need to wire 2 kt88/6550 in class a mode! - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 27th January 2002, 05:55 AM   #1
richt is offline richt  Puerto Rico
diyAudio Member
 
Join Date: Nov 2001
Location: puerto rico
Default need to wire 2 kt88/6550 in class a mode!

hi folks

what is the best way to run 2 6550 tubes in class a mode
(already in triode mode)

it is better with a grid leak cathode resistor or just raising the --
bias voltage??


thanks
richt
  Reply With Quote
Old 27th January 2002, 04:15 PM   #2
PassFan is offline PassFan  United States
diyAudio Member
 
Join Date: Aug 2001
Location: Central FL
How are they running now? What amplifier are they in? I don't have a big grasp of tubes but from what I've read it's not so simple sometimes. I'm sure some of the big guys will be weighing in here soon and they may need answers to the above questions.
  Reply With Quote
Old 27th January 2002, 05:11 PM   #3
diyAudio Member
 
Join Date: Feb 2001
Location: Columbia, SC
Class A isn't a question of how the gain devices are hooked up, it's whether the signal is present at all times in all devices. This devolves into a question of bias current (influenced in your case by the voltage offset between the signal grid and the cathode).
In general, it's not a good idea to simply run the tube harder without careful consideration of the circuit. The two main things to watch out for are:
1) The plate dissipation. A 6550 has a plate dissipation of about 35W (plus or minus a few depending on the version of the tube and who manufactured it). This is the amount of heat the tube can dissipate before it turns cherry red and slips sideways into the next dimension.
2) The power supply. For monetary reasons, most people don't put any more power supply into a circuit than they have to. Since biasing a circuit to class A requires much more current from the power supply, the rail may collapse.
There are other, more subtle points to consider, such as the change in load to the output transformer (not that big a problem, but be aware of it), the cathode current limits, possible changes to the feedback, etc. These aren't really limiting factors, as they can generally be fiddled, but changing the operating class of an amplifier is not something to be done lightly.
Do you have schematics for the amp? If so, begin there. Determine how close the 6550s are to their limits now before going any further.

Grey
  Reply With Quote
Old 27th January 2002, 06:21 PM   #4
PassFan is offline PassFan  United States
diyAudio Member
 
Join Date: Aug 2001
Location: Central FL
Grey:
I kinda figured a look would have to be taken at the load line of the tube. However I am not comfortable enough with tube circuits YET to give any advice. That is soon to change as I am about to delve into one of Gabevees SE designs very shortly. I have been unmoved by the printed spec for sometime now and I have to see what the fuss is all about.
From what I understand, SE tube amps love simple and efficient speaker circuits and I am considering a Lowther driver in the Medallion II cabinet. Do you have a favorite design of choice I might look at?
  Reply With Quote
Old 27th January 2002, 06:31 PM   #5
richt is offline richt  Puerto Rico
diyAudio Member
 
Join Date: Nov 2001
Location: puerto rico
thank you folks !!!

these tubes are in a mcintosh mc 275 !

the cathode from the 2 output tubes are tied to a winding
from the output transformer (local feedback) & center tapped to
earth. (class ab1) mcintosh specs.

they are biased at -57 volts ( tried -50 volts having good results.

(just for curiosity i am trying different types of designs like)

1= placed a 50 ohm resistor on the center tap of the cathodes
windings to earth (called self bias) but tubes runs cooler !
on the scope i dont see the switching distortion in this mode.

they are already running in triode mode (can be swithced to tetrode mode by a flip of a switch).


2= the most aggresive change was eliminating the negative bias voltage & intead of 50 ohm resistor i placed a 350 ohm resistor
(cathode bias) sounds good but i think that i only have 15 watts
from 75 watts & runs really hot (sounds really good but no power)

now i am playing around with negative feedback or not.


the thing is that an amplifier on tetrode mode is very efficient
but low on information (resolution) !

on triode mode is less efficient but with a lot of definition.
with no feedback you dont have the dynamics (slamming effect)
butis more detailed
i am trying to do it by a flip of a switch too ! (complicated)
because now i completly removed the first tube & making a jumper between grid & plate & a 200 kohm resitior to earh(ground)


any information will be appreciated !!!

p.s. this dosnt mean that mcintosh are bad amps .
they are great sounding amps. but me like all elct. engineer
want more !

thank you richt
  Reply With Quote
Old 27th January 2002, 06:34 PM   #6
richt is offline richt  Puerto Rico
diyAudio Member
 
