, and would it be worth it?

The major advantage will be Had's ability to make this month's payment on the BMW. And the enhancement of your ability to impress the fashionistas with your preamp.

Tube rectifiers have one real advantage: slow warmup. Other than that, they're costly, degrade regulation, inefficient, and hot.
 
Sherman said:


No,

Most (all?, I haven't checked) rectifier tubes have info in the data sheet regarding the max size capacitor you can use immediately after the tube in a circuit. For instance the 5U4-G can have a max of 40uF in a cap input circuit.


And with modern electrolytics, which surely must be different from those made 50 years ago, does that number change much?

Sheldon
 
baggystevo82 said:

Ouch!! Just a thought though, you know in PSU designer it tells you when you exceed the current rating of a given rectifier, couldnt you go a little bigger on the first cap as the rectifier turns on slowly? Not instantly as it is simulated. Also are the current ratings of rectifiers absolute, or an average current rating, i.e. more like a max dissipation rating?

For one thing, voltage across the rectifier rises with a certain slowness due to the sine wave input. For another, the tube is "soft"; unlike semiconductors, it doesn't have a hard exponential forward voltage characteristic. It's closer to a resistance. 5Y3 has approximately 300 ohms forward resistance (far less at low voltages, a bit more at higher voltages). Even during a full half cycle with an empty capacitor, this 300 ohm impedance limits current; this plus the series resistance of the supply winding (often comparable in the 50-500 ohm range depending on ratings) determines the hot switching transient current. If the tube isn't rated for this stress, ploop!

Since current through the rectifier depends on the value of the cap and its load, as well as the voltage left by the previous cycle, the equation would be differential and I don't care to come up with that, let alone solve it. You'll have to refer to PSUD to see how these conditions relate to peak current.

As for ratings, you can draw as much current as you want, until plate dissipation is exceeded. There isn't a rating on the 5Y3 datasheet for dissipation, but it shows a graph of efficiency (and how to determine this value) vs. current draw. Given the GT envelope, you might expect 10 or 15 watts before it starts glowing too much.

Sheldon said:
And with modern electrolytics, which surely must be different from those made 50 years ago, does that number change much?

No; ESR was, and is, a small fraction of the supply transformer's resistance. That's one thing that hasn't changed - the iron and copper alloys used to wind transformers haven't changed, and neither have the efficiency, size or regulation ratings.

Tim
 
rdf said:
I thought tube rectifiers also have softer VI knees and generate less rectification hash than typical diode bridges. Can be considered an advantage, YMMV.

If given a choice I'd always use SS... tubes don't know the meaning of the word regulation (relating to my above post, consider supply impedance for cap input is about four times the series resistance - that's 1.2kohm minimum for a 5Y3!). Reverse recovery is only for people who don't know how to handle it, pretty simple to avoid getting it into the signal path.

Tim
 
rdf said:
I thought tube rectifiers also have softer VI knees and generate less rectification hash than typical diode bridges. Can be considered an advantage, YMMV.

Yes, verifiably correct. Add to this that any noise generated on a B+ rail will, of necessity, appear, in part, on the output of the stage or amplifier in question. Avoided-noise, if you will, is absent in a way that filtered-noise never can be.

The above suggests the proposition that tube rectification might carry the advantage from a sonic point of view.
 
Tom, note well the difference between measuring something early on in the supply versus what actually appears on the rails. After all, there's 60Hz (50 for the metric set) across that transformer primary!

Rectifier noise is not at all difficult to keep boxed away from the audio circuit.
 
analog_sa said:
Heard? From whom? Doesn't sound very plausible in a commercial design.

I don't want to name the company - it might get diyAudio or myself in trouble... :hot:

Just because it's in a commercial design, it doesn't mean things are done properly. As could be expected, forums like Audioasylum get questions about why the rectifiers only last a few months and keep suffering internal arcs from owners of these amps.
 
Sch3mat1c said:
If given a choice I'd always use SS... tubes don't know the meaning of the word regulation....Tim

Isn't regulation only a factor if the circuit 'sees' the rectifier? Most of the tube PSUs I've seen here use R and/or L to drop the rectified voltage. In some multi-section cases it doesn't seem to matter much at all if the R is a tube or wirewound heating up the chassis internals. That's certainly the case for the tiny tunes amp I built, PSUD2 shows the same regulation of a step load current at the OPT primary for both the SS and 5AR4 cases once B+ is adjusted equivalently.

Using RC circuits across diodes to damp ringing is also a bit uncomfortable to me - fully admitting I haven't done any comparing - after all the surprising simulations results of RC PS filtering. Achieving good suppression at frequencies higher than ~70 kHz with common components is hard. With the present amount of junk on the lines, and talk of powerline broadband on the way, just the thought of opening paths to the junk present on the secondary is unappealing but I'm sure there's a way to handle it. I recall seeing common-mode chokes off the secondary of some of Thorsten's circuits.

But the real reason I used a 5AR4 is, of course, the kewl mega-glow factor and that it was cheaper than a delay relay, socket, diodes and circuit board. Had the tube socket, can't beat $11 CDN for excess coolerosity.
 
Sch3mat1c said:
If given a choice I'd always use SS... tubes don't know the meaning of the word regulation ..Tim


Unless you use a gas rectifier and/or a choke input power supply. Gas rectifiers have a relatively small & constant voltage drop, that's why old tube testers use them.

I'd think that in a preamp (which normally has a constant current draw) with decently filtered B+ , the choice of rectifier and the presence of a regulated power supply (or lack thereof) would be immaterial. But some folks who have a lot more time than I do to listen, say they can hear a difference. I'm not one to judge, some folks claim they can taste differences in beer that I can't taste, or like stuff I think is mediocre. Some folks like me like White Castles, others, ahem, have different opinions. I like a good paella, too.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
When it sounds good

I have used both HexFreds and various valve rectifiers in the same circuit when I was developing my diy preamp. Must have tested about 15 different NOS and current production tubes.
Findings are that valves sound clearly better than HexFreds. Its also amazing how strongly influential on the circuit was the sound of each valve even when it was same type, even same NOS make, but different production year! At a point I got stunned when I changed a wirewound series resistor leading to the first (20uf) capacitor before 3 cap-choke stages. That resistor got burned when a short occured by a misswired earth wire when a friend experimented on my circuit. I happened to have only a leftover Holco 1W at hand having the correct value. I did not put it in to listen for it...well I picked it up on first musical phrase reproduced!
My moral is: When it sounds good better use it. Then search why. Its always more productive to get technically theoretical after a pleasing audition of a proven upgrade.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.