Using a Variac for starting an old tube set

[olds...@tubes.com]

I have an old tube set that I want to fire up, but know this stuff can
be unpredictable, especially with old caps. I recently bought a variac
(variable voltage transformer). But I have never used one for this
purpose.

So, I plug in the 120vac old tube device, where should I begin with the
voltages? Should I start at half voltage (60v) or lower, or go higher?

I'm asking becuse I know that the filaments will be lower, but they can
handle this. Of course the DC inside the radio will be lower too, and
wonder how this will work for the tube rectifier.... But my big concern
is the power transformer in the device. Can too low of a voltage harm
the power trans.?

So, what voltage (or percentage of power) is the best place to begin?

My biggest concern are the power electrolytics (filter caps) being
shorted, (on any old tube device).
Yes, I do intend to replace them, but I still want to see what this
device does, before I tamper with it.

In this case, I'm testing out a Hallicrafters SX-99 radio receiver. It's
supposed to work, but I dont like the idea of just plugging in somthing
thats 60 years old, and waiting for smoke..... (if there is a problem).

Thanks

[Phil Allison]

** Well, I see it uses a 5Y3 as do many old guitar amps from Fender etc .

Based on my experience with lots of them, I would have the set open on the bench so you can see and SMELL everything !

Wind up the Variac in 10 volt steps, keeping an eye on on the 5Y3 as you go, and wait a minute or two before the next step. Bad smells, flashes or glowing plates from the 5Y3 are causes to back off fast !!

Monitor the AC current draw if possible, more than say 500mA is cause for concern.


.... Phil

[Dave M]

Not a good idea to try to reform the electrolytics this way. The reason is
that the rectifier tube won't start conducting until the Variac voltage is
pretty high, and if the electrolytics are very leaky, you risk damaging the
power transformer due to excessive current.

Best way to attempt reforming the electrolytic filters is to remove the
rectifier tube, then disconnect the wiring from the electrolytics. Then,
apply a current-limited DC power supply directly to the capacitors, slowly
bringing the voltage up to the normal operating voltage. Be sure to monitor
the leakage current through the capacitors, not letting it get much above
about 10mA or so.
Reforming time could vary considerably, depending on the condition of the
capacitors. If they are too dried out, it might be impossible to reform
them, in which case, replacement is the only recourse. Sixty-year-old
electrolytics aren't likely to be good candidates for reforming.
There is likely to be more electrolytics in the unit; treat them the same
way if they are tied into the B+ supply line. Any low-voltage electrolytics
aren't likely to pose such a big risk to the transformer.

If you want to limit the risk of damage to the unit, my advice is to replace
the capacitors first, then you can be reasonably sure that you won't damage
the power transformer. The transformer is the single most expensive and
hard-to-find component in the set. By all means, protect it.

Cheers,
Dave M

[Ralph Mowery]

In article <c26t9cde4gr0r0vfc...@4ax.com>,
olds...@tubes.com says...

Forget about that variac and just put about a 100 watt lightbulb in
series with the set. Turn the set on. If the light bulb is burning
near full brightness after a few seconds, pull the plug and look for
problems. If the bulb is not very bright, make sure you have a fuse in
the set for the rted value and plug the set in.

If the set had a solid state rectifier tube, you could pull all the
tubes and use the variac to see if the capacitors were shorted or some
other problems.

[John Robertson]

Replace primary electrolytic filter caps if it hasn't been running for
many years. Running it with probably bad caps just stresses everything
starting with the expensive transformer and working down from there.

You might try a 40 or 60 watt bulb to reduce the current available.
Otherwise that is the same way we power up old jukebox equipment after
it has been recapped.

John :-#)#

--
(Please post followups or tech inquiries to the USENET newsgroup)
John's Jukes Ltd. 2343 Main St., Vancouver, BC, Canada V5T 3C9
(604)872-5757 or Fax 872-2010 (Pinballs, Jukes, Video Games)
www.flippers.com
"Old pinballers never die, they just flip out."

[Foxs Mercantile]

On 2/11/2017 9:34 AM, Dave M wrote:
> Best way to attempt reforming the electrolytic filters is to
> remove the rectifier tube, then disconnect the wiring from
> the electrolytics.

The best way to reform electrolytics is by replacing them with
new ones and not wasting your time trying to reform the old ones.


--
Jeff-1.0
wa6fwi
http://www.foxsmercantile.com

[Ralph Mowery]

In article <o7nmri$nbp$1...@gioia.aioe.org>, jda...@att.net says...

