• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Transformers Specs for a 2x KT88 Amp

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi all,

I have good luck gettitng feedback on my projects in the forum, so I thought I'd throw my next project plan out there and see if you all agree with the direction I'm going.

I want to build a 2XKT88 amp, per the GEC data sheet. They have 2 design scenarios for a 100W tedrode mode and 70W UL mode... both with identical B+ (560vDC) and primary impedance requirements. As expected the screen supply in tetrode mode is about 300vDC. Maybe I'll have switchable modes, I definately want to experiment with both and see which I prefer. I figured I could get some transformers made specifically for the job, Heyboer came back with a resonable quote for the following:

OT:
100W
4.5k ct w/ UL taps at 40%
Anode current = .150A per side.
Secondary - 4/8/16 ohms preferably all usable via 1P3T switch w/ no rewiring

PT:
120vAC primary
395vAC (under .400A load) w/ CT (shooting for 560vDC under load w/ a full wave bridge rectifier)
5A at 6.3vAC heaters, or 3.5A at 6.3vAC and 1A at 12.6vAC


So I wanted to get some review and see if folks with more experience think this is the right direction regarding specs, if anyone knows of any caveats, or would make adjustments or additions given this is a one-off custom. For the record, I'll be deriving the Tetrode mode screen supply from the CT like this:

http://www.ozvalveamps.org/playmaster/pm117cct60w2.gif

Thanks to gingertube for tipping me off to this method in another thread.


Regards,
Brian
 
OK, that's fair. Running under 500vDC makes the power supply filtering a bit simpler as well... not sure what the tolerances are on the standard JJ can caps for voltage. I guess I should stack them with a voltage divider regardless just to be safe.


How about the arrangement in the link for getting 1/2B+? I can't for the life of me get my head around how it works and I've never seen it anywhere else. I'd like to hear from someone that has seen it and knows it works well (ie: is stable and low impedance).



Brian
 
You will need to series connect capacitors to handle that voltage, remember that if the B+ is 560V when running then it is likely to be around 660V (100V higher) at switch on before the tubes warm up and start drawing current. I'd be looking at series connectiong 2 off 350 or 400V rated caps. The standard JJ Caps are rated at 500V.

If you use that power supply arrangement then put a HV fuse (say 500mA) where it shows the standby switch.

Just finished rebuilding the power supplies of a 1965 Goldentone 60W Bassmaster which uses the same output tubes and transformers as the circuit you linked. I used a full bridge rectifier on the bias supply and suggest you do the same.

Cheers,
Ian
 
Last edited:
Gingertube,



Noted! JJ makes some 220uF 385vDC caps that I can put in series straight off of the supply. I'll use a third to filter the screen supply.

It looks like I'll need to add a Bias voltage winding as well due to the B+ arrangement correct? No way to tap off like in a marshall, right?

What is the benefit of doing full wave vs. half wave in the bias supply? Is it related to the output tubes? I'm going to use KT88s not 6DQ6 (never even heard of them before this...)


Thanks,
Brian
 
tikiroo,


Thanks for puttng it that way, makes some sense now how the voltage 'shifts' more positive. What I still don't get is how you have access to the 1/2B+ DC rail at the CT. Is the diode bridge forcing the entire transformer winding to float around 1/2B+?



Thanks,
Brian
 
Brian,
With a bridge across the full secondary like the posted example you will see that one end of the full winding is always clamped to 0V via a diode (the bridge negative to 0V link). The end changes with the half cycle of the AC signal. The other end of the winding is at full positive DC voltage and the CT will be at half that full winding voltage and always positive.

Cheers,
Ian
 
I probably wouldn't go too far over 500V on the plate with modern KT88..

Decades ago having built several KT88 amps around 50-100W+; I found the 50W versions with a slightly lower B+ sounded nearly as loud as the 100W which at 500V is an awkward voltage. So I specifically made amps around the 460V B+; the single 500V electrolytic is an advantage, against 2 staggered in series and the o/p tranny with a slightly lower A-A Z of 4Kohms; I ended up close to the GEC 65-70W version.
The attractiveness of 460V B+ is a smaller power transformer; and brings in the 6550 group, so one can compare tubes.
Only a hint; amp builders like myself and other users, often find out the hard way when todays tubes don't seem to last long when run with dissipations higher than 2/3 of specified plate ratings.
At present, I am putting the E Harmonix KT90 though it's paces, which I find is quite stable (currently 2000 hrs) being run at 600V at 80mA quies which is 48W plus a bit for the screen.

Those of us who remember the original KT88; this OEM tube in it's 1950's hayday could take 600V at 50mA easily; that's 30+ Watts sitting for 10,000hrs before replacement.

rich
 
Go for the same curves as per KT88 (tempting to say KT66) but back off the curve at 33W dissip, even though the data sheet quotes max plate dissipation at 42W..These new tubes don't like higher and with lower 450V B+, rules out 100W power per pair. I run each at 75mA quies max for a bit more class A at low listening levels and have a couple of amps with such that give distortion comparable to KT88's. With some ears one can reduce the quiescent even further.
A pair on 75mA I quies each will on max signal give peak current of ~130mA (each) on music equating to around 65W power..no more.
The TungSol new edition can't take the duration when compared to the to the famed GE 6550A; that tube was practically a KT88 straight swap.

pic of 3 tubes, A GEC KT88; 6550A JAN; New Edit Tungsol 6550.
Of the 3; the Tungsol has the smallest area anode. You won't find a 6550A in catalogues without a having big wallet. Out of the 3, it has the largest area anode.

r:-
 

Attachments

  • 88-6550A Tung.JPG
    88-6550A Tung.JPG
    149.5 KB · Views: 223
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.