1st powerup bias question plz..

[GrooveGrinder]

Hi folks....

For a first time powerup of a ground up rebuild..
Everything new except for iron/chassis,
Stock circuit/replica board,
JJ E34L ...not EL34 (the hot ones)...

Where should the bias pots be..
Full clockwise, full counterclockwise, or centered??
(and why plz?)

Thanks in advance!!!

Greg

[Bob01605]

Groove,

If you rotate those controls fully counterclockwise you will be at the lowest setting when you power up. The bias will be low - but you won't hurt anything. Then turn each one up to the correct bias > 1.56 volts. You will have to go back and forth between the two channels because adjusting the left channel has a slight effect on the right channel.

Dynaco biased their EL34 tubes a little on the high side. 1.56 volts gives 100 milliamps per tube pair or 50 milliamps per tube. IMHO 40-45 milliamps per tube will give longer tube life and sound the same.

To bias a stock Dynaco ST-70 at 40 milliamps per tube = use about 1.25 volts bias voltage per tube pair.
To bias a stock Dynaco ST-70 at 45 milliamps per tube = use about 1.40 volts bias voltage per tube pair.

Bob Latino

[Bob01605]

On initial start up what I usually do is first plug in just the 7199 tubes and leave the rectifier and the EL34 tubes out. Turn the amp on. Do the 7199's light up ? If they do you know your 6.3 volt AC line from the transformers are OK. If they don't or you blow a fuse > You better go back and check your wiring ..

If the 7199's light up OK, stay on and don't blow the fuse then I would plug in the GZ34 rectifier and the EL34's in one channel and bias that channel - Shut the amp off - plug in the remaining EL34 pair - turn the amp back on and bias the other channel. Go back and forth between the two channels until the bias is correct on both. If the bias seems to jump around a lot then connect the speakers. I have seen some amps that would not bias correctly without the speakers applying a "load" on the circuit.

My comments above about biasing a little lower than Dynaco recommends goes back to the Mark IV. When Dynaco came out with the Mark IV in 1960, the year after the introduction of the ST-70, they used a 13.5 ohm bias resistor. In conjunction with the 1.56 bias voltage recommendation that gave 115 milliamps per tube pair or almost 58 milliamps per tube. This is really on the "hot" side for an EL34 and almost assuredly cut down on ouput tube life in these amps. Dynaco got away with it primarily because of the great sounding and extremely durable Mullard output tubes that they supplied. Mark IV users IMHO would do better to replace that 13.5 ohm resistor with the 15.6 ohm bias resistor from the ST-70. Dynakitparts, by the way, stocks new bias resistors for all Dynaco power amps.

Bob Latino

[GrooveGrinder]

Hi Bob..

Thanks for your reply!

The procedure you mentioned is very helpful!
I'll do just that.
Very interesting history re: mkIII, etc.
I love knowing those little tidbits!

You will have to go back and forth
between the two channels because adjusting
the left channel has a slight effect on the
right channel.

Oh how I remember that from the days of using
a couple of stock 70s.
Back when they were already seriously deteriorating,
before eventually retiring them
to sit around waiting for the rebuild I'm just finishing.
With sooo much going wrong with them, the back
and forth process took a looooong time...
Bias supply going... tube sockets in such poor
shape you couldn't tell whether it WAS the bias supply..
Can cap iffy at best..
And that very pretty sparkling in the rectifyer........
<sigh> such nostalgia..........

If you rotate those controls fully counterclockwise
you will be at the lowest setting when you
power up. The bias will be low -

Ok... let me try to get this straight here..
Usually I can dope these things out fairly easily,
but I got so much going on right now, that brain
fog rules the day...
"lowest setting" = less negative bias against the
tubes "running away", right?
Wouldn't you then be better off with them fully clockwise
then?
------------------

I've another question for you, if you'd be so kind..
The tubes I'm using are the JJ *E-34-L* not
EL34.
I mention this, both as part of my initial question,
and also in regard to your biasing recommendations.

When reading what others said about them around
the net, I noticed that many
(at least by those who mess with this sort of thing)
felt that they sound best biased "hotter" than
normal for EL34s.

Most suggest 1.65 or even 1.7v.
They say these tubes just don't sound great at
1.56v.
Dull, cold, are the terms used....

This copied from the tubedepot details page
for the JJ E34L:
"With a slightly higher grid voltage than other EL34
(-13.5 to -16.5 vs -10 to -13.5 volts for the EL34)
this tube offers tight low end with smooth mids."

(The last line is of course, irrelevant.)
Perhaps they mean higher grid voltage spec, or
draw, requirement? capability?
I'm a bit lost...
I assume its draw... hence more bias?

