 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
75.71.207.145
In Reply to: RE: CCS Questions - Part II posted by JimBop on September 23, 2009 at 12:29:50
Here's the schematic:
Follow Ups:
Jim, while am sure that is a fine schematic, that is not even remotely close to the George Wright WPA 3.5 schematic dated 1/26/98 from Wright sound Company. The one I hold is hand drawn by George himself give to me by Hug from Seattle, Washington. Out of respect I will not post that schemo here. There are numerous differences and the one you hold is not even remotely close.
Kind Regards,
Ken
The Mind has No Firewall~ U.S. Army War College.
Congrats Cleantimestream on your having the original WPA 3.5 schematic, but why insist on stating it everywhere if you have no intention of sharing it with the community?
Many owners of WPA 3.5 (including myself) have re-created the schematic of their amps and have no problems sharing it, after all it is not some secret Butler Audio black box.
Jim,
Apart from the CCS replacing a resistor in the driver stage, and the errors you noted, is this in fact the schematic of the Wright Sound Mono 3.5 with hum pot installed? That would be helpful for future searches, since George Wright reportedly never wrote his schematics down before he died.
One other minor correction: I think there must be a jumper from the center tap of the +300-0-–300 winding to the CT of the 5V rectifier filament winding.
Yes, there is a center tap. However, with the hum pot installed, the center tap is disconnected.
...as I know that is disconnected. I am referring to the B+ and rectifier windings. The "connection" can be through a shared ground, see Tre's post. It is not a practical concern, but just for the sake of logical completeness.
Is this the Wright amp? I had thought George used an unbypassed cathode resistor on the first stage but maybe he changed the design from time to time. My WPA 3.5's appear to need about 5V RMS to reach full power, so they have even less gain than what was predicted for your circuit.
"I think there must be a jumper from the center tap of the +300-0-–300 winding to the CT of the 5V rectifier filament winding. "
?
The 300-0-300 winding's center tap must be grounded. But it would not be connected to the 5 volt winding's center tap even if the 5 volt winding had one.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
The connection I was referring to would indeed be through the grounds, as you mention. Also, you are right that it is not a "jumper" because the current has to go through the filter caps and load to the cathode side of the rectifier! And it is effectively only a virtual center tap to the rectifier cathode and its filament winding. (I assume it would only connect to one side if an indirectly heated rectifier was in use.)
My main point was that the amp would not work without some connection to the 300-0-300 CT, and Jim didn't show that on his schematic.
.
Have Fun and Enjoy the Music
"Still Working the Problem"
JimBop,
Let me write something of a different perspective - I do understand your circuit. To annunciate, it is a follows: a resistively plate loaded triode, the first half of a 6sn7 comprised of pins 4,5, and 6, direct coupled to a cathode follower, likewise the second half of the 6sn7 comprised of pins 1,2, and 3. The cathode follower is capacitively coupled to the 2a3 output stage. 'Nuff said.
Basically, your question goes to loading the first stage with a CCS, and the requisite limits thereof - in effect replacing the 100k plate load resistor. The answer is as follows. You measure the voltage across the cathode resistor, from pin 6 to ground and divide by 2k to arrive at the current through the first stage. Then measure the voltage between pins 6 and 5. Finally multiply this current times this voltage to arrive at the power dissipation of the first section. It must not exceed the power rating of the first section. I'm not gonna reach for my tube manual - you can look that up.
Further, you need look at the total dissipation of the 6sn7. Again, consult the tube manual. Adding the power dissipation arrived at before for the first section, do the same for the second section. Add these together and compare with what the manual says.
Also, look at the pate to cathode voltage rating and make sure you're not exceeding that. In this case with the supply voltage mentioned, you're likely not.
So long as you're with these aforementioned limitations, anything or operating point is fair game. Adjust your CCS to taste...
I'll go on to suggest the following for thought in doing so. When you load a tube with a CCS, you'll want twice the voltage swing desired from the stage dropped voltage across the CCS. That's my rule of thumb anyhow.
You're not going to realize any gain from the cathode follower, so in this circuit all the gain for swinging the grid of the 2a3 need come from the first section of the 6sn7. I'm thinking you have a preamp or a hot input source...
You'll want to look at the voltage across the cathode resistor (2k) and make sure this voltage exceeds the peak ac input signal voltage.
Some numbers...the mu of the 6sn7 is 20 and you'll realize this with the CCS. So, assuming something like 50 volts of bias on the 2a3, 50 divided by 20 is 2.5 volts. You need at least 2.5 volts across your cathode resistor (2k), and this reflects the input sensitivity of your amplifier.
The swing would be twice the bias on the output stage, or 100 volts in my example, meaning and applying what I said that you would want at least 200 volts across your CCS.
That really doesn't leave much across the tube....
