Re: IGNORE this DAMN FOOL



AJ wrote:

In article <98ce1feecfb8f1a8284862d6d044fbd6
@localhost.talkaboutaudio.com>, bratzirules@xxxxxx says...
Setting bias on any amplifier by reading the total power draw of the mains
is ***-nutty.

You need a resistor from ground to cathode to serve as a shunt, unless
you have a very accurate DC clamp on milliammeter and can get it 'round a
opt primary winding lead. (No one does, or hardly anyone.)

That said, if you aren't very experienced, whatever the manufacturer
charges for a schematic with factory recommended bias set procedure is
money very well spent. The unit's owner should ante up.

That ALSO said when I fixed guitar amps I worked in a shop that had one
of the really cool LabVolt panels that had a true wattmeter, voltammeter,
a power factor meter and voltage and current metering-completely redundant
except to show students in electrical shop class the relationships. It used
the big Westinghouse gauges used in utility power house panels. We got to
where we could pretty successfully instantly diagnose certain problems on
certain pieces we saw all the time, like the shorted windings on Peavey
CS400 power amp power transformers in a certain range of production. That
isn't technicianship though. Technicianship is learning to systematically
go from principle to fault finding in a systematic way.

--
Message posted using http://www.talkaboutaudio.com/group/rec.audio.tubes/
More information at http://www.talkaboutaudio.com/faq.html


It could be how they set them up at the factory on a regulated supply.
I've being biasing my guitar amps(push/pull, grounded cathode, fixed
bias class AB), by first measuring the resistance of the transformer
secondaries(unplugged, power off, caps discharged). Then with the amp
warmed up for 10-15 minutes with no signal, measuring the plate voltage
and the voltage drop(to 2 decimal places) from B+ to a plate terminal on
each side of the circuit. The rest is an exercise on a scientific
calculator. voltage drop/resistance = current in amps * plate voltage =
dissipation in watts / max dissipation for the tube type * 100 = idle
percent of max. Someone posted 66% and that works for me. For 4 output
tube amps divide the current by 2, but this won't show if each is
pulling equally. It also doesn't include screen current. Schematics that
don't show an idle current setting, often show B+ and plate voltage so
some idea of what the designed had in mind is there.

When guitar amps come to me for a fix the first thing I do is add
10ohms, 5 watt wire wound R between each cathode and 0V.

This allows safe measurement of the total anode and screen currents. I
check screen current at the series screen R,
and usually for two tubes Ig2 are within +/-20% of each other, and as
the Ig2 is less than 10% of Ia, then it can be neglected and the bias
set so the V across the 10 ohms is equal.

In an Ampeg SVT I am working on now, they have the 3 cathodes of 6550 on
each side of the PP circuit tied together and taken to a 1 ohm R for
reading the bias current on each side of the PP circuit. Its unreliable
because you don't know what the bias current difference is in each 6550.
So I abolished that damn 1 ohms, and have added 6 x 10 ohms, 1 for each
cathode, and made 6 test points at the rear of the chassis so all six
bias currents can be read off. Suppose the tube are a bit old. They will
all have different bias cathode currents by a maybe +/- 30% of what they
should be. The Ampeg POS doesn't have 6 adjust pots like it should. The
POS has all 3 grids on each side direct coupled to a cathode follower
cathode and the follower grid is biased. So Ampeg can only have one
adjust pot for each set of 3 tubes each side of the PP circuit. Just why
they have a CF driver is a mystery, because the 6550 grids each have a
47k between the grid and driver cathode on the follower. The 47k
prevents grid current in class AB2 working which isn't safe for any amp
in the hands of destructive "musicians". Ampeg coud have omitted the
follower buffer driver, and had 6 bias pots with 6 R&C couplings from
the anodes of a normal LTP with 12BH7.

So how does one bias the amp now there are going to be 6 different bias
current readings? Well, there isn't a great deal of choice
because the tube with the highest bias is the one that will determine
the setting. With Ea = 600V the Pda doesn't want to be more than 1/2 the
maximum Pda rating, say 20W, so max Ia should be no more than 33mA, so
each 10 ohm test point should real no more than 0.33V
What if one tube measures 0.55V, another is 0.33V, and another is 0.22V?
There is 110mA total.
But that one high Ia tube is idling at Pda = 33watts, and if ever the
amp is overloaded with a speaker load being too low, guess which tube
will over heat with a red plate first? It will be the one that is idling
the hottest. I'd be dropping the bias down for all 3 tubes so that bias
would become 33mA, 16mA, 10mA, and this would mean slightly more THD but
who cares?, this is a guitar amp where THD doesn't matter. Suppose there
is a similar range of bias currents on the other 3 x 6550. OK, set the
bias so none are above 33mA. The totals can be added and usually,
although differences exist, the totals for each side will be about the
same while the tubes are still serviceable. Once they go beyond the
point where one or two have Ik which is 100% higher than the those with
least Ik, then its time for tube replacement. Its a great pity that
Ampeg and many other dumbasss designers decided to make sure amp users
couldn't bias their tubes correctly by adjusting bias for each tube. The
tube life would be extended to what it should be when each tube runs
with the same temperature. The SVT is supposed to make 300 watts. Each
pair of 6550 is expected to make 100W, and in fact with Ea about 700V,
its quite possible. The replacement Power tranny I am fitting can only
make 600V, but if there is 240W, I will be happy, and won't have to
service the amp again too soon. With multiple output tubes, there is
always a high possibility that bias Ia will be very different in older
tubes when there is only 2 grid voltage adjustemnts for 6 output tubes.
Anyway, the way I have altered the SVT will allow the owner to check the
bias very easily on each output tube and perhaps he will see trouble
coming before the darn thing blows up yet again. The one I have for
repair was something someone in the US was glad to get rid of and was
just shipped to Oz without the PT. The OPT has been recently replaced,
and I wouldn't mind betting that after replacing the OPT, the amp didn't
last long and began humming and smoking because of troubles I found with
most of the electros. So they cut their losses and sold it.
Maybe the PT they removed was fried; I am not to know, they don't send
the amp with a US 110V PT because it adds to the freight cost. If the PT
is good, its a spare that's useful. Its not wanted here in Oz because
there isn't any way to change the mains voltage for 240V.

Patrick Turner.
.

Quantcast