>
> On 2/11/2017 9:34 AM, Dave M wrote:
> > Best way to attempt reforming the electrolytic filters is to
> > remove the rectifier tube, then disconnect the wiring from
> > the electrolytics.
>
> The best way to reform electrolytics is by replacing them with
> new ones and not wasting your time trying to reform the old ones.

Well said. Anything very old just shotgun the electrolytics. They are
not that expensive.
You will probably save time in the long run.

[Foxs Mercantile]

I don't like call backs and I don't like working on the same radio
twice. And I certainly don't like replacing shit that got ruined
because I was too cheap to replace a $4 part.

[Benderthe.evilrobot]

<olds...@tubes.com> wrote in message
news:c26t9cde4gr0r0vfc...@4ax.com...

> I have an old tube set that I want to fire up, but know this stuff can
> be unpredictable, especially with old caps. I recently bought a variac
> (variable voltage transformer). But I have never used one for this
> purpose.
>
> So, I plug in the 120vac old tube device, where should I begin with the
> voltages? Should I start at half voltage (60v) or lower, or go higher?
>
> I'm asking becuse I know that the filaments will be lower, but they can
> handle this. Of course the DC inside the radio will be lower too, and
> wonder how this will work for the tube rectifier.... But my big concern
> is the power transformer in the device. Can too low of a voltage harm
> the power trans.?

A lot of people do it that way, but I'd worry about stripping the HT
rectifier cathode.

I'd take all the tubes out and strap temporary silicon rectifiers on the
rectifier socket.

You have to bring the voltage up very gently, but only the first 2 reservoir
electrolytics are directly in the firing line.

The lower Vf of silicon rectifiers mean you can bring the HT up slightly
higher than normal, so there shouldn't be any surprises when you re assemble
it all with the original rectifier tube.

In theory; you should leave the capacitors energised for at least 12h to re
form, but there must be some means of current limiting in case they break
down - I certainly wouldn't leave a variac hooked up.

With all the tubes out - the temporary silicon with a 40W bulb in series
with it should prevent any dangerous current flow.

[Benderthe.evilrobot]

"Ralph Mowery" <rmower...@earthlink.net> wrote in message
news:MPG.3309391e3...@news.east.earthlink.net...

A lot of people go to extremes to maintain authenticity - for some; that
means gutting the old electrolytics and hiding new modern ones in the
original cans.

It doesn't save any time, but if you do a neat job its hard to spot.

Old electrolytics that need re forming isn't a fault condition - its a
natural characteristic of aluminium electrolytics.

The electrolyte is caustic and very slowly etches away the oxide layer
dielectric. Thinner dielectric means higher capacitance, but lower breakdown
voltage.

Re forming simply grows the oxide layer back to its original thickness. The
leakage current is what's doing the work - but you have to do it slowly or
it overheats and/or breaks down.

[Foxs Mercantile]

On 2/11/2017 3:15 PM, Benderthe.evilrobot wrote:
> A lot of people go to extremes to maintain authenticity - for some;
> that means gutting the old electrolytics and hiding new modern ones
> in the original cans.
>
> It doesn't save any time, but if you do a neat job its hard to spot.

BUT....those are NEW capacitors, not old ones that have been reformed.

> Old electrolytics that need re forming isn't a fault condition - its a
> natural characteristic of aluminium electrolytics.
>
> The electrolyte is caustic and very slowly etches away the oxide layer
> dielectric. Thinner dielectric means higher capacitance, but lower
> breakdown voltage.
>
> Re forming simply grows the oxide layer back to its original thickness.
> The leakage current is what's doing the work - but you have to do it
> slowly or it overheats and/or breaks down.

Which is still a waste of time.
Once you've "reformed" the electrolytic, you're right back where you
started from. A 50-75 year old part that needs replacing.

[Phil Allison]

Dave M wrote:



>
>
> Not a good idea to try to reform the electrolytics this way. The reason is
> that the rectifier tube won't start conducting until the Variac voltage is
> pretty high, and if the electrolytics are very leaky, you risk damaging the
> power transformer due to excessive current.
>

** That is very doubtful.

Leaky electros get hot fast and will soon pour smoke.

OTOH the power tranny will survive a moderate overload for hours while a bigger overload will pop the fuse.

>
> The transformer is the single most expensive and
> hard-to-find component in the set. By all means, protect it.
>

>

** Monitoring the AC current is all you need to do.


.... Phil

[jurb...@gmail.com]

I haven't done alot of tube stufff lately, but over the years I got the idea that all this variac and dim bulb stuff was for solid state where it can totally fry in microseconds.

