Page 1 of 2

#1 Shuguang KT66

Posted: Wed Feb 25, 2015 9:13 am
by RobW
Hi I'm using a pair of these Shuguang KT66 valves which I purchased off ebay. They seem ok but when it comes to being biased correctly in a fixed bias amp I'm not so sure and would value any opinion on this. I've biased them with a scope turning the amp up with an input signal until it just starts to distort the output tubes. The crossover notch can be seen quite clearly and it disappears completely by the time the valves are drawing about 17ma average, the HT then is at about 465dc and bias voltage is -67.5dc. Everything I've read seems to contradict the settings as the valves are only drawing a small current about 30% when calculated. Actually it's slightly lower than this as I took the current readings at the cathode and haven't taken the screens into consideration. Everything I've read suggests that they should be biased at around 70% or at least hotter than I have them now, but my scope findings seem to contradict this. The sound is good and I get the full RMS at the speaker jack so wander is this the norm for these valves? :?

#2

Posted: Wed Feb 25, 2015 9:23 am
by pre65
I seem to remember reading (ages ago) that some KT66 "copies" were more like EL34 in use.

#3

Posted: Fri Feb 27, 2015 11:53 am
by Mike H
Do not exceed 25 Watts of anode power dissipation. Otherwise, whatever! :D

#4

Posted: Fri Feb 27, 2015 9:41 pm
by RobW
Mike H wrote:Do not exceed 25 Watts of anode power dissipation. Otherwise, whatever! :D
Melt down :D

I’m not sure who to believe,
1. The guy who said that valves needed to be biased using a scope.
2. The guy who said that they should be biased by transformer shunt method or
3. They guy who said draw graphs of it and work it out.
To further confuse why on earth put an adjustable pot on the bias supply if the range of acceptable values is so vast? I think it’s to confuse the tone deaf tech maybe into tweaking the pot by 1ma at a time? Oh dear just ran into another problem, the mains voltage its 244 volts now, back to the drawing board. :oops:

#5

Posted: Sat Feb 28, 2015 11:31 am
by Mike H
There is not a vast range of acceptable values, there is a best optimum operating point, but which must also keep the valve within its absolute maximum ratings, else it will quickly expire.

E.g. you can't just increase the HT for more power, you'll find you need to then reduce anode current to keep it under the max. anode power dissipation, but then that screws up your carefully calculated operating point. So then you need to find a more powerful version of the valve, or start thinking in terms of doubling valves in parallel, to increase the effective Pa....

Fixed bias is made adjustable with a preset because no valve is exactly like its datasheet, so same types will have different bias Voltages, if only somewhat. Also bias changes over time, as the valve ages, so will need periodic checking and readjusting. (Which is why I think buying matched pairs or quads is a waste of time, they'll all be different from each other within a month.)

#6

Posted: Sat Feb 28, 2015 11:33 am
by Mike H
PS: bias variations applies to identical type transistors too. Unless they're all on the same chip. :D

#7

Posted: Sat Feb 28, 2015 7:48 pm
by Greg
Mike H wrote:(Which is why I think buying matched pairs or quads is a waste of time, they'll all be different from each other within a month.)
Very interesting observation. I once bought a very mismatched quad of 6550's and when fitted, the PP amp hummed like hell on both channels. I returned them and replaced with a matched quad and all was humless, the quad retaining their hum free integrity for their lifetime. Maybe the valves were faulty?

#8

Posted: Sun Mar 01, 2015 5:01 pm
by Mike H
Cannot comment, can only relate my own experience :D ~ used to buy matched pairs of Golden Dragon KT66's, the amp's had bias balance adjusters to get zero DC in the OPT's ~ I found myself readjusting to get the balance back every 2 weeks on average.

#9

Posted: Wed Mar 04, 2015 8:46 pm
by RobW
I’m still faffing with these but I think 30ma seems about the best setting, . Might try some Golden Dragon ones next, what are they like? :D

#10

Posted: Wed Mar 04, 2015 11:52 pm
by Mike H
Long time since I used them, so can't remember! Still got 3 of 'em I think BUT getter mostly gone, suggesting they're going gassy.

#11

Posted: Wed Mar 11, 2015 10:22 pm
by RobW
Hi I’ve read on many forums that shuguang KT66s can’t handle high HT voltages. I did a check on the ones I have and one of them is just starting to red plate at 35ma. From plate dissipation calculations this is 60% anode dissipation, I thought that was actually quite ok, obviously not.
Can anyone recommend a better KT66, one that can handle voltages in excess of 450 volts please?.
Any help appreciated, Rob  :D

#12

Posted: Thu Mar 12, 2015 1:13 pm
by pre65
As I understand things, the plate voltage (plate = anode) of a valve is the anode voltage minus the cathode voltage.

So if plate max is 450v and cathode volts are 30v, then the max voltage actually on the anode is 420v.

Thats why valves with grid bias need less HT.

If you already know that then sorry.

#13

Posted: Thu Mar 12, 2015 4:44 pm
by Paul Barker
He hasn't stated what his bias method is but his writing reads like someone who is not fully cognisant of the facts.

#14

Posted: Thu Mar 12, 2015 8:00 pm
by Mike H
There is no KT66 that will take much over 450V.

Well, two datasheets I've got say 500V max for anode, 400V max for screen, in push-pull pentode mode only where anode current can be low, remember anode power dissipation must not exceed 25W.

For higher anode currents, anode Voltage must be reduced accordingly.

#15

Posted: Sun Mar 15, 2015 11:34 pm
by RobW
Hi thanks to everyone for the answers.
Mike H wrote:There is no KT66 that will take much over 450V.
The amp's design voltages are quite high but no higher than other vintage amp running kt66s or 6l6 and I totally agree from the data sheets I’ve read the screens are over the specified voltage but that’s how the original JTM45 was, the screen voltage was high close to that of the anode and the GEC valves were ok with it.
Paul Barker wrote:He hasn't stated what his bias method is but his writing reads like someone who is not fully cognisant of the facts.
Current draw for plate dissipation was measured at the cathode across a fixed 1ohm resistor. Currently running at 30ma at about 450 volts, however is still red platting. I think I have a faulty valve. 8)