Tackling the New Beast: KT-120 Tube

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

schmidlin

Well-known member
Joined
Jul 7, 2008
Messages
165
Location
Ohio
So this guy is obviously suitable as a replacement to 6550, KT88 etc, but I want to see what extra I can pull out of a pair of these for an audio amp.

yhst-8476489043850_2111_3192985


"With a plate dissipation rating of 60W (compare that with the 35W of the 6550, the 42W of the KT88, and the 50W of the KT90), a pair of these KT120 tubes in push-pull configuration can deliver 150 watts or more of audio power. And when used in a vintage circuit designed for 6550 or KT88, the KT120 will deliver unbelievable amounts of clean headroom, low end rumble and dynamic punch all without overloading."

Max plate 850 volts, screen 600.
http://www.newsensor.com/releases/KT120TungSol.pdf  (data sheet)
http://www.tubedepot.com/ts-kt120.html  (retail)

So I have a PT that can supply ~650VDC under load, and my question is how to handle the screens. 

Option 1:  Get an OT with screen taps and add some say 1K-2K screen resistors, getting say, 140 watts out of a pair.  I know this is pushing it, but it is common for power pentodes to get a higher max screen voltage rating in UL mode, and I just don't think they got around to spec'ing this just yet.

Option 2:  Put in a second smaller PT that will supply 400VDC for the screens and the pre tubes.  Yes, it's playing it safe, but I suspect I will only get ~100 watts out of this configuration.  But I can do that with 6550s.  Yawn.

Any thoughts?
 
Maybe this beast will spawn some ideas?

http://www.timeelect.com/6550a-ex.htm

http://www.prowessamplifiers.com/schematics/fender/400_ps-300_Watt_Bass_Amplifier.html


I wonder how many PS owners will be tempted. . .

 
Thanks lassoharp,

Wow, I had the 160 version in my shop last week, and I considered using their OTs that are being sold as NOS on the 'bay.  Not *quite* enough iron for me.

I think I will opt for the KT-90's as their screens go to 650 for design max.  Funny, I feel I'm the only one who thinks that screens are the weak link, not plates.  :p
 
> So I have a PT that can supply ~650VDC under load, and my question is how to handle the screens. 

That's actually no problem at all...  string another RC section into your power supply, and your value for the R.  Say your screens and preamp are pulling 10mA, if you need to drop 60V that's only a 6k resistor.

Maybe I'm missing some nuance?  You'll need to drop your B+ for the preamp tubes anyway...
 
> Maybe I'm missing some nuance?

Screen current is variable, maybe 10mA idle and 40mA at full roar.

Using 6K and adding your 10mA preamp load, you get 120V to 300V drop.

Or counting down from 650V, 530V to 350V from quiet to loud.

That's just too much. Even if your preamp does not mind large variations of B+ (slowed and delayed by filtering), your maximum output may (probably not) be limited by the 350V at screen, while the required driver signal is large due to the high screen voltage at idle.

Set your screens to carry the load, no higher. IIRC, 6550 can do fine at 600V plate 300V screen and 5K load. In overdriven guitar use, somewhat higher screen defers blocking; but double is just too much.

I'd say: do NOT pick your power transformer first. Understand your tube first.

I wrote this last week but the dog ate it.

This KT-120 is NOT good for new designs. It is 1.5 of a 6550 at 1.5 the cost. 1.5 times power is not a "Wow!" difference. The KT-120 is not sure to stay in production for the life of your amplifier. If you need "more power!", run four 6550 or eight 6L6/EL34. (Looks more impressive anyway.)

The KT-120 is of special interest for a few older amplifiers which pushed the limits of genuine made-in-Newark Tung-Sol 6550s and the golden-age KT88s. Some of these amps EAT modern "6550/KT88" tubes for lunch. If the KT-120 does not die so fast, it keeps these old amps running as designed. But for "modern rational" amp design, stick with the highly popular tubes and team-up to reach your power goal.
 