Tube stuff I would just plug it in. Of course watch the rectifier, and then whatever other tubes, especially power tubes in there. Anything redplates pull it out and start again. Eventually you get down to seeing how much ripple is on the B+ and go from there. Things with selenium rectifiers are of course a different story, I do not need FEMA here evacuating the neighborhood and I am sure these ninnies today would do some shit like that.

But generally, tube stuff is very forgiving.

[Benderthe.evilrobot]

<jurb...@gmail.com> wrote in message
news:3a89fcc5-e11a-4400...@googlegroups.com...

> I haven't done alot of tube stufff lately, but over the years I got the
> idea that all this variac and dim bulb stuff was for solid state where it
> can totally fry in microseconds.

Most people overlook the PTC characteristic of light bulbs that result in a
switch on surge. The few microseconds it takes the filament to heat can
destroy semiconductors if you're unlucky. Most modern stuff with an SMPSU
has a UVLO, so the variac often isn't a viable alternative.

A plain ordinary power resistor to limit fault current is safer. Tube
equipment is more forgiving - but you can still come unstuck.

[jurb...@gmail.com]

>"Most people overlook the PTC characteristic of light bulbs that result in a
switch on surge. "

That should not bother tube equipment. Possibly if it has solid state rectifiers...

With solid state stuff though that surgge save a little bit of time by letting the main filters charge up. In fact certain switching power supplies would not start without that surge.

Bottom line is that the lowly incandescent light bulb has more advantages than disadvantagges in this appplication. One is a fairly positive visual indication. For example I use the bulb to discharge the main filter caps. In fact I am looking to get one (or a few) from the UK or other country that uses 230 volts on thier mains. this because many newer SMPSes are automatic voltage select and some double the line voltage in the US, and simply don't when contronted by the higher line voltage. Sometimes the filter caps are like 400 volts, and that will eventually burn out the bulb. I have no desire to make another jig like that to have two of them in series, plus they would have to be identical as well. I think I would just rather have a British bulb.

[pf...@aol.com]

On Friday, February 10, 2017 at 11:53:25 PM UTC-5, olds...@tubes.com wrote:
> I have an old tube set that I want to fire up, but know this stuff can
> be unpredictable, especially with old caps.

I sat on my fingers until just now. This rant goes back a few years, but is as relevant now as then.

RANT WARNING:

Isolation transformers are critical for any workbench, and after good
hand-tools the first thing any radio person should get... perhaps even
before said tools.

Variacs, on the other hand, are either very handy diagnostic tools or
worse-than-useless anchors fit only to let the magic smoke out of an
otherwise salvageable radio. I am not really sure if there is an in-between
other than their original function as light dimmers.

Without both current and voltage metering, they are worse-than-
useless. With the proper metering, they can be very handy tools.

Myths: A variac will help reform caps (with specific reference to tube/
valve radios).

Fact: Only on those vanishingly few radios with solid-state rectifiers
that pass B+ current at any voltage, and only if done over more time than
one cares to consider. Bench DC supplies are best for this typically futile
purpose in any case.

In the case of a tube rectifier, most of them do not start to pass DC
until the filament voltage reaches between 65% and 75% of nameplate rating. If the B+ on the particular radio is say.... 350V, that means that the first voltage the caps-to-be-reformed would see would be something between 227V and 263V... not exactly a soft-start.

Myth: A variac will allow one to apply voltage gently to a radio, so
as to discover problems before they become fatal.

Fact: No, not at all. True, a variac does allow a radio to see the
minimum amount of current to trigger its functions... an AA5 with bad filter
caps may begin to hum at 80V rather than 120V, similarly with a transformer
radio. But if the caps are not so bad as to show audible hum, the variac ceases to be useful. And a variac will *not* tell you whether a repaired radio is OK or not.

However, if the proper meters are put on the Isovariac, then some
real diagnosis can happen: An AA5 which should, by calculation draw say...
35 watts or so (0.3A), and draws 42 watts is dissipating 7 watts of heat
somewhere... perhaps the output transformer. And, unless one has calibrated
eyeballs, the difference of 7 watts will not show up on a dim-bulb tester.

Comes down to a current meter with fine enough increments to give
meaningful information. Such a tool is useful right down the line from initial diagnosis to testing the completed results.

End Rant. And it could have been much longer.

There are various sorts of metered isovariacs out there. I keep a Heath IP 5220 (Nice write-up here: http://www.byan-roper.org/steve/steve-at-play/antique-electronics-and-2/heathkit-ip-5220-variable.html ) that has two ranges 0-1A and 0-3A. So far, nothing I have had on the bench has been beyond the 3A quiescent range - I don't do televisions, which helps. But they are also made by VIZ, B&K, and a bunch of other top-flight companies. Some even have a leakage function.