Join Date: Nov 2001
Location: puerto rico
i forgot


driver / predriver/ & phase inverter are already running class a
mode



richt
  Reply With Quote
Old 27th January 2002, 10:24 PM   #7
diyAudio Member
 
Join Date: Feb 2001
Location: Columbia, SC
Skippy,
I haven't messed with SE tube amps yet. The speakers I'm using aren't efficient enough to benefit from low power like that. I have some rather vague ideas for a speaker that may or may not pan out, but it's going to be a while before I get around to that. Until then, the smaller power amps aren't going to do me much good. I'll get there.
richt,
You would go and pick one of the most complicated tube amps there ever was...
<i>Sigh.</i>
I'm at work--don't have all my schematics with me--but as I recall, the MC-275 used the arrangement that Frank came up with where the output transformer had about sebenty-lebben taps on it, going every which a way. Oi! This is not going to be easy.
Honestly, I'd recommend building something from scratch rather than tear into a perfectly good Mac. (Pixie, you wanna hear <i>good</i> McIntosh, go find a MC-275 amp in good condition [not all of them are]. That was back when McIntosh was a serious contender...and I mean <i>serious.</i> Yes, it's tubes. You'll live though the experience. Note that McIntosh is back into tube design again.)
1) Yes, if you're cathode/self biasing (resistor under the cathode) the tube in addition to fixed bias (separate DC voltage applied to the grid) you'll get less bias, not more, unless you lessen the amount of fixed bias to compensate. You'll be moving further towards class B operation.
Keep in mind that if you start running 6550s in triode mode, you're going to <i>completely</i> change the characteristics of the tube. It won't mate with the ouput transformer as efficiently.
2) You're into two different things, here. One is the value of the resistor you use to bias the tube(s), meaning whether you've kept the same bias point, gone cooler (less bias) or hotter (more). I can't tell from here. The other point is that an un-bypassed cathode resistor gives you a truck-load of negative feedback, so your gain goes down the tubes (ahem...). Takes a heap of voltage swing from the driver stage to overcome that much NFB, and when it falls short, you simply don't get the power out that you're accustomed to. Try bypassing the cathode resistor with a couple hundred uF electrolytic and see if you don't get back most of the gain that you lost.
The most important suggestion I can make is to keep an eye on your rail voltage. Make sure you don't unload the rear end so much that the rail pops up and exceeds the voltage ratings of the main caps. Conversely, don't increase the bias so much that you draw too much current from the power supply and blow the power transformer. You will not have an easy time replacing those transformers if you destroy one. If you burn up a tube, those can be replaced, but it'll be easier on your wallet if you don't exceed the ratings in the first place.
Good luck.

Grey
  Reply With Quote
Old 28th January 2002, 02:10 AM   #8
dice45 is offline dice45  Germany
diyAudio Member
 
Join Date: Jan 2002
Location: Munich, Bavaria
Hello Grey, skippy, richt,









Quote:
Originally posted by GRollins




Skippy,





I haven't messed with SE tube amps yet. The speakers I'm using aren't efficient enough to benefit from low power like that. I have some rather vague ideas for a speaker that may or may not pan out, but it's going to be a while before I get around to that. Until then, the smaller power amps aren't going to do me much good. I'll get there.



Grey,




it always is the 1st Watt of an amp that counts. With a speaker of 94dB/W/m eff. you should get tears of joy and goosebumps




Quote:





richt,





You would go and pick one of the most complicated tube amps there ever was...





<i>Sigh.</i>



Agreed, (sigh)




Quote:




1) Yes, if you're cathode/self biasing (resistor under the cathode) the tube in addition to fixed bias (separate DC voltage applied to the grid) you'll get less bias, not more, unless you lessen the amount of fixed bias to compensate. You'll be moving further towards class B operation.






Here I got lost, Grey, maybe we both are meaning the same.









If a tube runs in fixed bias mode, the cathode/filament is grounded and the grid is pulled via a grid resisitor of some 10 or 100 KOhm or a grid choke to a negative voltage (called bias) referring to ground.









If a tube runs in self-biasing mode, the grid is pulled to ground via a grid resistor (or grid choke) of about 3 to 5 times the grid resistor in fixed bias mode and the quiescent current flowing through the cathode resistor (a few 100 Ohms) lifts the cathode/filament to a positive voltage referring to ground. The bias voltage referring to the cathode however still is negative.









If the tube has a combination of both, the voltage across the cathode resistor and the fixed bias negative voltage in reference to the cathode have to be added, overlayed. What let me get lost is that "less bias" means "less negative bias voltage" to most of my buddies and me. If I add fixed bias to self-bias, I get more bias and the resulting quiescent current is reduced. You say, the tube is pushed more to class B operation, to me this means the tube's operation point sits in the knee of the transfer characteristics. Yes, Grey, that fits, agreed.









