Author Topic: GU-50(Ray Pentode) VS KT88(Pentode) Triode connected.  (Read 11039 times)

0 Members and 1 Guest are viewing this topic.

Offline assendor

  • God Member
  • *****
  • Posts: 488
  • Got to DIY!
    • View Profile
GU-50(Ray Pentode) VS KT88(Pentode) Triode connected.
« on: December 18, 2009, 12:50:29 PM »
Hello all,

Want to Rewire an amplifier to use GU-50 instead of KT88.
it is/or was a Audio Note Kit2 Amplifier.
-Output transformers 2.6k Ohm.
-Power supply B+ is 400v-420volt.
-KT88 can handle 600volts at g2.
-GU-50 is rated only 250volts g2 voltage.

Amplifier are now Triode connected.
This works wonderful with KT88, but what about GU-50 ??

If we put a 100ohm resistor between g2 and Plate
to run tube "Triode Connected",
it's quite obvious the voltage ends up higher than 250volt at g2.
OR is this misunderstood?
How can this be solved, still running the GU-50 tube similar to a Triode connected
KT88??

It is possible the GU-50 is much more linear already, so that the Triode connection
is not as different sounding from a triode connected GU-50.
It can appear from others experiences that this GU-50 tube is FAR more linear than KT88/6550/6L6 types of tubes.
Still it possibly gets even more linear Triode Connected, and is therefore the
preferred type of connection.

Have already looked at some possible solutions to the problem,
still it would be interesting to get other opinions how to do it.
Also some have tried with 375volts B+ voltage, connecting the amp in
Triode with only a resistor, and the tube did not fail.
Could it be these maximum limits are set WAY low, or is it likely to be failing
when we apply much higher voltages..??
Thanks.

-Jonas

Offline yeo

  • YaBB Administrator
  • God Member
  • *****
  • Posts: 2043
  • diy rules!
    • View Profile
    • diyparadise
Re: GU-50(Ray Pentode) VS KT88(Pentode) Triode connected.
« Reply #1 on: December 18, 2009, 03:58:48 PM »
hi jonas

check the datasheet. see how much current g2 conducts. with any resistor you put in, the voltage should be lower than b+ already.

but with b+ at 600v and this requirement of <250v... hmm...

you may try to drop the voltage down via a voltage divider and provide this voltage to g2.

some folks regulate it, via a string of zener diodes. or some gas discharge tubes like 0a2 which looks nice too...

yeo

Offline assendor

  • God Member
  • *****
  • Posts: 488
  • Got to DIY!
    • View Profile
Re: GU-50(Ray Pentode) VS KT88(Pentode) Triode connected.
« Reply #2 on: December 18, 2009, 11:08:49 PM »
Hello yeo,

thanks a lot for information.
I try to understand these things better, it's a bit hard to get a complete understanding..  :-\

Here comes like always some questions... :)

1. Maximum power dissipation is average 5W in g2.
This would in my "mind" mean that with 400volts Plate voltage, the maximum allowable continous current
running into g2 would be 12.5mA. This would give exactly 5W in g2??
So to keep it some lower would be preferable, so it does not exceed this value during signal situation.
Or maybe this is no problem since the signal are supposed to swing same amount negative as posistive?

2. with 250volt at g2, it would then be allowed an maximum continous current of 20mA??

If I understand this correctly with the Currents and power,
This makes your idea of measuring the g2 current a very clever way to check if everything is running fine.. :)
If it does not exceed these maximum power limits, then everything is OK?

3. Voltage Dividers.. Yes this sounds like a good idea.
When getting + voltage from plate then, and divide it with two resistors to ground, and letting the
divided voltage at 250v enter g2, Should there be used any capacitors in this circuit anywhere?
Would it be preferable to have a capacitor bypassing the upper resistor, or lower resistor, or both??
Will this still leave the GU-50 tube in triode connection=?==

4. I don't at all understand when the tube is running like a (Ray)Pentode or when it is a triode.
If we apply a voltage 250volt from B+, then it is Pentode connected.
If we apply the same voltage from Plate, then it is Triode??
Is this because the voltage applied from Plate varies with signal, and therefore vary also g2 voltage
according to Plate voltage?? Is it this that makes it operate more like a Triode?

