KT 90

[Jana]

I'm in the early stages of thinking about an amp to build. I don't know what it will be yet. But, I want to try something different, I am so bored with working with EL34's, 6l6's, all the "normal" tubes.

I am intrigued by what little I have read on the EH KT90.

Does anyone have experience with these tubes? Reliability? Quirks? Opinions? Facts?

Peace,

Jana (who is contemplating trying to shoehorn a pair of KT90's into a champ size amp but probably won't do it because of sensibility reasons and will build a regular sized head instead).

[breakfastbuddy]

hey they use kt 88 and thats close enough , pp Deliverance 60w pp
i heard that kt90 gives a better distortion than kt 88 .

[Andy Le Blanc]

I currently have an amp up and running with KT90, Incredible tone.
If you can supply the current the plate has max dissipation of 50w.
I used it an open mic/jam yesterday and was blown away again, the power,
definition, nice break up. But again its finding a power supply that can handle
the current demand. I have a PT from a Twin supplying a pair of KT90 and
I can only bias to around 35w plate dissipation each, the tranny is not adequate
to bias further, but it is sufficient for 6550 and KT88.

[Jana]

Thanks Andy.

I have iron that can supply the current. I have a pair of power transformers that are rated 700ma and will give me 480volts DC. The 6.3vac is rated at 8 amps.

Are you using the Electro Harmonix KT90's?

I am having trouble finding data on the EH tubes. I have found data on the earlier EI version but from what I have read, some say it is not the same tube. I emailed EH requesting data but haven't heard anything yet. Are they similar enough to a 6550/Kt88 to use that data for transformer impedance, etc.?

[Andy Le Blanc]

Yes, EH KT90. The only tube data that pops up on the net is the EI data.
Its supposed to be a drop in for KT88/6550. Ive got mine set with a 3.5k load
which is on the low side, you have to watch out and not to go too low because the
tube will draw significant current with too small of a primary resistance, most of
the data for KT88/6550/KT90 puts a load optimum for a pair between 3.5K
and 5k. You should be able to generalize from KT88 data but the improved
plate dissipation of the KT90 puts it heads and shoulders above the others.
It allows you to get over 100w output with two tubes with a very conventional design,
and not have to use a transmitting type.

What was the source of your PT? I've wanted to really push these tubes and
have started to consider Hammond or Piltron toroids for the VA ratings.
If EH is going to market a KT90 they damn well better be able to be biased to
the max plate dissipation as per existing published tube data.

[Jana]

I had special ordered three sets of transformers about 10 years ago through an amp parts supplier who is now defunct. If memory serves me right, they are made by Pacific. The OT is rated 150 watts at 100% duty cycle, with an impedance of 2500 ohms P to P. They have 2, 4, 8, 16 ohm taps so they can be used as 5000 ohm transformers with pair of tubes (originally I had planned to use a quad of 6550's).


I would part with a set, PM me if you're interested in a trade or ?

[Phil_S]

This kind of power supply is something you can troll for on ebait with fairly good success. The demand for such is not that high. Examples:
http://tinyurl.com/dg8ro5
http://tinyurl.com/ccpfzb
http://tinyurl.com/db6luc (if you find this guy on his regular business site, it costs less)
I think you get the idea. Old organ PT's generally have plenty of filament supply but you might have trouble getting one with high enough HV. I've got one that's good for 6.3v @ 9A, but the HV is only 285-0-285. It probably weighs 15 lbs.

[Andy Le Blanc]

Thanks Phil, thats not a bad lead, whats his regular business site? His ebay store?

Jana, you have 3 sets? I scrounged around for trannies and built a test chassis,
I have a light weight sheet metal brake that needed to be set up for the gauge,
and the Idea was get close to the published tube data for KT88/6550 by
using load to up the P-P impeadance with a less than ideal opt. Which has
worked wonderfully. I used a 1750 ohm primary and 4/8/16 sec. with an
8 ohm load to get a 3.5k load. The KT90's were an after thought that worked
out well. The cost was as I stated earlier, that the DC resistance of the OPT
is too low to keep the tube from really taxing the PT. The pay off is in the tone.
It has so far proven to be valid approach, the amp plays very well.
It might be a good opportunity for a joint build with the same trannies.
The amp I have up and running is very simple but has these design features.

The load mismatch to create the p-p impedance as discussed.

KT90........ fixed bias, to explore KT88/6550 and the max plate dissipation of EH KT90.

A split load inverter....... no global feed back

And, the tube that serves as the pre amp is diode biased, very simple.

I paired mine with a JBL E120-8 in an open back cab and was completely
floored. I'd be very interested to use the same approach with a more ideal
transformer set at the very least, post the tranny specifics

[Jana]

Digging through my notes I found my original scribbling for the specs that I requested for the transformers.

OT
150 watts
2500R P to P
2,4,8,16 ohm taps
no less than 200ma DC current capability per side of primary

I have measured the DC resistance of the primary and it is ~28.5R on one side and ~35.5R on the other side

PT
120VAC primary

Secondaries
340 VAC at 700ma. (no Center tap, requires full wave bridge)
100 VAC bias winding
6.3VAC at 8 amp
Copper shield

They are big, each weighs about 10 pounds. Upright mounting on both.

I had to order three sets.

I have used one set for a 4-6l6 amp which I sold. It puts out 98 watts clean.

I have another set in a 2-6l6 amp which puts out around 60 watts clean.
This is the amp that I would try the KT90's on. There is space on the chassis for two more tubes, purposefully planned that way. All I have to do is punch two more holes and rewire the output leads for a 2500R impedance instead of 5K, which it is now.

I have one set which I had on an amp but cannibalized the amp for other parts before I completely finished the amp (It happens in my world, lol) So, they have been mounted on an amp, but for all intents and purposes they are new from an electrical standpoint. I would part with that set.

I figure that with the increased current load of the KT90's, the B+ from these transformers will probably be about 460. Biasing a KT90 to ~70% for AB1 means biasing them to about 75ma. each. I would feel comfortable doing that with these transformers. Even with a quad.

[Andy Le Blanc]

That puts the plate dissipation at 34.5w at that V. The design-max for KT88
is 35w, for a 6550A, its 42w. I was able to eek 32w out of a Twin PT
before it had issues with heat. Here's the rub. A full wave bridge with a
capacitor input supply will give .62 x sec. I AC, that's 434ma., about 100ma
more than a generic twin replacement, on par with the Hammond
twin replacement with a 450ma sec. I AC. The familiar full wave (with ct)
works out to 1 x sec. I AC. It should be enough even to push the tube if you choose to.
Mind the screen grid resistors. The OPT sounds good. How long are the leads?

[Jana]

The leads on the primary are 9 1/2 inches long minimum, the red center tap is longer. On the secondary the common is 9 1/2 inches long and the 2,4,8, 16 ohm taps are 12 inches or longer.