Citation II hack

[s-petersen]

Wow that must be some kind of tube! it has over 600v across it, and only about 30-40k in series! 24w of resistors.Am i missing something?Is p1 for balance or gain?I admit I don't fully understand the phase inverter in this one.
I am a complete newbie as far as building from scratch,I have been collecting and repairing for about 30-35 years,So I may ask some dumb questions,Please be gentle.It's easy to trace,but hard to draw.

 

[s-petersen]

You might consider a CCS in the shared cathode, or tail, of the phase splitter and make the anode load resistors equal. It will improve the phase splitting.

John

Isn't there supposed to be a change in voltage/current in the 68k resistor for the Phase inverter to work properly?(newbie question)

 

[s-petersen]

Years ago I made a phase inverter pair for a pair of my Dynaco Stereo 70's to series the the amps,it seemed to work well, based on a similar schematic to this one,fron an Altec 340a
I didn't do any measurements,but it sounded good.
What is the advantage of the long tailed pair vs this phase inverter?

 

[nerdorama]

Isn't there supposed to be a change in voltage/current in the 68k resistor for the Phase inverter to work properly?(newbie question)

The current of the CCS must be adjusted to match the current that is in the 68k resistor. If the current is matched, the voltages will come out the same. The current in the shared cathode will theoretically remain constant as the two triodes will increase and decrease as a balanced pair. This will change the anode voltages but the cathode voltage will remain constant as the two tube currents alternate. The CCS makes for a more theoretically perfect LTP. The reason the original anode resistors are unmatched is because the cathode resistor just isn't large enough to make a "perfect" phase splitter.

John

 

[diamondsouled]

Wow that must be some kind of tube!it has over 600v across it,and only about 30-40k in series! 24w of resistors.Am i missing something?Is p1 for balance or gain?I admit I don't fully understand the phase inverter in this one.
I am a complete newbie as far as building from scratch,I have been collecting and repairing for about 30-35 years,So I may ask some dumb questions,Please be gentle.It's easy to trace,but hard to draw.

Not sure if Bruce Moore used the Citation 2 PT as well. :scratch2: Here's the PS for the MFA120.

 

[Kegger]

I was going to try hosting,but my provider dosen't offer it
I cropped out the part I have questions about I hope this works.

Going off of your schem, an replacing the ECC83 with ECC81 this would be
changes to make to that circuit to slip that ECC81 in there with good drive.

The right 33K plate load to 36k, the tail resistor to 20k, and the pot to 20k.
(the circuit should balance up with the pot centered, and adjustment room)

You'd also like to see 100-120v at the plate of the EF86 to bias the splitter.
(maybe replace that R3 47k resistor with a 50k pot to tune that voltage in)

.

 

[s-petersen]

Going off of your schem, an replacing the ECC83 with ECC81 this would be
changes to make to that circuit to slip that ECC81 in there with good drive.

The right 33K plate load to 36k, the tail resistor to 20k, and the pot to 20k.
(the circuit should balance up with the pot centered, and adjustment room)

You'd also like to see 100-120v at the plate of the EF86 to bias the splitter.
(maybe replace that R3 47k resistor with a 50k pot to tune that voltage in)

.

Thanks,This is great,This is the kind of advice I'm Looking for,I was trying to keep the basic circuit like the original and merge them together.
Next question:How do I adjust the feedback loop?I have a sig gen and scope.

According to the info on the BTH(driver)the cap C3(470pf)and R7(33k)are to correct for phase in the feedback circuit,I'm pretty sure something will hsve to change there also.

 

[s-petersen]

The current of the CCS must be adjusted to match the current that is in the 68k resistor. If the current is matched, the voltages will come out the same. The current in the shared cathode will theoretically remain constant as the two triodes will increase and decrease as a balanced pair. This will change the anode voltages but the cathode voltage will remain constant as the two tube currents alternate. The CCS makes for a more theoretically perfect LTP. The reason the original anode resistors are unmatched is because the cathode resistor just isn't large enough to make a "perfect" phase splitter.

John

So the CSS acts as a varible resistor,adjusted by the current,to provide a wider voltage swing at the cathode?(does the opposite resistance wise as the first stage?)

 

[nerdorama]

So the CSS acts as a varible resistor,adjusted by the current,to provide a wider voltage swing at the cathode?(does the opposite resistance wise as the first stage?)

Not exactly. It acts as an extremely high impedance to AC signals. Sometimes on the order of 100M ohms or more. It regulates the DC current through the circuit, but acts almost as an open circuit to AC forcing the AC to balance between the two halves of the tube. If one goes up by a certain amount, the other has to decrease the same amount thereby having nearly exact balance at AC assuming the anode load resistors are identical.

