power tube burn-in rig

[JC@]

I'm trying to figure out the easiest way to burn-in power tubes for a future matching.
Still in my mind the project begun to grow in complexity: circuit brakers or resetable fuses for each tube socket, then a LED each to show me the faulty ones...

I did a google search this morning and I found this has already been done in the present days:
http://www.upscaleaudio.com/rare/tpm.htm
By the way, I discovered here something I did not take into account, 2nd grid current draw meassurements.

Not to mention that my original idea was to really BURN-IN the tubes!
That is, basically a mega-huge simple amp with signal applied to it and putting real power on the loads. I believe that could be the only way to burn-in power tubes.

So I don't know what to do.

[Enzo]

First off, in the audiophile world, the whole game is to look for esoterica that someone else doesn't pay attention to. Like matching screen currents... As soon as the novelty of that wears off, someone will figure out a way to match cathode temperatures, socket pin inductances, and lunar phase compensations.

Just how many tubes you need to heat? 4 or 40? Got an old Fender amp kicking around? Slap four of them into it at a time and let it run for a half hour. Certainly won't hurt the Fender to have - gasp - mismatched tubes in it.

40? OK, mount 40 octal sockets and wire heater current to them all. Do you really have to load the tubes and run them at power to get them to settle in? To me, they just need to get hot. After all this gadget isn;t doing the matching. 40 tubes? Well, that 40 amp 6v transformer ought not to cost much...

Have you settled on a budget for this?

If you simply HAVE to run the tubes, then you need a power supply.bonus. Those 640 watts should keep the shop nice and warm on a winter day.

And the sudden increase in electrical use ought to get you reported to the DEA, so expect a visit. You won't be lonely.

OK, I am poking fun, but you need to define your goals a bit.

[JC@]

OK Enzo I get the point. After all heating them is a lot more than doing nothing.
I will begin my project with just a big enclosure, 25 octal sockets and a big 40A 6 volt tranny.

You know a lot better than me that a tube fails because it comes faulty from the factory, or then as soon as it is put under real working conditions or in the couple of weeks later. After that period of time they tend to stay stable for quite a long time unless you knock them hard while they are working (vibrations and such)

[Enzo]

I don't buy my tubes in bulk. I did 20 years ago, but I threw too many away. Now I buy them already in matched sets from reputable dealers - who put a warranty on the tubes they sell. They have already burnt them in in the match process. That doesn't buarantee 100% all good tubes, but I get a low fewer bad ones than I did buying them by the crate. But that is just me.

But really, what volume do you really need to handle at a time? I am serious about an old Fender Twin, or a Traynor something. Slap a set in there at a time and let them simmer while you work on other stuff. Not as elegant as a tube burner-inner, but it would check the tubes under real world conditions. Just only four at a time.

As to the the B+ requirements, you could set up a basic B+ rail with enough current for one row of tubes rather than all at once, and it could be switched row to row for a period of time. That way you wouldn't have to have a 1000 watt B+ supply.

[JC@]

I don't buy my tubes in bulk. I did 20 years ago, but I threw too many away. Now I buy them already in matched sets from reputable dealers - who put a warranty on the tubes they sell. They have already burnt them in in the match process. That doesn't buarantee 100% all good tubes, but I get a low fewer bad ones than I did buying them by the crate. But that is just me.

But really, what volume do you really need to handle at a time? I am serious about an old Fender Twin, or a Traynor something. Slap a set in there at a time and let them simmer while you work on other stuff. Not as elegant as a tube burner-inner, but it would check the tubes under real world conditions. Just only four at a time.

As to the the B+ requirements, you could set up a basic B+ rail with enough current for one row of tubes rather than all at once, and it could be switched row to row for a period of time. That way you wouldn't have to have a 1000 watt B+ supply.

Enzo I know you were talking serious that's ok.
Instead of a real tube amp, I can use my maximatcher for that matter, but I really need to put some stress to lets say twentyfive EL34, or 6L6 at the time.
Right now I am interchanging messages for an used 1000 Watt DC power supply it's a bit pricey but it's just what I am looking for...

So I am almost done with the filament supply and the B+ supply

I am a little stuck on how to wire the sockets for the bias, maybe using cathode bias. That way I could implement somekind of easy circuit braker near the cathodes in case one of the tubes shorts out. Or maybe just a fuse per tube would be easier.
I saw somewhere in the past photografs from a russian factory and they used a light bulb placed by each tube socket in their rig... Interesting indeed. No fuses necessary I guess here. Sensor and warning light all in one.

Then I would need some advice about the simplest / best way to wire a row of tubes. Just to avoid the disaster I could propitiate with a wrong way of doing it.

[Steve Conner]

Why bother buying a power supply when you can just rectify your AC line in a voltage doubler? I've built some real nasty ghetto power supplies for my Tesla coils that way. My record is 600V DC at about 7A, off a 240V line at 30A.

Of course it's unregulated and non-isolated from the line, but if you just want to stress some tubes, that may not be a problem. It's also not the sort of thing you want to go near without *both* your hands in your pockets

[bob p]

if you want to do this sort of thing with isolation from the line, think about using one of those big industrial control transformers wired up backwards. its easy enough to find single phase transformersx in the 1-3 KVA range on eBay for next to nothing. i paid $20 and $40 for transformers like this, and i've even seen them go for $0.01 plus shipping.

[bob p]

i've got to admit, i like Enzo's idea about buying matched tubes in small quantities from reputable dealers. i don't go through enough tubes to make any other sort of purchase worthwhile to me.

on the sbuject of uber-geekdom and tube matching... has anyone thought about building a tube characteristic plotter?

[Enzo]

Me, i would use fixed bias with appropriate isolation resistors at each grid.

A cheap fuse protection might be a polyswitch at each cathode. A self-resetting semiconductor fuse.

[Steve Conner]

Yes, one of those industrial transformers would be good. If it's for an audiophile application, and you're worried about it being too cost-effective, you can always paint the whole transformer with 10 bottles of C37 lacquer.

[JC@]

Me, i would use fixed bias with appropriate isolation resistors at each grid.

A cheap fuse protection might be a polyswitch at each cathode. A self-resetting semiconductor fuse.

The poliswitch is nice, but that way I will not see wich tube is failing.
If I could lift the cathodes to 3 volts instead of direct to ground, then I could use a led (when it goes off) as a warning indicator. Anyway, I am not sure if this can be done..

Instead of that I have googled a bit and found something called miniature circuit breakers (just like the one on the back of the Carvin 3212 amp), resetable, but the smallest one in value I found is 250mA. They cost about 7.50 each!

These ones are just what I would like, http://www.altechcorp.com/PDFS/TR11-Series.pdf with it's red press button and all, but again, the weakest one is 1A I believe...

[Enzo]

The failure current through the tube that set off the polyswitch should light an LED. Put the LED and series resistor in parallel with the polyswitch. There are commercail amps out there with tube failure LEDs that work just like that

SO what is wrong with a 1A? If it trips, the tube was already bad, so we are not protecting the tube. Plenty of 100 watt MArshall out there with 1A HT fuses, so there is your precedent. And if it loads down the test rig supply for a moment before blowing, so what?

[JC@]

The failure current through the tube that set off the polyswitch should light an LED. Put the LED and series resistor in parallel with the polyswitch. There are commercail amps out there with tube failure LEDs that work just like that

SO what is wrong with a 1A? If it trips, the tube was already bad, so we are not protecting the tube. Plenty of 100 watt MArshall out there with 1A HT fuses, so there is your precedent. And if it loads down the test rig supply for a moment before blowing, so what?

Yeah.. I can see myself getting closer and closer to success