I want nice long tube life like the next guy, but
I'm not sure whats best for the E34Ls, including
the sound quality consideration.
Maybe I should start out at the "official" 1.56v spec?

Btw, triode has 15ohm resistors too for $1.15.
(which I used, btw)
Here:http://store.triodestore.com/15ohm2wcf.html
They're big.. a pretty red... look like nice quality,
and are made in Spain :-)


Thanks so much for your help...
Can't wait to get 'er running!

Greg

[GrooveGrinder]

While I'm at it...

Do you have a fuse recommendation?

A lower value for use during first powerup..
Also, perhaps a slightly lower rating for normal
running?

Maybe 2A for pwr-up and 2.5A for normal use?

Thanks again!

Greg

[Shannon Parks]

Greg, you could just use a fast acting fuse for testing and a slow blo for everyday use.

[EWBrown]

Last year, I found a box of five hundred 6.8K, 1W, 5% resistors of that same Spanish make (Piher) and type at the local town dump. Just sitting on the swap table. Free is ALWAYS good One person's trash is another person's treasure.

(I've used a few of these in my Grommes LJ restoration (with a better LTPI curciut and no tone stack).

I've left my share of goodies there as well, I still wonder just what happend to those two Codex DES ATM encryptors that I "contributed". Yellow_Light_Colorz_PDT_20

Just what do I do with 500 resistors of the same value... Just think series-parallel combinations. Yellow_Light_Colorz_PDT_03

/ed B in NH

[kheper]

This copied from the tubedepot details page
for the JJ E34L: "With a slightly higher grid
voltage than other EL34 (-13.5 to -16.5 vs
-10 to -13.5 volts for the EL34) this tube
offers tight low end with smooth mids."

See -

http://www.jj-electronic.com/pdf/E34L.pdf

What they are referring to is that the typical
operation of the EL34-L's g1 voltage is
usually set higher than the EL34's. You do
not need to run it there.

Setting the bias higher makes the entire tube,
not just the grid, run hotter than it would.

I use the JJ EL34-Ls. They plenty run hot at
50ma per tube. I leave it at that.

[GrooveGrinder]

Greg, you could just use a fast acting fuse for testing and a slow blo for everyday use.

Hi Shannon!
Thanks for the suggestion..
What values do you suggest?

I remember reading a thread on the subject a
long while back...
Couldn't find it again though..
It contained an analysis of ST-70 draw, vis a vis
the Dynaco rating of 3A, and found it to high,
based on industry norms...
Hence it wouldn't provide best protection for the amp,
but more margin than needed for draw..

I think those (much more expert than I) in that
thread suggested 2.5A for running and 2A for testing.

I'd ge greatful for further thoughts!
(especially since I gotta go buy some fuses today
to get up and running)

G

[EWBrown]

When you're first starting out, even before installing any tubes, set the two bias pots for the most negative voltage levels (at the bias test points) , as the lower the voltage (meaning the most negative) , the lower the tubes' plate / cathode currents will be upon startup . That way, you won't be inadvertantly smalling teh tubes with too much bias current.

The bias negative voltage range is somewhat limited by the three-resistors voltage divider, and you can't set the voltage as being "too negative" to start out.

With the original stock Dynaco 15.6 ohm resistors, the idea was to get a reading of 1.56 volts for each pair of 6CA7 / EL34 / E34L, or 50 ma for each tube of the pair. Similarly, the MKIII used a 11.2 ohm resistor, to get 70 mA per tube, or 140 mA for the pair of 6550s or KT88s. THis would also result in a 1.56 DC volt reading at the bias test jack when everything was set correctly.

The reason behind that seemingly "odd" voltage reading, is that it was the normal DC voltage of a fresh carbon-zinc D cell battery, and in the days before digital volt meters, it was the practice to read a fresh D cell, mark the (old ) meter's needle position (with crayon, tape, etc) and then set the bias pot fdor a matching indication, so that any old "dumb" meter was good enough, in that it was being used to compare two voltages rather than read the actual voltage to 3 places. Most modern designs (or upgrades) it is best to use 10 ohms as this eliminates any math errors, and can give a "true" current reading when used with a small LED or LCD panel-mount DVM display unit.

All of this, of course, presumes that the pairs of tubes are reasonably well matched, the original supplied ones usually were, allegedly, though I remember seeing some departures from what would be considered as being close enough...

WIth the new 15 ohm resistors , measure them first and determine their exact value) then proceed accordingly.

The easy / lazy (like I prefer) method is to use 10 ohm, 1% resistors, 1/2 or 1 watt is sufficient, and then set for 1.00 volts to get 50 mA per tube.
Assumint they are "exactly" 15.0 ohms, then set for 1.50 volts per pair.

