Hoffman Amplifiers Tube Amplifier Forum

Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: Bub on May 08, 2010, 01:42:54 pm

Title: 6550 bias in ultraliniear connection
Post by: Bub on May 08, 2010, 01:42:54 pm
Hi Gents,

I'm a little confused as to what to set my bias at. Ive got 4x Tung-Sol 6550 originals with individual bias adjust on each tube.
The Tung-Sol data sheet says 35W max plate dissipation in triode mode, no mention of UL. Another data sheet states 70W mpd in UL mode, does this sound right?. Ive installed a switch to change from triode to UL (much prefer UL mode). Plates are at 565V so I set the bias at 43mA for triode mode but I don't want to go near 86mA  for 70W mpd in UL mode. I have no idea what to set the bias at in UL mode.

Thanks in advance for any help.

Rob
Title: Re: 6550 bias in ultraliniear connection
Post by: FYL on May 08, 2010, 02:11:04 pm
Max dissipation is 42 W, recommended dissipation is 35 W per tube (design center value), or 70 W for two tubes. With a 565 V B+, you should not go above (35 / 565) = 62 mA. Using the 75% empirical rule, begin with (62 * .75) = 46.5 mA.



(Edit: corrected typo)




Title: Re: 6550 bias in ultraliniear connection
Post by: Bub on May 08, 2010, 02:15:47 pm
Thanks FYL

Rob
Title: Re: 6550 bias in ultraliniear connection
Post by: PRR on May 08, 2010, 11:29:15 pm
Stay under 42 Watts per plate; in fix-bias push-pull you can usually stay well under that figure.

> prefer UL mode). Plates are at 565V

Then screens are at 565V? GE 6550 sheet (http://www.mif.pg.gda.pl/homepages/frank/sheets/135/6/6550A.pdf) says 440Vg2 in pentode, 500V in triode, and 450V where screen is connected to tap on the plate winding.

You are "way over" the rated screen voltage.

My gut says that most good "6550"s will tolerate this for a very long time. Sunn 2000-S and Model T ran 500V on UL 6550; but higher-volt UL Sunns are marked for KT88.
Title: Re: 6550 bias in ultraliniear connection
Post by: Bub on May 09, 2010, 10:03:27 am
Thanks PRR, your gut has more info than my brain could ever absorb, but anyways.
What I've done is wire my switch so that when I'm in triode mode the screens are wired to the plate via a 1k 5W resistor.
My transformer is a Hammond 1650T with 40% ultraliear taps so when I switch to UL mode I've tied the taps to the screens through the 1k 5W resister, hope this is correct. In either mode I still have 565VDC on my screens. I'll use 35W so I should come in around 43mA. Will I be safe in this range? or should I just wire my tubes to pentode mode and be done with it.

Thanks again

Rob
Title: Re: 6550 bias in ultraliniear connection
Post by: Bub on May 09, 2010, 11:13:46 am
Thanks for the link PRR, I didn't realize the huge difference in screen gid current depending on the connection type. Please bear with me on this one. I checked the schematic where I should be hooked up to a fixed positive voltage for pentode mode which is after my choke but it's still 565VDC. So I'm wonder the best course of action if I go Pentode mode, hook my screens to the fixed voltage at 565VDC and bias lower to be nice to my tubes or derive a lower safe voltage for my screens while my plates are at 565VDC.
I would post the schematics but it is out of KOCs books so I don't no if I should do that.

Thanks

Rob
Title: Re: 6550 bias in ultraliniear connection
Post by: ernie_jr on May 09, 2010, 11:26:37 am
I have several sunn 200's that have about 550 volts on the plates ni UL mode, never had any problems with 6550's, currently have sovtek 6550we that are about 5 yrs old and still going strong
ernie
Title: Re: 6550 bias in ultraliniear connection
Post by: Bub on May 09, 2010, 11:38:45 am
Thanks Ernie jr.
I'm curious as to the difference in sound. I understand a triode connection should produce more even order harmonics where as the ultralinear connection should be cleaner (colder I guess) at a higher volume but where does the pentode connection come in. Now I want to add another switch.

Thanks

Rob
Title: Re: 6550 bias in ultraliniear connection
Post by: ernie_jr on May 09, 2010, 12:00:49 pm
for my bass use i have sunn 200s in UL and a homemade ampeg b25 with a sunn power supply working in pentode mode,
the sound pretty much the same. I am sure the HI FI guys can tell the difference, but for loud bass use, pretty much
the same sound.
ernie
Title: Re: 6550 bias in ultraliniear connection
Post by: Bub on May 09, 2010, 12:09:12 pm
That's kind of what I've done here, Ampeg B15 preamp with Traynor YBA-3 output section but 6550s instead of el34s. I can't even describe how good this thing sounds in triode or UL mode.
Title: Re: 6550 bias in ultraliniear connection
Post by: PRR on May 09, 2010, 08:05:46 pm
> for pentode mode which is after my choke but it's still 565VDC.

That was always "in excess of specified MAX".

Ernie says it works for him, and I believe it.

Do keep ~~1K resistors in series with the screens, all modes. They won't have much effect in normal operation, but allow a little "give" in abnormal conditions.

> I understand a triode connection should produce more even order harmonics

It's a push-pull amp. All even-order distortion essentially cancels.

Triode will damp speaker-flap, but won't make a lot of power.

Pentode will make a lot of power, but does not damp the speaker.

UL is the in-between. It has some damping action on the speaker, and nearly as much power as pentode. And some other benefits not real useful for a stage-amp.

The classic Rock Guitar sound is NOT highly damped. Many early amps were naked pentodes, and speaker-flap was selected to complement the dry tone of a raw steel string.

The classic 1960s Fenders had "light" speaker damping via a NFB loop from speaker to driver. The early Marshalls were similar but with more damping. This difference may not have been so much philosophical as practical: some speakers "need" more damping than others, especially at high power. And there is always a "need" to have a "New for 1962!!" sound... speaker selection and damping factor can give an amp which sounds different (and perhaps "better") than your sloppy old 1961 amp. Like tailfins on a 1956 Plymouth. Gotta churn the customers and make a profit every year.

Bass amps use less speaker coloration, and often need good damping to control speaker slap.

> how good this thing sounds in triode or UL mode.

It seems like your speaker and style "like" damping. At this power level, that's very reasonable: 150+ Watt shocks will make a cone quiver for days if the amp does not suck-out the excess energy flapping around. There's other ways to do this, but if you are liking UL then go with what works.
Title: Re: 6550 bias in ultraliniear connection
Post by: Bub on May 09, 2010, 08:51:58 pm
Thank you for the explanation PRR, that fills in alot of the gaps. When I read about a pentode to triode switch they refered to it as a half power switch but I thought you got more power in triode mode so I skipped pentode all together. I'm still concerned about my screens so I will bias at 40mA for now and maybe hook my switch to work from pentode to UL just for something to do. I need to go back read about power supplies. I would like to get my screens down to an exceptable level even though it may not cause any damage.

Thanks

Rob