I don't know why, but I've always been fascinated with the 6550 tube.
Anyway, since it's been decades since I've fooled with tube amps I am really
rusty. I'd like to build a good, clean amp around the Hammond 1650T ot and
the 6550's. I think I could use four 6550's in a ppp configuration with the
1650T ot. Any thoughts on how viable this is? Any suggestions on existing
schematics I could work from? I've looked at the Ampeg V4B and was wondering
if the 6550 could be subbed for the 7027 the V4B uses? Another question I
had about the V4B is that it doesn't utilize screen grid taps in the
transformer but resistors from the power supply. Wouldn't the screen grid
taps on the 1650T work better?

As far as power, I was looking at the Hammond 278CX xfrmr. It has 400-0-400
at 465mA. If my old math doesn't fail me that should give me about 565VDC
for the plate. Since the 1650T is 1900 ct that should be 800 on each side of
B+ which would give me a bit more than the ot can handle?

Anyway, any help would be appreciated. I don't want to get ahead of myself
making assumptions about what tube & xfrmr combo is best. I just really like
the Hammond xfrmrs and 6550 tubes!

Thanks again, Mike

take a look at the Quicksilver amps... used to build them...can't get
much better than their design...Paul

"Mike Lawson" wrote in message
...
I don't know why, but I've always been fascinated with the 6550 tube.
Anyway, since it's been decades since I've fooled with tube amps I am
really
rusty. I'd like to build a good, clean amp around the Hammond 1650T ot and
the 6550's. I think I could use four 6550's in a ppp configuration with
the
1650T ot. Any thoughts on how viable this is? Any suggestions on existing
schematics I could work from? I've looked at the Ampeg V4B and was
wondering
if the 6550 could be subbed for the 7027 the V4B uses? Another question I
had about the V4B is that it doesn't utilize screen grid taps in the
transformer but resistors from the power supply. Wouldn't the screen grid
taps on the 1650T work better?

As far as power, I was looking at the Hammond 278CX xfrmr. It has
400-0-400
at 465mA. If my old math doesn't fail me that should give me about 565VDC
for the plate. Since the 1650T is 1900 ct that should be 800 on each side
of
B+ which would give me a bit more than the ot can handle?

Anyway, any help would be appreciated. I don't want to get ahead of myself
making assumptions about what tube & xfrmr combo is best. I just really
like
the Hammond xfrmrs and 6550 tubes!

Thanks again, Mike

Hi Mike,
The 6550 in the V4B might overload the filament supply if it wasn't designed
with this tube swap in mind. The 6550 draws ~700 mA more filament current
then a 7027. Others out there probably know more about this amp than myself.
One thing to rmember about the Hammond 200 pwr tx series is the rating of
the primary at 115 v, at the more common 120+ v these days, the figure will
be closer to 415- 420 v. About swamping the ot, depends on the application.
If your building a guitar amp and the low end is much lower than say 50 Hz,
the T will probably be ok. If your looking at HiFi and response down to 20
Hz, that's different. There's the W version that's good for 280 watts at ~
double the cost. This should get you started.
Mark

On Wed, 13 Apr 2005 20:06:13 -0400, Mike Lawson wrote:

I don't know why, but I've always been fascinated with the 6550 tube.
Anyway, since it's been decades since I've fooled with tube amps I am really
rusty. I'd like to build a good, clean amp around the Hammond 1650T ot and
the 6550's. I think I could use four 6550's in a ppp configuration with the
1650T ot. Any thoughts on how viable this is? Any suggestions on existing
schematics I could work from? I've looked at the Ampeg V4B and was wondering
if the 6550 could be subbed for the 7027 the V4B uses? Another question I
had about the V4B is that it doesn't utilize screen grid taps in the
transformer but resistors from the power supply. Wouldn't the screen grid
taps on the 1650T work better?

As far as power, I was looking at the Hammond 278CX xfrmr. It has 400-0-400
at 465mA. If my old math doesn't fail me that should give me about 565VDC
for the plate. Since the 1650T is 1900 ct that should be 800 on each side of
B+ which would give me a bit more than the ot can handle?

I'm building an amp currently that uses P-P KT88s - one per side, not
parallel as you're doing with the 6550s. Ratings for the KT88 are similar
to the 6550, as you no doubt know. I'm using the 1650N and am also using
the 278CX (I'm building a stereo amp). Using 1N4007s, I see 550 VDC as
you predict. I'm getting 80W out of a pair of KT88s, more than the 1650N
is rated for (60W) Admittedly, the 1650R would make a better choice for my
amp... At clipping, I'm drawing about 270 mA out of the B+, which drops
to 510V at that point.