Thanks for any input, hope it's not to many questions to answer here..  :-[ :-[ :-[

-Jonas

Offline KenC

  • God Member
  • *****
  • Posts: 700
  • Got to DIY!
    • View Profile
    • Tinkering
Re: GU-50(Ray Pentode) VS KT88(Pentode) Triode connected.
« Reply #3 on: December 19, 2009, 03:37:15 AM »
Jonas,
When the data sheet says "max power 5W" then you shouldn't give more the 5W to the grid. Your priority is the Voltage & if the tube is in good condition & you did your calc correct, the current drawn by the grid should be lower the the max (normally). Seriously, I would avoid letting the grid getting too near to their max power dissipation.
There are more then one way to supply the grid voltage. you can tap from B+, drop the voltage with a resistor, or set up a voltage devider between the B+ & the ground, current source, regulate it with zener diode or other means.... just a few to name. Take out your Morgan Jones' book & they will explain.
How much current at which voltage at the grid? check the data sheet, you should have a table/graph to estimate.
Cap for the voltage divider? No, this will block the DC & only allow AC through.
For your 4), you need to read through the working principle of triode, tetrode, penthode. & the history of why triode, tetrode & penthode came about.
But in short, in penthode, the grids act an an extra electrode to "help" focusing the electron flow, so apply a voltage to it, it will do its job, but when connected to the plate, it become part of the plate & in effect the tube works only with 3-electrodes a.k.a. triode.

Hope this help.

Merry X'mas

KenC

Offline assendor

  • God Member
  • *****
  • Posts: 488
  • Got to DIY!
    • View Profile
Re: GU-50(Ray Pentode) VS KT88(Pentode) Triode connected.
« Reply #4 on: December 19, 2009, 04:23:51 AM »
Hello KenC,

Oh thanks for information,, :) :) :)

Yes I'll look up all those things in Morgan Jones book.
It's a lot to read, but still it would be worth it to be able to understand it.

If g2 voltage is limited to 250volts, then this amplifier no matter what cannot be run in
"pure" Triode connection, as I understand it, with higher B+ than 250volts, if we are to follow the
datasheet's limitations.
Now some have claimed this g2 value set to maximum 250v is very low, just to be completely
on the safe side.
Guess it can stand a lot more, since some also claim it works fine with higher voltages.

I will read in Morgan Jones Book, then come back with more questions how to understand
what I have read.. HAHA┬ž!

Thanks again.

-Jonas

Offline soundbrigade

  • God Member
  • *****
  • Posts: 168
  • Waiter - give me a double triode!
    • View Profile
Re: GU-50(Ray Pentode) VS KT88(Pentode) Triode connected.
« Reply #5 on: December 20, 2009, 06:00:48 AM »
Do youhave the triode curves?
Else here you are: http://www.pastisch.se/faktiskt/GU-50.jpg
/M

In a distant tree sat a little bird, and when he saw the spaceship disappear he wisely nodded his head.

Offline assendor

  • God Member
  • *****
  • Posts: 488
  • Got to DIY!
    • View Profile
Re: GU-50(Ray Pentode) VS KT88(Pentode) Triode connected.
« Reply #6 on: December 20, 2009, 06:16:04 AM »
Hello Magnus,

thanks a lot, that's GREAT!!

-Jonas


Offline assendor

  • God Member
  • *****
  • Posts: 488
  • Got to DIY!
    • View Profile
Re: GU-50(Ray Pentode) VS KT88(Pentode) Triode connected.
« Reply #7 on: December 21, 2009, 02:51:26 AM »
Hello Magnus,

Found an interesting wiring of the GU-50 tube, is it your hard-drive this Image comes from??
Some claims this is a Triode connected GU-50.

Guess it is not a 100% triode operating GU-50, since then g2 should be "electrically" directly
connected to Anode/plate, so that there is no difference in potential between Anode/Plate and g2.

   In this case the potential of g2 is not the same as Plate/Anode, But AC voltage vary with Plate,
at least his is what I understand here.
How about adding a Capacitor bypassing R12, then the AC signal swing follows even more exactly
like it would do when normally Triode connected.

Anything I have misunderstodd completely here now??

Thanks for any input

-Jonas