Here is a good thread to read on the subject.

http://www.diyaudio.com/forums/tubes-valves/160312-balance-ccs-long-tailed-pairs.html

John

 

[jaymanaa]

Hmm, so I'm not the only one who doesn't care for the 12BY7. Heck it's a dang TV tube.

 

[s-petersen]

You have to admit it has wide bandwith,but the amp gets so hot that it cooks the circuit boards,and sounds almost transistorish,it just dosen't seem to have the openness,it's hard to explain,also my amp is in pretty poor shape besides,if things don't turn out,I can always go back,no irreversable mods.

 

[Kegger]

Thanks,This is great,This is the kind of advice I'm Looking for,I was trying to keep the basic circuit like the original and merge them together.
Next question:How do I adjust the feedback loop?I have a sig gen and scope.

According to the info on the BTH(driver)the cap C3(470pf)and R7(33k)are to correct for phase in the feedback circuit,I'm pretty sure something will hsve to change there also.

Yes more then likely you'll have to tweak the feedback, I'd build it with the feedback
as is, measure/check the amp and "tweak" it from there to look good with 1k and 10k,
sweep it on down to oh 20hz and out to 20-30k to see how everything looks after that.
(without being there I can't really guide you in this part as well as others may be able to)

 

[s-petersen]

Yes more then likely you'll have to tweak the feedback, I'd build it with the feedback
as is, measure/check the amp and "tweak" it from there to look good with 1k and 10k,
sweep it on down to oh 20hz and out to 20-30k to see how everything looks after that.
(without being there I can't really guide you in this part as well as others may be able to)

Have you seen anyplace a guide,perferably with scope pics on applying feedback,C vs R, to shape the square wave response? I have to do a little more reading,I have the Morgan Jones books,there may be something in there

 

[Kegger]

I do not believe, nor at least I don't remember seeing one.
There probably is, but I haven't searched trying to find one.

I can do it myself, from experience that had plenty of trial n
error to where I'm pretty decent at it now, but I'm no expert
and certainly not the one to try an lead someone on "how to".

Loop feedback is not my expertise by long shot, I also haven't
tried to be that proficient with it as I prefer other forms of FBK
but they do not always apply to every circuit, so it is "needed".
(feedback is a hands-on design that's different for every piece)

So you'll have to work through it if your doing a redesign..

 

[DavinCarter]

Hmm, so I'm not the only one who doesn't care for the 12BY7. Heck it's a dang TV tube.

The 6SN7 was developed as a TV tube.
Many tubes developed for TV applications are great for audio.

 

[jaymanaa]

The 6SN7 was developed as a TV tube.
Many tubes developed for TV applications are great for audio.

I keed, I keed.

 

[s-petersen]

Not exactly. It acts as an extremely high impedance to AC signals. Sometimes on the order of 100M ohms or more. It regulates the DC current through the circuit, but acts almost as an open circuit to AC forcing the AC to balance between the two halves of the tube. If one goes up by a certain amount, the other has to decrease the same amount thereby having nearly exact balance at AC assuming the anode load resistors are identical.

Here is a good thread to read on the subject.

http://www.diyaudio.com/forums/tubes-valves/160312-balance-ccs-long-tailed-pairs.html

John

I think I'll start off simple,I can try that after it's up and running

 

[s-petersen]

Well I fired it up.(literally)kegger's changes aren't in yet(Parts on order),neither is feedback,or inrush limiting.I ramped up the input voltage with my variac,and everything seemed to work properly(no audio tests performed)I didn't have any outputs hooked up to test voltages around the amp.Well somehow I brushed against the input with my test lead and there was a small arc at the tube socket for one of the output tubes.(it blew a 5 amp fuse,I may have toggled power without realizing it)I shut it down, replaced the fuse and powered it up again,rechecked bias,and the tube with the arcing socket was not drawing any current.so I checked voltages.There was no voltages at the screen,and when I checked the plate,there was a small spark(a poor solder joint)but it killed my meter(first one in 20 years)I guess the sparking caused a high voltage spike to my meter.Anyway,I grabbed another meter after re-flowing the joint, and plate voltage was fine,but screen was 100v low,
I shut it down and checked the screen resistor,it looked fine,but was open.
My question is:If you lose the plate connection will it burn up,or overload the screen resistor? If the cathode resistor opens up will 400+volts go to the bias meter(if selected)?

I hooked it up to some test speakers,and without feedback it is over sensitive, I think it could play a magnetic cartridge at reasonable volume directly,
it also picks up hum with the bottom cover off,with it on it is very quiet. It's a little late to make noise tonight,I'll have to check it out tomorrow during the day.

 

[ccheath]

Did you put a load whether a resistor or a speaker hooked up before you power up? Tubes get fussy with unloaded output transformer.

 

[s-petersen]

I didn't have a load hooked up,I figured that with no input and no feedback loop there would be no output,hindsight, I should have put a resistor on it,as I accidentally gave it an input