Ah yup, true, true.  You could use a backwards power Zener then, a 5W one would let you pull up to 50mA at 100V.  :)

You could also get a PT with another tap so you can get the right screen voltage after the rectifier, or just use a whole separate PT for the screens and pre.  (Not a dumb idea.)

Happy breadboarding  \m/
 
As I recall the original SVT used 6146 transmitter tubes for audio outputs.  The fluctuation in screen current and voltage drop is what caused these amps to burst into flames.  RevB had a separate supply for screens and it had the same problem.  RevC used 6550's.  Problem solved until the world as we know it ran out of real 6550's.

>I want to see what extra I can pull out of a pair of these for an audio amp.

Great fun with guitar amps.  They sound fantastic when pushed to the edge.  I had a single ended 6V6 amp running with a plate dissipation of 16.2W.  Key word is "had".  Sounded absolutely bad ass right up until it arced.  That output transformer is now a paper weight.  The amp now dissipates 12 watts and still sounds pretty good although it's vintage OPT is no more.  Champ OPT's are cheap.  An OT for this project will be as least a C note.  I don't think I'd try to push a pair of these to the edge.  Big ass high powered tube amps have a fairly consistent history of not lasting.  100 watts seems to be the limit of 100% stability.  Sure there's plenty of higher powered tube amps but they are in constant need of attention.  I had a friend with a pair of Conrad Johnson amps.  When he hit is 3rd set of outputs, he sold em.
 
Related Article:

Are you aware of the relatively rare and not widely used 200w Marshall Major? This was a production amp between 1967 and 1974 (first called the Marshall 200 then the Major starting in 1968) which had four KT88 tubes – giving it 200w of brain-pounding power.

Or actually, more like 140w. John “Dawk” Stillwell, Ritchie Blackmore’s tech and the guy who modified Ritchie’s Majors, says on his forum that “the spec of maximum output rating on a KT88 is only really 35 watts RMS.”

(Btw, according to the book “Jim Marshall: The Father of Loud,” Jimmy Page had his stage Super Leads modded to run four KT88s for 200w!)

Apparently Blackmore was the one who pushed Jim Marshall into 200w territory, but Marshall couldn’t design a 200w amp that wouldn’t quickly fry itself. So Marshall asked the Genelec tube company to design the amp.

What they got was an amp that by all accounts was sensitive, eats tubes and can’t take distortion boxes thanks to something having to do with an “ultra-linear” output transformer. (I’m not an amp tech by any means, so if I’ve botched any of that explanation please let me know!)
Major History

Anyhow, here’s what Dawk says about the Major on his forum (all info is his, edited):

> Some of the history of this came from Marshall’s own Ken Bran and the rest is from my years of experience of working on [Majors]. It all started when Deep Purple went from playing clubs to 20,000-seaters. Many prototypes of a 200-watt amp were built by Marshall, they all failed. So they had GEC design one for them. It’s the company that made the famous Gold Lion KT-88 tubes.

> To get the 200w RMS rating, GEC had to boost the b+ [plate voltage] to 650 volts dc. The maximum designed rating for a KT-88 is 450 volts plate voltage. Then to top that off they used an ultra-linear-designed output transformer. It’s basically designed like a hi-fi amp. The Major has a very clean sound and low gain when compared to the rest of the Marshall line. That’s why Blackmore’s sound on Made in Japan sounds so good.

> Those people at Marshall back in the day didn’t know what rock and rollers were going to do to that amp. When a booster or some kind of other pedal was used to make some distortion, the 650 volts of b+ would jump to 1,800 volts – that’s called “prv” voltage. The optx [output transformer] is working sort of like a fly-back transformer in a TV set. That’s what’s producing that high prv voltage.