I am *not* a believer or supporter-in-concept of reforming electrolytic caps more than 40 years old. That is not suggest that 40+ year old caps cannot be good - I own many such in-circuit. But if a cap at any age does not pass ESR/Capacity then reforming is a fools' errand as it is only a matter of time. Further, that *time* will be at its most unfortunate when it arrives.

Peter Wieck
Melrose Park, PA

[Benderthe.evilrobot]

<pf...@aol.com> wrote in message
news:55d0da30-cdce-475c...@googlegroups.com...

> On Friday, February 10, 2017 at 11:53:25 PM UTC-5, olds...@tubes.com
> wrote:
>> I have an old tube set that I want to fire up, but know this stuff can
>> be unpredictable, especially with old caps.
>
> I sat on my fingers until just now. This rant goes back a few years, but
> is as relevant now as then.
>
> RANT WARNING:
>
> Isolation transformers are critical for any workbench, and after good
> hand-tools the first thing any radio person should get... perhaps even
> before said tools.
>
> Variacs, on the other hand, are either very handy diagnostic tools or
> worse-than-useless anchors fit only to let the magic smoke out of an
> otherwise salvageable radio. I am not really sure if there is an
> in-between
> other than their original function as light dimmers.

IME: The magic smoke usually comes out of the variac - they generally seem a
bit fragile.

[pf...@aol.com]

On Tuesday, February 14, 2017 at 2:31:21 PM UTC-5, Benderthe.evilrobot wrote:

> IME: The magic smoke usually comes out of the variac - they generally seem a
> bit fragile.

Never had that problem. But I tend to underfuse my devices. I could see it happening with anemic devices and/or dead-shorted loads with improper or no fusing.

[Benderthe.evilrobot]

<pf...@aol.com> wrote in message
news:27191fe3-4673-44dd...@googlegroups.com...

It doesn't take a dead short to kill a variac - and a moderate overload only
blows a fuse after a time delay.

You have to use a fast fuse and rate it considerably less than the variac
spec.

[tabb...@gmail.com]

On Monday, 13 February 2017 21:37:35 UTC, pf...@aol.com wrote:
snips

> RANT WARNING:

> Variacs, on the other hand, are either very handy diagnostic tools or
> worse-than-useless anchors fit only to let the magic smoke out of an
> otherwise salvageable radio. I am not really sure if there is an in-between
> other than their original function as light dimmers.
>
> Without both current and voltage metering, they are worse-than-
> useless. With the proper metering, they can be very handy tools.
>
> Myths: A variac will help reform caps (with specific reference to tube/
> valve radios).
>
> Fact: Only on those vanishingly few radios with solid-state rectifiers
> that pass B+ current at any voltage, and only if done over more time than
> one cares to consider. Bench DC supplies are best for this typically futile
> purpose in any case.
>
> In the case of a tube rectifier, most of them do not start to pass DC
> until the filament voltage reaches between 65% and 75% of nameplate rating. If the B+ on the particular radio is say.... 350V, that means that the first voltage the caps-to-be-reformed would see would be something between 227V and 263V... not exactly a soft-start.

That's just wrong. If reforming by running a complete set from a variac, wind the voltage up _very_ slowly until rectifier conduction just begins. At this point the rect won't pass much current, and any bad cap present will conduct, keeping the voltage on the lytics low. After a while the variac can be inched up _slightly._ Bear in mind that even a 2v increase on 240v mains increases rectifier conduction many times 1%, so very slowly does it.

It's not the ideal method but it does work.


NT

[Foxs Mercantile]

On 5/17/2017 10:39 AM, tabb...@gmail.com wrote:
> It's not the ideal method but it does work.

And it's an absolute waste of time.
Any cap that needs to be reformed is going to fail.
Just replace them damn things and get on with it.

--
Jeff-1.0
wa6fwi
http://www.foxsmercantile.com

---
This email has been checked for viruses by AVG.
http://www.avg.com

[pf...@aol.com]

> That's just wrong. If reforming by running a complete set from a variac, wind the voltage up _very_ slowly until rectifier conduction just begins. At this point the rect won't pass much current, and any bad cap present will conduct, keeping the voltage on the lytics low. After a while the variac can be inched up _slightly._ Bear in mind that even a 2v increase on 240v mains increases rectifier conduction many times 1%, so very slowly does it.
>
> It's not the ideal method but it does work.
>
>
> NT

The thing about rectifiers is that they are go/no-go devices.
The thing about Variacs is that they are not current-limiting devices but voltage limiting devices.

