KT-88 UL screen Q

[clyde1]

Finally I have my KT-88 build done, unfortunately the matched pair I purchased from CE Distributing were a little too unmatched, only a slight 40ma
discrepancy! I put a spare pair in from Mojo, .4ma out, in the ballpark I'd say.
My question is: Do I need screen resistors running the KT's UL? I have a DPDT switch installed to switch between pentode and UL mode. I'm running
1K screen resistors right now for both classes of operation, plate volts around
590 to 600, screen volts trail by 1 volt. I'm biased at 46ma for about 26-27
watts dissipation. If I move the screen resistors to the pentode side only, the screens may run higher than the plates in UL, posssibly over the 600V limit. Is this a problem? Thanks.

[mhuss]

A series screen resistor is not normally used in UL mode. The advantage of UL is that the screen always stays at a lower voltage then the plate, so it is generally safe to omit the resistor. Note that the 600v rating is for static conditions, I don't think it applies to peak signal conditions.

--mark h

[clyde1]

Thanks for your reply Mark. My screens running straight UL (sans resistor)
have more voltage than the plates. This is normal as the 43% or so tap on the OPT refers to impedance, not voltage. As I am near the 600V screen limit
I'm wondering about a margin of safety.

[mhuss]

Sorry if I was unclear. The static voltages on the screen will be very slightly higher then the plates (as they have slightly less resistance W.R.T. the center tap), but the AC voltages will always be lower on the screens.

600 volts is an awfully high B+ -- how come you're running so hot?

--mark

[clyde1]

Why so high?-My client wants loud and clean for steel as well as six-string, so
I'm at the max for this TX, even running the primary through the unused 5V tap for a 15 volt boost arrangement. I just fired it up, definitely loud, clean
and huge bottom. I bypassed the 470R screen resistors when in UL and like the sound better so out they go and I'll go to one 1K feeding the screens from the pentode feed. Do you think 5W will be enough? Thanks for your input Mark.

[Ray Ivers]

clyde1,

Just FWIW... in this situation I would use individual 100R or 220R screen resistors to the UL/pentode switch, switching in a common 680R or 1K resistor to B+ in pentode mode (I would use a 10W resistor to be safe, at least until you can measure your actual full-power screen current).

Ray

[Steve Conner]

Yes, I heard that if you use no screen resistors at all, it can lead to RF oscillations. Those individual resistors (aka screen stoppers) should be mounted close to the tube sockets with as short a lead length between the resistor and the screen pin as is practical.

I don't know how bad this is if you're only using two tubes. Most tales I've heard of RF oscillations have been in amps that have several pairs of power tubes in parallel. But I always like to use at least grid stopper resistors.

I'e also seen UL circuits that need series R-C networks between the plate and screen taps on the transformer. I guess this is for poorly made OTs that have a lot of stray inductance, though.

BTW- A good pair of KT88s should eat 600 volts for breakfast and ask for more!