So assuming you get about twice the power out of a pair of 6550s, I'd
expect to be drawing about 540 mA out of the 465mA-rated power transformer,
and delivering about 160 Watts through the 120 Watt-rated 1650T.
So you'll be limited by the iron, not the tubes- but just barely.


I wouldn't worry much about the output xfmr- it sounds about right. You
probably won't be getting 160 W due to the P.S. limitation, and even if
you do, the transformer's iron will saturate at the low-frequency end only
when you're near clipping-- the "power bandwidth" will show signs of LF
rolloff. Drop the signal level by a few dB and the OP xfmr will be happy.

And although the 278CX won't quite produce enough current, it'll be pretty
close. You should wind up with a very decent amp using the lineup you
propose!

Anyway, any help would be appreciated. I don't want to get ahead of myself
making assumptions about what tube & xfrmr combo is best. I just really like
the Hammond xfrmrs and 6550 tubes!

I must say I'm fond of Hammond iron, too. You have to go a LONG way to
improve the performance just slightly. They hit the price/performance
point about right in my opinion.

I've always had a thing for KT88s, much as you do for 6550s. I really
don't know which is the better tube; they seem awfully close in looking at
specs. I'm just having the time of my life getting refreshed in all this
old technique that used to be second-nature to me. It's like a second
childhood, I suppose, after designing with solid state for over 30 years.

Anyway, HAVE FUN designing this beast- it ought to sound mighty fine!
Keep us posted on how it's coming along-- What do you plan to use for a
splitter, voltage amp, etc?

First Listen is going to be a very pleasant surprise- you're gonna love it...!



Thanks again, Mike

Keep at it!

Marty

The 1650T with its 1900R impedance is just about fine for a parallel
push-pull of 6550s in UL arrangement, fixed bias. Its max. power of 120W
means that it would be unwise to squeeze ultimate power out of these mighty
tubes. Since 500V condensers are easily available I'd keep B+ at about
450-460V while using indirectly heated rectifiers (say 5AR4, now available
from Sovtek and JJ). Note that using these tubes in UL or triode the max.
allowable V is 450V due to G2 constraints.
I'd just design the amp around this basis and I'll have a power transformer
custom wound. It is not difficult nor too expensive.

Ciao

Fabio


"Mike Lawson" ha scritto nel messaggio
...
I don't know why, but I've always been fascinated with the 6550 tube.
Anyway, since it's been decades since I've fooled with tube amps I am
really
rusty. I'd like to build a good, clean amp around the Hammond 1650T ot and
the 6550's. I think I could use four 6550's in a ppp configuration with
the
1650T ot. Any thoughts on how viable this is? Any suggestions on existing
schematics I could work from? I've looked at the Ampeg V4B and was
wondering
if the 6550 could be subbed for the 7027 the V4B uses? Another question I
had about the V4B is that it doesn't utilize screen grid taps in the
transformer but resistors from the power supply. Wouldn't the screen grid
taps on the 1650T work better?

As far as power, I was looking at the Hammond 278CX xfrmr. It has
400-0-400
at 465mA. If my old math doesn't fail me that should give me about 565VDC
for the plate. Since the 1650T is 1900 ct that should be 800 on each side
of
B+ which would give me a bit more than the ot can handle?

Anyway, any help would be appreciated. I don't want to get ahead of myself
making assumptions about what tube & xfrmr combo is best. I just really
like
the Hammond xfrmrs and 6550 tubes!

Thanks again, Mike

Fabio Berutti wrote:

The 1650T with its 1900R impedance is just about fine for a parallel
push-pull of 6550s in UL arrangement, fixed bias. Its max. power of 120W
means that it would be unwise to squeeze ultimate power out of these mighty
tubes. Since 500V condensers are easily available I'd keep B+ at about
450-460V while using indirectly heated rectifiers (say 5AR4, now available
from Sovtek and JJ). Note that using these tubes in UL or triode the max.
allowable V is 450V due to G2 constraints.

The limit for G2 supply is 450v but it only applies to beam tetrode
operation.
UL or triode operation of 6550 or KT66 allows 560v with few problems,
eg, in the mono Leak 60, which used KT88/6550 with a high Ea
to get its 65 watts class AB1 UL output.

When you have over about 38% UL taps the screen dissipation is reduced .
When the anode swings -ve, the screen also swings negative
instead of remaining at a high fixed voltage, when it attracts more
electrons.
Triode operation gives the lowest G2 dissipation, and I have seen a few amps
with
Ea = Eg2 = 600v.