Which means: When the rectifier "goes" the voltage that hits the caps is a function of the voltage from the transformer secondary to the transformer primary. In a Transformer-type radio. Or the minimum trigger voltage for the rectifier on an AA5 type.

The typical tube rectifier will operate somewhere between 75% to 80% of the rated filament voltage. In some few cases, as low as 70%.

The typical tube rectifier as in those devices under discussion operates at a filament votage of 5V for transformer-types and 50V for an AA5-type set.

B+ for the typical transformer-type set is about 350 VDC. About 100 VAC for an AA5. Caps are rated (usually) at 450V and 150V respectively. "Typical", not "EVERY".

Meaning that 245V is what the caps will see in the one case, and 70V for the other.

Again, not hardly a soft start. Nor worth bothering about. Replace the poor, silly things outright and be done with it.

[tabb...@gmail.com]

On Wednesday, 17 May 2017 18:00:39 UTC+1, pf...@aol.com wrote:
NT:

> > That's just wrong. If reforming by running a complete set from a variac, wind the voltage up _very_ slowly until rectifier conduction just begins. At this point the rect won't pass much current, and any bad cap present will conduct, keeping the voltage on the lytics low. After a while the variac can be inched up _slightly._ Bear in mind that even a 2v increase on 240v mains increases rectifier conduction many times 1%, so very slowly does it.
> >
> > It's not the ideal method but it does work.

> The thing about rectifiers is that they are go/no-go devices.

wrong. bzzt.
As folks familiar with valves know that they will pass only small current when the heater is run well below rated voltage.


NT

[pf...@aol.com]

On Wednesday, May 17, 2017 at 7:23:15 PM UTC-4, tabb...@gmail.com wrote:

> As folks familiar with valves know that they will pass only small current when the heater is run well below rated voltage.

a) Why is it that so many mistake *voltage* for *current*?
b) Please explain to me how a Variac delivering 84 VAC on a transformer-type device, and that being the minimum trigger voltage on say... a 5Y3, will deliver any less current than that rectifier is able to pass at that voltage?
c) Further, explain to me how 245 VDC constitutes any sort of soft-start for the on-board capacitors.

If one happens to have a metered Variac, one can observe the onset of B+. It does NOT hover when the rectifier starts to pass DC.

[ohg...@gmail.com]

On Wednesday, May 17, 2017 at 10:49:22 PM UTC-4, pf...@aol.com wrote:

> b) Please explain to me how a Variac delivering 84 VAC on a transformer-type device, and that being the minimum trigger voltage on say... a 5Y3, will deliver any less current than that rectifier is able to pass at that voltage?

Because the filament voltage of the 5Y3 (for example) will be reduced a similar amount. Low filament voltage makes a weak tube. With low filament, the current capability of the rectifier will sag with a load causing a corresponding voltage drop.

Whether the reduced current capability of the sagging rectifier is enough to keep a leaky cap from trashing some other component is something I don't know or really care about - I replace all old paper caps if I should happen to be working on vintage stuff. I use my variac mostly for isolation but also for troubleshooting.

[tabb...@gmail.com]

On Thursday, 18 May 2017 03:49:22 UTC+1, pf...@aol.com wrote:

If you show at attempt to follow what I explained I'll gladly explain further.


NT

[tabb...@gmail.com]

On Thursday, 18 May 2017 13:43:33 UTC+1, ohg...@gmail.com wrote:
> On Wednesday, May 17, 2017 at 10:49:22 PM UTC-4, pf...@aol.com wrote:
>
> > b) Please explain to me how a Variac delivering 84 VAC on a transformer-type device, and that being the minimum trigger voltage on say... a 5Y3, will deliver any less current than that rectifier is able to pass at that voltage?
>
> Because the filament voltage of the 5Y3 (for example) will be reduced a similar amount. Low filament voltage makes a weak tube. With low filament, the current capability of the rectifier will sag with a load causing a corresponding voltage drop.

it will of course drop to the voltage where the cap no longer leaks too badly, if circuit resistances are satisfied with current.

> Whether the reduced current capability of the sagging rectifier is enough to keep a leaky cap from trashing some other component is something I don't know or really care about

Nothing is getting trashed when the rectifier only delivers 10mA.
You can only reform lytics, not paper caps. Normally lytics go across the supply, often via a dropper, and paper caps go from anode to grid.

> - I replace all old paper caps if I should happen to be working on vintage stuff. I use my variac mostly for isolation but also for troubleshooting.

Memory a bit vague but I think rectifier conduction started somewhere just over 60v for a 1937 240v set.


NT