Quote:





Keep in mind that if you start running 6550s in triode mode, you're going to <i>completely</i> change the characteristics of the tube. It won't mate with the ouput transformer as efficiently.








From own experience, I would not see it as dramatic.




A pentode usually is run in near-to-short-circuit operation (load resistance is small compared to plate resistance, say, 10 to 15 %) and the load resistance has to be within close close tolerance in order to yield the maximum output power specified for a given THD: leaving this narrow optimum lewads to dramatic THD increase.









A Triode OTOH runs best with the opposite, the load being 2.3 to 3 times as big *** the tube's plate resistance. at this load the output power for a specified THD is maximum. But the maximum is not near as narrow. A 2A3 e.g. having 800 Ohm Rp is mostily used with an OPT's primary impeadance of 2-3 kOhm but also recommended with 3 or 3.5 kOhm.









Not a rule but a pleasant coincidence: most pentodes or tetrodes love a certain load impedance and happen to love the same load impedance wired as triode. With an EL34 and a KT66, I have used the same OPT for pentode and triode mode, no problems, power loss, deteriotated sonics. richt wrote his McIntosh
already is triode-wired so this would indicate the same.









I would feel comfortable to use the same trannie and just rewire the output tube. One other thing, Operation slightly moved to AB1 position should not deterioate sonics (provided there is a lil'bit of negative loop feedback) and the tubes live much longer. In a PP amp, I would not force the tubes into pure class A.




Exception: a purely differential output stage with a common CCS in the cathode.









Quote:




2) You're into two different things, here. One is the value of the resistor you use to bias the tube(s), meaning whether you've kept the same bias point, gone cooler (less bias) or hotter (more). I can't tell from here. The other point is that an un-bypassed cathode resistor gives you a truck-load of negative feedback, so your gain goes down the tubes (ahem...). Takes a heap of voltage swing from the driver stage to overcome that much NFB, and when it falls short, you simply don't get the power out that you're accustomed to. Try bypassing the cathode resistor with a couple hundred uF electrolytic and see if you don't get back most of the gain that you lost.








Technically no objection, Grey. If you tell my which electrolytic to buy and where, I mean, an electro not messing up the sonics, I kiss your feet. (and a huge foil cap is as problematic simply due to its mere size; huge caps never sing).









This cap is a coupling cap in the ground circuit. In a PP output stage with a common cathode resistor (which is not a bad idea at all, provided way enough wattage !!) it becomes less dramatical as the coupling cap is a common mode cap then, but still clearly (messy) audible.









Better: hunt more gain in the input and driver stage and live with the un-bypassed cathode resistor. Much better. Maybe get rid of the nice but gainless concertina phase splitter (dunno if the amp uses one)









Even better (as you suggested, Grey): leave the amp as it is and build a new one from the scratch. Better for the long-term blood pressure









Quote:




The most important suggestion I can make is to keep an eye on your rail voltage. Make sure you don't unload the rear end so much that the rail pops up and exceeds the voltage ratings of the main caps. Conversely, don't increase the bias so much that you draw too much current from the power supply and blow the power transformer. You will not have an easy time replacing those transformers if you destroy one. If you burn up a tube, those can be replaced, but it'll be easier on your wallet if you don't exceed the ratings in the first place.








Good advice, very good advice!









And, ... sorry for the smart-assing, I thought, if I misunderstood you, Grey, then others may join me ... and better some of this than a fried amp.









Yawn !!!, ... need some sleep
__________________
Greets,
Bernhard
  Reply With Quote
Old 28th January 2002, 03:23 AM   #9
diyAudio Member
 
