DCAudioDIY.com

DC Area Audio DIYer's Community
It is currently March 28th, 2024, 6:41 am

All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 44 posts ]  Go to page Previous  1, 2, 3, 4, 5  Next
Author Message
 Post subject: Re: PP KT88 Amps
PostPosted: September 6th, 2020, 4:01 pm 
Offline

Joined: June 4th, 2013, 2:39 pm
Posts: 487
SoundMods wrote:
Roscoe Primrose wrote:
David McGown wrote:
with an additional feedback loop and a weird resistor bypassing the coupling cap to each output tube

With that resistor bypassing the coupling cap, are you sure you actually had a negative voltage on the grid?

Roscoe


Excellent question! Me thinks the builder mistook a grid stopping resistor as a resistor installed across the coupling cap. That would place +DC on the power tube's grid. Not good. :o

The tube would pull full current until self-destruction. :shock: Either fixed bias on the power tube grid of say -50-vdc (a S.W.A.G.) or cathode bias.

But under no circumstances bypass the inter-stage coupling cap.


Well, with all due respect, this circuit actually works as designed. The coupling cap bypass resistor improves LF stability by, if I understand it correctly, adding some local feedback between the driver and output stages. But in order for it to work, you have to construct the bias circuit carefully so that the positive voltage sent to the grid of the output tube is countered by the negative voltage supply. Without that resistor, there's way too much negative voltage and the tubes won't bias. It's very cleverly done. The driver stage has also been altered from the original Williamson to add some degenerative feedback in the driver stage's cathode circuit. Here's the original article that explains how the circuit works:

http://www.technicalaudio.com/pdf/Gordo ... n_6550.pdf

In the Acro catalog, the circuit is duplicated except for the addition of an adjustable pot in the bias supply to allow for an option to use either EL34s or KT88s.

ETA: I can also attest that, having gotten one monoblock up and running, the amp is dead quiet and sounds very nice. It compares well to my "original" 807 Williamsons, with a crisper sound and about double the power. I'm eager to hear it in stereo. ;-)


Top
 Profile  
 
 Post subject: Re: PP KT88 Amps
PostPosted: September 6th, 2020, 4:55 pm 
Offline

Joined: February 28th, 2013, 1:19 pm
Posts: 914
Grover,

It is heartening to hear that you were able get the circuit to work, and it was your success that I had my "Ah Ha" moment and realized what I did wrong.

It is all about current from the bias supply. For there to be proper voltage at the grid, the current through the resistors bypassing the coupling caps needs to be the same current as sunk through the bias supply. You want a large value resistor across the caps, and a smaller one in the bias supply and a suitably negative supply to start with, since it acts as a voltage divider ensuring the proper negative bias voltage at the grid. The negative bias voltage and resistor on the bias supply is critical for the circuit to work. Putting in an adjustable bias pot (not the balance pot) is a recipe for disaster, this needs to be a fixed value resistor to ensure bias stability, so the voltage divider provides the correct negative voltage at the grid. Also means you cannot bias this amp unless the driver stage is powered.


David


Top
 Profile  
 
 Post subject: Re: PP KT88 Amps
PostPosted: September 6th, 2020, 5:28 pm 
Offline

Joined: February 28th, 2013, 1:19 pm
Posts: 914
I was not so clear in my last post.

Lets assume the voltage at the plate of the 6SN7 is 210 volts. And the negative voltage at the grid of the tube is design as -45V. This means the current through the 470000 ohm resistor is 255V/470000 = 0.54 mA. Then we consider that the gridleak resistor of the output tube being 100000 ohms and a voltage of -45V means 0.45mA flowing from the ground to the grid node. This gives about 1 mA of current that needs to be sunk into the negative supply from each tube, or 2 mA at the wiper of the balance pot.

The resistance of the balance pot (if perfectly balanced) is 5000 ohm, so the voltage drop at the wiper is 0.002 x 5000 = 10V, so the supply needs to be -55 at this point. With an 18000 ohm resistor in the bias supply, the voltage drop is 0.002 x 18000 = 36V, so the DC voltage at the first cap of the bias supply needs to be -91V.

Change the values of anything, and it comes crashing down unless it is carefully engineered for DC balance. It should work, but not the way I was trying to do, that is for sure.

David


Top
 Profile  
 
 Post subject: Re: PP KT88 Amps
PostPosted: September 6th, 2020, 6:09 pm 
Offline

Joined: February 28th, 2013, 1:19 pm
Posts: 914
I screwed up my analysis, forgetting R19 and R20. This complicates matters a little.

This means we have the current from the grid to the negative supply running through the 47000 ohm grid leak (R19), i.e. 0.54mA x 47000 ohm = 25.4V. Which means the voltage at one end of the pot (assuming a -45V bias voltage) is -70V. The current from ground to the end of the pot (R22) is 70/100000 = 0.7mA. This means the current flowing towards the negative bias supply from each tube is 1.24mA, or say 2.5 mA total. Voltage drop across the 10K balance pot (if perfectly balanced) is 5000 x 0.000124 = 6.2V on each leg. This means -76.2V at the supply. Voltage drop across the 18K resistor (R29) is 18000 x 0.0025 = 45V. This means the bias supply needs to be -121 Volts at the first capacitor past the rectifier. I had nothing close to that.

