2xKT88's anyone?

[clyde1]

I'm doing a UL KT88 amp and have never ventured this way before. The design calls for an AB763 variant so it can be used for both guitar and steel guitar,
no vibrato. The player would like it clean and loud, hence the UL hook-up
in my mind. What voltages should I be shooting for with the KT's and what bias point? Will the standard Fender PI drive them? Will a single GZ34 handle the current (3 or 4 12AX's) and if not, would the sag from a 5U4G be a problem
to someone who plays steel through this amp? So many questions, sorry. I
know Ray is the resident (one of them) KT88 man here. Thank-you.

[MWJB]

If you're just looking for an AB763 style tone with improved headroom and dynamic response, then you probably don't "need" to go UL...unless you want a really clean guitar tone?

I've tried KT88 (& variants) with tweed, brown and blackface long tailed PI's, they work fine (though that's not to say you might not want to optimise). You might want to experiment with a 12AX7 in the PI rather than a 12AT?

I'd keep the plate voltage in the typical Fender ballpark, 470v to 510v-ish. Current-wise, they'll take 55-60mA but have a listen at lower currents too (30mA+?), especially with the steel.

I've never had the guts to try them with a GZ34, how about a pair of 5u4's? 4x 1N5408 is what I've been using, a lot of the common PT's with 6A or more heater current don't have rectifier taps anyway.

Just my £0.02...

[Ray Ivers]

clyde,

If it were me I'd go with 600V B+ (since UL pretty much eliminates screen-overvoltage issues, and the KT88's are rated for 600Vs anyway) along with a 5K 100W OT like the Hammond 1650T (120W) or similar; I think 4.3K is that max-power match for this B+, but I'm not 100% sure. This should give nearly 100W output, so the GZ34/5U4 is out, and the 12AT7/47K Rp Fender PI should work fine; I'd also include the Ruby Zener mod since it works so well when needed but vanishes when not, and allows you to choose practically any bias point you'd like to try; in light of this. I would start at about a 30% P.D. idle point for a loud/clean application like this, YMMV.

Ray

[clyde1]

Thank-you for the replies gentlemen. 600V and the 12AT7 PI was what I was
thinking too. My PT will handle 2 rectifiers so I'm going to try 2 GZ34's.
Has anyone had experience with the EI KT-90? I assume operating points are the same but I'm wondering about reliability and tone. The KT-88 is a much more robust tube construction-wise.

[Ray Ivers]

clyde,

You might keep an eye out on eBay for pre-war Ei KT-90's, with a lot of spot welds on the plate structures; these are good tubes IMO. I recently did some very basic power testing of the E-H KT-90 in a 600Vp/Vs UL circuit, and it seemed to put out about the same power as the other tubes I tested (E-H KT88, early Svetlana (now Winged C) KT88's, Teslovak/Groove Tubes KT88, recent-production Ei KT-90's, etc.). The old Ei KT-90's were pretty ruggedly made, and have thick Pyrex bottles too - plus the KT-90 has a lower rp, and seems to pull a bit more bottom end from average guitar-amp OT's IME.

Unless you're absolutely dead-set on thermionics, the Weber Copper Cap rectifiers might be a good alternative - apparently you can even 'roll your own' and spec out a rectifier that never existed, which is way cool IMO - and unloaded 5V windings means just that much more 'core headroom' in your PT.

Ray

[clyde1]

Thanks again Ray. Yes I do have pre-war KT-90's. They've been around awhile and I'd like to use them in a build. As far as Copper Caps go, I've tried them and found their turn-on characteristics to be instantaneous, for which reason I'd go hollow-state, having chassis with double rectifier holes. You've been a greaat help.

[Ray Ivers]

clyde,

OK, no problem.. as long as I'm here, here's a couple items that have been on my 'to try' list for a long time:

1) FWB, two 1N4007's plus one GZ34 (possibly w/separate tube rect. to screens?)

2) FWCT using SS diodes + series-pass tube

There are others, they're just not coming readily to mind. You can always just parallel two rectifier tubes like the Sunn 2000S, etc., just wanted to offer up some other options.

Ray

[clyde1]

The 1650T has a primary of 1900R. I'm thinking unless I run it at 16 ohms for an 8 ohm speaker I'll probably use the 1650R, primary R of 5000, but only 100 watts. Hopefully Hammond's conservative ratings will save the day.

[Ray Ivers]

clyde,

I actually meant the 5K one - the 1650R - sorry about that (but the 1650T would also work very well connected as you posted, come to think of it).

I'm going to be trying out the 1650K (3.4K, 50W) and 1650N (4.3K, 60W) in a two-KT88 600V UL design myself, I'm hoping one of them will give me 100W down to 60Hz or so without all the mass; if not, I've got a 1650R (and Heyboer ;>) waiting in the wings.

Ray

[Steve Conner]

I'm working on a tube hi-fi design at the moment that looks like it's going to be 2x KT88s in UL, and I noticed you guys were discussing alternative tubes. How about 6550s? Would something like the new Svetlana 6550s work OK in UL the same as KT88s?

[clyde1]

I have used 6550's and KT88's interchangeably but believe 6550's have marginally lower ratings for some parameters which would have to be confirmed by searching tube data sheets. Shouldn't be a big problem for most applications, tonal differences may be your determining factor.