But what must be remembered is that a high B+ over 500v
will usually reduce the class A part of the AB total power, unless a higher load
value
is used.

As B+ is reduced, the load value is also reduced and idle current increased.

With the 1650T, I would agree that a quad of 6550/KT88
with Ea = about 425v would be about right.
RLa-a = 1.9k, so the load seen by two tubes is
3.8k, somewhat for UL or triode if a large amount of
class A power is wanted.

Alernatively, a six pack of 6CA7/5881/6L6 would also do well
with 1.9ka-a, since each pair sees 5.7k.

More info on loading 6550/KT88 are at
http://www.turneraudio.com.au/htmlwe...50pographs.htm

Consider using Ea = 400v, and UL taps at 50%, total Pd for
each tube = 25 watts. Ia + Ig2 = 63 mA.
For each tube to work solely in class A for a pure class A PP output,
the load has to be about 5k, so the PP load for a pair is 10k a-a,
and for a quad its 5k a-a, and PO = 48 watts of pure class A,
with thd at about 1% using cathode bias which is ok becaues
the bias won't change much since its class A.

The 1.9k a-a load offered by the Hammond 1650T
is too low for class A unless one sets the OPT up
to use the 4 ohm outlet, and connect a load of 10.5 ohms.

However, when loaded with 1.9ka-a, each tube in
the quad sees 950 ohms as its load during the class B portion of its
load during each cycle, and about 80 watts class AB1 is possible from the quad,
but 27 watts is class A power.
thd at 80 watts will be about 3%.
At 10 watts average po, the average thd will be less than 1%.

With a class AB amp with such a small amount of class A %
of output power, fixed bias is the only way to keep the
bias stable unless only low power use is envisaged, and
if normal speakers are used, ie, 90dB/W/M sensitive,
this will be the case; a watt of power into each of two speakers
will drive most people out of a room.
So it pays to know just how much power one really needs.

I would only ever use the 4 ohm tap of the OPT to keep the
load value seen by the tubes as high as possible to achieve
healthy PO but the lowest thd.

But since most speakers are now 6 ohms with dips to 4 ohms,
the 4 ohm tap should be all that is used, and speakers with
a nominal Z = 8 ohms with dips to 6 ohms should also
only be used with the 4 ohm outlet.

The 1650R would be a better transformer to use imho,
since it allows 5k to 4 ohms, allowing full class A if wanted,
or class AB with a higher Ea with greater efficiency.


I'd just design the amp around this basis and I'll have a power transformer
custom wound. It is not difficult nor too expensive.

I would suggest the Hammond range of power trannies has something to offer.
But perhaps not for a stereo amp using 8 x 6550.
The heater power is 6.3v x 14.4 amps, plus more for the driver amp.

The only way I would now ever use a quad of 6550 is in a monobloc,
since a stereo amp weighs too much.

I don't know anyone needing more than 50 watts for their
modern speakers with 90 dB/W/M sensitivity.
Ad for that one only needs Ea = 500v and 2 x 6550 per channel.
See the 5050 schematic at
http://www.turneraudio.com.au/htmlwe...0ulabinteg.htm
The 1650R would be quite suitable with 5k to 4 ohms,
with only the 4 ohm match ever being used.


Patrick Turner






Ciao

Fabio

"Mike Lawson" ha scritto nel messaggio
...
I don't know why, but I've always been fascinated with the 6550 tube.
Anyway, since it's been decades since I've fooled with tube amps I am
really
rusty. I'd like to build a good, clean amp around the Hammond 1650T ot and
the 6550's. I think I could use four 6550's in a ppp configuration with
the
1650T ot. Any thoughts on how viable this is? Any suggestions on existing
schematics I could work from? I've looked at the Ampeg V4B and was
wondering
if the 6550 could be subbed for the 7027 the V4B uses? Another question I
had about the V4B is that it doesn't utilize screen grid taps in the
transformer but resistors from the power supply. Wouldn't the screen grid
taps on the 1650T work better?

As far as power, I was looking at the Hammond 278CX xfrmr. It has
400-0-400
at 465mA. If my old math doesn't fail me that should give me about 565VDC
for the plate. Since the 1650T is 1900 ct that should be 800 on each side
of
B+ which would give me a bit more than the ot can handle?

Anyway, any help would be appreciated. I don't want to get ahead of myself
making assumptions about what tube & xfrmr combo is best. I just really
like
the Hammond xfrmrs and 6550 tubes!

Thanks again, Mike