1/2 watt is sufficient, as the actual current is 100 mA< 10 ohms generates 1.00VDC across teh resistor, for a total of 100 milliwatts of dissipation, so a giant power resistor isn't needed. BEsides if the worst should happen and a tube arc over or internally short out, the 1/2 or 1W resistor will act as a fuse and prevent any Chernobyl-style "meltdowns" or glowing red plates...

If you have one of the new Triode Electronics PA060s, they have more current capacity, and the tubes' cathode currents could be set higher if desired, to perhaps 55 or 60 mA (110-120 mA for the pair). This may produce better sound, but will age the tubes a bit faster. 6550s and KT88s can go up to 70 ma each, but then this is coming very close to hitting the limit of the transformer's capabilities... Don't do this witn older / vintage PA060s, they were already "hitting the wall" at 200 mA (plus a few more mA for the driver board).

HTH

/ed B in NH

[GrooveGrinder]

See -

http://www.jj-electronic.com/pdf/E34L.pdf

What they are referring to is that the typical
operation of the EL34-L's g1 voltage is
usually set higher than the EL34's. You do
not need to run it there.

Setting the bias higher makes the entire tube,
not just the grid, run hotter than it would.

I use the JJ EL34-Ls. They plenty run hot at
50ma per tube. I leave it at that.

Thanks for the great info!
Very helpful..
I think I'll do just that... stay with stock bias setting.

How do you like the JJs?
Like the sound?
Have you tried others?

G

[EWBrown]

I've used both J/J EL34s and E34Ls, and have changed the tubes in one ST-70 to the J/J KT-77s from Eurotubes. These are also very nice, and do require a minor bias readjustment to reach 50 mA cathode current for each (I use the VTA-70 12AT7 based driver board in this one and it has four separate bias setting pots).


Some of the early release J/J KT-77s had rather skinny pins and would be a loose fit, and a poor contact in older, used sockets. The current crop of KT-77s have the normal diameter pins, and this is no longer a problem.

The skinny pins work OK with new "virgin" octals socket, preferrably those with full-surround contacts.


/ed B in NH

[kheper]

How do you like the JJs?

A tube to remember...

As Alex says, 'Real horrorshow'.

http://soomka.com/nadsat.html

Like the sound?
Have you tried others?

They are a good tube for the money. Other
than certain NOS tubes, they cannot be
beaten.

[erichayes]

A couple of comments, FWTW . . .

Going back to Jr. High math, there was this little, but very useful term called the "absolute Value" of a number, and it was always positive ( |-57|=57, |23|=23). If you think of the bias voltage in terms of its absolute value, or magnitude, you won't get confused with semantics such as "higher", "greater", "lower", etc. Thus, if I tell you to increase the bias voltage to decrease current flow, what I'm really saying is "increase the amount of bias voltage". Most amp manufacturers wire their bias pots so that, as the pot is turned CW, the amount of bias voltage actually decreases, which causes the cathode current to increase.

Before I switched from EL34s to KT88s, I found the door-opening point for (JJs, at least) their idle current was 60 mA/tube. Since I'd designed the amps to run with KT88s, this wasn't a problem as far as the power supply was concerned--I don't know if I'd want to try it using original Dynaco iron. At 60 mils, the '34s became different tubes and sounded like I'd hooked up a better pair of speakers.

[kheper]

If you think of the bias voltage in terms of its absolute value, or magnitude, you won't get confused with semantics such as "higher", "greater", "lower", etc. Thus, if I tell you to increase the bias voltage to decrease current flow, what I'm really saying is "increase the amount of bias voltage". Most amp manufacturers wire their bias pots so that, as the pot is turned CW, the amount of bias voltage actually decreases, which causes the cathode current to increase.

Before I switched from EL34s to KT88s, I found the door-opening point for (JJs, at least) their idle current was 60 mA/tube. Since I'd designed the amps to run with KT88s, this wasn't a problem as far as the power supply was concerned--I don't know if I'd want to try it using original Dynaco iron. At 60 mils, the '34s became different tubes and sounded like I'd hooked up a better pair of speakers.

That precisely explains the specs of the
EL-34-L tubes (not just the JJs) and the
'subjective interpretation' of the tube when
they are biased at a higher current.

However, I cannot get my stock st-70 bias
circuit to do more than -44VDC. 'Upping'
this can easily be done, but may not be a
good idea with an old PA-060.

I picked up two transformers on e-bay, they
are rated 80VAC@150ma. These will make
nice bias circuits to try 60ma on the
EL-34-Ls in some future projects. There
remains the heat issue; EL-34s run too hot
for my liking.