David


Last edited by David McGown on September 6th, 2020, 6:14 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: PP KT88 Amps
PostPosted: September 6th, 2020, 6:13 pm 
Offline

Joined: June 4th, 2013, 2:39 pm
Posts: 487
Yes, I agree, it all has to balanced carefully. I'm not sure why an adjustable resistor in the bias supply is a liability, though. Keroes points out that you may need to adjust his bias-supply resistor to ensure the proper voltage at the KT88 grids. The Acro schematic simply allows for this, and for swapping in EL34s if you desire. I had to fiddle a bit further, adding some resistance in series with the pot, to allow for excess bias voltage caused by the reverse tranny I used for the supply. But right now I've got about -36 volts on the EL34 grids, and they are running at about 65mA each. I think that's about right. It's definitely the case that you can't adjust the bias without the full circuit on, and that's tricky, but it's not terribly hard if you start with more negative voltage and ease it down to where you need it.

And the fact is that if you swap output tubes, even different EL34s, and then change out the driver 6SN7 for one that runs a bit hotter or cooler, everything's going to change a bit, so the ability to make an adjustment seems pretty key to me.


Top
 Profile  
 
 Post subject: Re: PP KT88 Amps
PostPosted: September 6th, 2020, 6:20 pm 
Offline

Joined: February 28th, 2013, 1:19 pm
Posts: 914
Grover,

I think the adjustability has to be designed to take into account those variables, but not so loose as to result in running the tubes too hot. I never had a bias voltage that was negative enough for it to have a chance at working, and did not understand what I was doing on the bias supply. You were right, build it as it was designed to start with.

David


Top
 Profile  
 
 Post subject: Re: PP KT88 Amps
PostPosted: September 6th, 2020, 6:26 pm 
Offline

Joined: February 28th, 2013, 1:19 pm
Posts: 914
Grover,

Perhaps my punishment for this is to build a proper Williamson :lol:

Seriously, I am keeping with the Mullard for now, enjoying the amps too much to touch them.

David


Top
 Profile  
 
 Post subject: Re: PP KT88 Amps
PostPosted: September 6th, 2020, 9:34 pm 
Offline

Joined: June 4th, 2013, 2:39 pm
Posts: 487
I'm sure the 6550 amp sounds lovely. I'd love to hear it....sometime!

Jim has a pair of my 807/KT66 "Musician's Amplifiers" over at Dewey Beach and I'm hoping that when this Covid nonsense is over he can bring them to a meet so everyone can hear them.

Truth be told I threw these amps on a pair of Heathkit W4 chassis, . I'm getting lazy (and disrepsectful) in my old age. ;-) The power tranny is just adequate for the EL34s. If I want to crank up the KT88s I've got to swap in a bigger power tranny and different choke. But they sound pretty nice for now.


Top
 Profile  
 
 Post subject: Re: PP KT88 Amps
PostPosted: September 6th, 2020, 9:40 pm 
Offline

Joined: January 14th, 2015, 11:15 pm
Posts: 499
With so many pp amp designs, what is special about the Williamson design? Is there some special magic or is it a rite of passage?

Just asking, no axe to grind or anything.

ray


Top
 Profile  
 
 Post subject: Re: PP KT88 Amps
PostPosted: September 6th, 2020, 11:14 pm 
Offline

Joined: June 4th, 2013, 2:39 pm
Posts: 487
ratbagp wrote:
With so many pp amp designs, what is special about the Williamson design? Is there some special magic or is it a rite of passage?

Just asking, no axe to grind or anything.

ray


You might get a lot of different answers on this (or not!) but to my ears, the beauty of a true Williamson amp is the sheer naturalness of tone. There is nothing "hi-fi" about it--there's no artificial brightness or edge to the sound, no weird phasiness, no incoherency. Instruments and voices sound utterly realistic and honest. The original was an "all triode" design, using 6J5 triodes for the voltage amp, phase splitter and driver stages, then KT66s in triode mode. The triode-wired KT66 was considered a significant upgrade from 2A3s or 300Bs in that you got more power, more transconductance, along with the convenience of indirectly heated power tubes, which meant *much* less noise and a vastly simplified power supply--one filament supply could serve for *all* the tubes.

The other huge leap was *bandwidth*. The Williamson was the first amplifier to employ global negative feedback. In the late 1940's a typical 15 watt push-pull triode power amp using 2A3s (300Bs were rarely used outside industrial theater amps) could reasonably produce a frequency range on the order of 100Hz to 12kHz, or 15kHz if it was a quality piece. The Williamson boasted "flat" response of 20Hz to 200kHz!! This meant that transients were reproduced with astonishing faithfulness. It also employed a very high load on the output tubes--10K plate-to-plate--to achieve incredibly low distortion.

But there was a catch. The Williamson has a daunting number of frequency poles, all threatening to send the amp into oscillation, and the output transformer had to be capable of enormous bandwidth to accomodate the amount of feedback employed without introducing deadly phase shifts at the output. You *cannot* build a Williamson with any old output transformer. In England, the Partridge was built to Dr. Williamson's specifications. In America, the Peerless S-265-Q, the Stancor A-8054 and a handful of others were found to be adequate to produce a reasonable stability. The Peerless is a killer transformer, and that's what I used on both mine and my friend Jim's amps, after an incredible run of luck on eBay at sourcing *six* of these rara avis within the space of three months.

But then there was *another* catch. When all is said and done, and everything is in order, the stock Williamson can produce low and high frequency oscillations if you look at it the wrong way. *Any* vintage Williamson amp needs a fair amount of work to be stabilized. Once stabilized, what was a great amplifier becomes a *brilliant* amplifier, because all the beauty is there without the distortions that supersonic oscillation can produce.

There's a great read at Audiokarma about cleaning up the Heathkit W2, which was Heathkit's answer to the Musician's Amplifier. It's an interesting and enjoyable read about the Williamson design, and anyone interested in the history of hi-fi would find it interesting:

https://audiokarma.org/forums/index.php ... 2m.767851/


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 44 posts ]  Go to page Previous  1, 2, 3, 4, 5  Next

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 45 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group