> Then the b+ at 1,800 volts would arc over to the closest ground, usually between pins 2 and 3 on the output tubes, causing a lot of smoke, burning of the tube socket and a real bad smell…leaving a big black carbon mark inside the chassis where it arced. This would happen so fast that it didn’t have time to blow the fuse…blam! Sometimes the fuse holder itself would explode. [This what Dawk calls "the Smoke on the Water effect."]

> [Because of] the cost of warranty repairs on these amps, Marshall took some steps like covering the primary leads with extra insulation and changing the tube sockets to ceramic. Then the prv would arc over in other places like on top of the tube socket or inside the tube base – the worst being inside the output transformer itself. The transformer could test good, but then only to arc over again when the first hard chord was played. Most of all the original Dagnall output [transformers] are damaged in this way. The amp will play okay until it gets a loud signal, and then blam! Because there no signs of an arc over in the chassis, you’re not realizing that the arc over was inside the optx itself.

> At that time Marshall couldn’t keep up with production let along sending parts to America. So they (Unicord) hired this old guy Otto to make transformers in a back room in their warehouse on Long Island. I had Otto wind a lot of mylar insulation between the primary and secondary in the new output transformers he was making. Dagnall had used masking tape for this – masking tape! What a piece of crap that transformer was. The Majors that are not blowing up when pushed hard (now) probably have that upgraded Otto transformer in them.

> The warranty replacement got very costly for Marshall…new optx and new KT-88s as well. So that was the main reason for this amp to be discontinued by 1974 [though] there were many more reasons for this decision…. Lack of sales and their high price also played a part.

> Ken told me once that all the parts [would] be installed by the girls [in the Marshall shop and] only then would the heavy transformers go on the chassis. This made the amps too heavy for the girls to pick up and move around on their benches. [Ken said] “We here at Marshall were loaded with back orders,” [and] the girls doing the assembly could be making three 50-watters and two 100-watters in the same time it took to build a 200w Marshall Major. So case closed on the 200w Majors.
The Players

Blackmore and Paul Kossoff are the most famous Major users, along with Mick Ronson of David Bowie’s band who called his Major “the Pig.”

A lot of mods were done to everything back in those days. Not sure about Koss’ and Mick’s amps, but Blackmore’s majors were messed with quite a bit to keep them stable and to help get Ritchie where he wanted to go tone-wise. Apparently Dawk added an extra gain stage (preamp tube) to help that along, and did who knows what else.

Ritchie boosted his signal with a Hornby Skewes treble booster and in the studio with an Aiwa reel-to-reel tape recorder, among other items. His guitars were anything but stock. So while the Majors contributed to his tone, it certainly didn’t end there – which I know for a fact because I had the opportunity to play through one of his actual Majors recently!

However, the head in question had had the Blackmore mods removed, so the Blackmore gain wasn’t happening. The head was run through an old Marshall 4×12 and an old Marshall 4×15 (which I believe were recommended for the Majors). It sounded GOOD, and actually started to break up halfway up even with the 4×15 cab.

Was it loud? Heck yeah, with two cabs especially. But just remember that 200w (or 140w) is not twice as loud as 100w.
 
I never dreamed I would challenge the venerable PRR, but I will do so.  I think the KT 120 is *ripe* for new designs.  Those Marshall Major designers were morons at best, and the SVT designers had it spot-on.  Marshall basically ripped-off American designs and then thought they could push it: not so. 

A lot of the SVT success IMHO derives from the lower 6550 screen voltage.  And much of it comes from the cathode-driven grid driver of the 6550s.  Brilliant and stable.  I plan to use both of these in my design.

Just for yuks, check this out:http://www.bonavolta.ch/hobby/en/audio/6550_1.htm
I have been staring at this for years now and find it full if great ideas.  I really like the non-cap coupled PI/Grid link.  I will see if I can make this happen. 

My last design had cathode follower PI/Cap/Grid link and it has much acclaim.  A fart-less bass amp.  Not sure if the cap effects impedance...but the buyer is in love.
 

Latest posts

Back
Top