Join Date: Feb 2001
Location: Columbia, SC
Bernhard is clearly trying to give me a run for my money in length of posts...
--I don't have any objection to efficient speakers, per se, just the use of horns which tends to lead to compromises that I'm not willing to make. Add that to the fact that I'm currently playing with line sources--which are rarely so efficient--and anything less than 30 to 50W won't fly for me. Yet. We'll see what happens later when I get time to play with speakers again.
--Bias, to a solid state person, nearly always means current. Bias, to a tube person, can mean either the negative voltage on the grid, relative to the cathode, or the current that flows through the tube, essentially the same way that a solid state person would look at things. Unfortunately, the two uses mean opposite things: less bias voltage (a smaller negative number, that is) at the grid means more bias current through the tube, conversely more bias voltage (a larger, more negative number) leads to less bias current. Sometimes I even confuse myself if I get to rushing headlong through these things. If I was unclear above, I apologize. If the original fixed bias (approximately negative 50-odd volts applied at the grid) was left alone, and a cathode resistor was added, then the combination of the two would lead to a net increase (a larger negative number...confused yet?) in bias voltage, which would lead to a decrease in (bias) current through the tube. It is possible to mate the two biasing schemes, but it requires a deft hand to achieve optimal results.
How's that? Did I manage to say it more clearly?
--The McIntosh MC-275 wasn't wired for triode use at the factory. It will have been modified for that somewhere along the line. My point here--triode vs. pentode--is that the McIntosh output concept, with all the extra taps coming off of the primary of the output transformer, lead to a particular output design. One I wouldn't tinker with lightly. Give me something like a Conrad Johnson (Ultralinear output) or an Audio Research (I believe they usually use straight plate-to-plate outputs), and I'll fiddle the thing until the cows come home. An old Dynaco...fire up the soldering iron, let's get wicked. But a Mac...uh unh. That amp walks a rather sharply defined knife edge, and I'd take it or leave it unless I was intending to use another output transformer. At which point, you no longer have a Mac.
--You don't need to kiss my feet. I wasn't suggesting that he live with an electrolytic bypass, just try it as an experiment to see if the gain came back. Not having a schematic handy, I don't remember how they handled the cathodes. If they're wired together (i.e. differential) then it wouldn't do much good. If the cathodes are separate, then it will make a huge difference in gain.
An electrolytic bypassed with a decent film cap is always a possibility, I suppose, but I prefer to go all film when I can afford the (admittedly) steep price.
It's more likely that folks will have an electrolytic in their junkbox than a large film cap. Solen has metalized polypropylene caps up to, I think, 330 uF (250V), somewhat smaller values at 400V or 630V. That would be the cap I would use in that application, perhaps bypassed with an MIT or Hovland or something. But...the chances of most people being able to pull one of those out of a hat are slim.
Note that using cathode bias will decrease available power anyway, since it will eat about 50 volts of the potential voltage swing. Fixed bias is a lot more efficient in that respect. If I'm remembering the front end of those amps, I don't think anyone will be able to find much in the way of unrealized gain given the design. A whole new front end would probably be required, which leads us back to the possibility of building an entirely new amp.
--Yeah, it'd be a pity to destroy a classic amp like that. There aren't that many of them hanging around. I wouldn't mind getting my hands on one...but would do no more than update the parts.

Grey
  Reply With Quote
Old 28th January 2002, 04:35 AM   #10
richt is offline richt  Puerto Rico
diyAudio Member
 
Join Date: Nov 2001
Location: puerto rico
thank you very much folks for all of your replies.

i have mcintosh amps since 4 years ago.
they are a great sounding amp & solid as a brick wall, if
it says 75 watts ,you feel like you have a 150 watts amp !

the mcintosh topology is very simple ! you have to think that
this amp was made for multipurpose use like 70 volts/
600 ohms out (for p.a. systems) & lab reference amp.

thats why you see more winding in the output transformer than usual. i consider that transformer one of the best sounding
availabe today ! that why i choose to mod mines (i have 3 of them)

remember this ++++++++++++++++++++++++++++++++

in mcintosh designes there are 3 negative feedback !!!!!!!!!!!

1 is located between the 2 cathodes of output tube
(as the same design in audio research vt-100)

2 is located between the 12 bh7 driver & the grid of the 2 ouput
tubes ,plates are fed with another winding from out transf.
it was easy to eliminate this tube (12az7) & bypass it.
this circuit when was working creates the effect that the speakers
will eat you alive !!! (some kind of slamming effect)
very good for rap & dj.s (kidding)

3rd is the global feedback (as usual in all amps) output sample feed on the first tube.
(i designes a very simple circuit that witha flip of a switch you
have negative feedback or not without altering gain)
this is useful for a delicate piece of music you want to really feel !

as i told ithe first post i placed another switch to make it triode
or tetrode.
the wierd thing is depending of the music you want to hear
you can combine these 2 switches.

for very low levels of music i prefeer triode with no feedback !

the best choice but remember that you have only 40 % of power.
(audio research from a good friend , iwired it triode mode &
no global feedback obtaining grat results !) 6 months ago with no complains.

i am happy with my amps but i want to hear it class a on the
outputs.

i dont find good info for that. i only have a lot of power amp schematics that i studied a lot

thank you folks

schematica available upon request (fax only)

richt
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
New kt88's or 6550 nhuwar Tubes / Valves 35 18th April 2009 02:02 AM
KT88 and the 6550 duderduderini Tubes / Valves 56 28th October 2008 04:47 AM
Driving the kt88/6550 in set mode duderduderini Tubes / Valves 0 15th October 2008 12:38 PM
KT88 or 6550 cjng74 Tubes / Valves 4 25th November 2007 12:33 AM


New To Site? Need Help?

All times are GMT. The time now is 12:11 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2