Now, enough of Ohm's law. Let's discuss some more advanced concepts. The distortion spectra of the input stage and cathode follower are in phase in your amplifier. With two plate loaded triodes in succession, they are out of phase and you can use them in contravention to each other for distortion cancellation. That's not available to you. Further, you can often run two successive plate loaded stages of the same type at relative opposite ends of the operating range, getting different complementary sonic signatures, if you will. This is somewhat available to you.
To wit, when two stages are direct coupled, they act more like a composite single stage than they do as two distinct stages that are resistive coupled, for example. Generally, I've found that direct coupling the driver and the output stage is superior to direct coupling an input and driver stage and then RC coupling to an output stage. Just my experience...
Pay attention to the first and second stage talking through the heater and try and keep this potential difference to a minimum by looking at the relative cathode voltages of the first and second stages.
The cathode follower (CF). Contrary to what many believe or understand, while CFs are good in terms of impedance matching (the purpose of the CF in your amp is to provide a lower source impedance for driving the grid of the 2a3), they are not particularly good sounding when loaded down. IMO, 47k in combination with the miller effect of the bi-plate 2a3 is getting there in terms of loading....
What about leaving the 100k on the first stage and using the ccs in the cathode of the CF? In other words, replacing the 30k resistor from pin 3 to ground??
Few other things, I'd ditch the grid stoppers and take a good hard look at you're power supply - there's some caps there and other stuff that make me go hummmmm...
Gotta run, RMAF and all, HTH,
LD Moore
Just a quick note. I have had the experience of a 2A3 oscillating badly without a grid stopper. Adding a small stopper (220 ohm carbon composition IIRC) cured it immediately. I now use them on anything with more than 2000 micromhos of transconductance, and I'm cautious about checking the rest with a good RF 'scope.
Hello Paul,
Listen to 120 Ohms next on the 2A3, and compare it to 220 Ohms, by ear.
Jeff Medwin
I've been meaning to ask this question: In my amps, which are the grid stoppers?
JD
The 4.7K on the input tube and the 220 ohm on the 2a3.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
That's what I thought. Any recommendations on what type of resistors work best in these positions?
Thanks again.
Paul just recommended carbon composition, IIRC brand.
Make sure the Mosfets in the CCS has them for each gate also. I use 1K there.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Sorry to drop in unannounced.
One suggestion. Try to make the length of grid stopper resistors to the sockets as short as possible. This helps keep the amp stable, by minimizing the lead capacitance. You prolly already knew this, but I felt obliged to buzz-in. Little angel on my shoulder told me.
:^)
Yeah, I'm still trying to think how to put them inside the glass. :-)
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Can you drop me an email using Asylum email? I have a question regarding a possible trade. Thank you.
Larry,
Thanks for the info. It will take me quite a while to understand it all.
Jim D
I didn't even see the CCS in the power supply.
That's a whole different thing. A power supply thing.
I thought you wanted to load the plate of the tube with a CCS because it makes the tube have more gain and have less distortion.
What you are doing there is different, with a different purpose.
Sorry.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
I did want to load the plate of the tube. I misunderstood which resistor was the plate load resistor.
There are two mistakes on the schematic:
- The 100K resistor from pin 1 of the 6SN7 should connect to the next line to the right (the output of the CCS).
- I missed a jumper and 522K resistor between pins 2 and 6 of the 6SN7.
I don't have time to scan and post a corrected version right now.
Jim
OK, what you have there is a gain stage, pins 4,5 and 6 followed by a cathode follower, pins 1,2 and 3.The gain stage is direct coupled to the cathode follower stage.
The 30k resistor is the cathode resistor for the cathode follower.
You can do this two ways.
Replace the 100k plate resistor with a CCS. The easy way to do this is, with the circuit hooked up as shown, measure the voltage across the 2k cathode resistor of the gain stage.
Do the math, current = voltage / resistance.
Now you know how much current the original gain stage was drawing. Measure the plate voltage while you're at it.
Replace the 100K plate resistor with the CCS and set the CCS, by measuring the voltage across the 2K cathode resistor so that voltage is the same as before.
The other way you can do this is forget about the second half of the 6sn7 and take the signal off the mu output of the constant current source. That would be at the top of the current set resistor instead of the plate of the tube.
This will give a low output impedance. Not as low as the output of the cathode follower but low enough.
One advantage of not using the cathode follower is, you can now set the gain stage to whatever current and voltage you want. You can really find the "sweet spot" without (as many) constraints.
BTW If you keep the cathode follower make sure the plate voltage of the gain stage ends up the same as it is now. You can adjust the plate voltage of the gain stage without changing the current of the CCS by changing the value of the 2K cathode resistor. The larger the resistance value the higher the plate voltage will be, the lower the resistance value, the lower the plate voltage will be.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 09/24/09 09/24/09 09/24/09 09/24/09
This is one driver topology that I know little to nothing about, so perhaps someone more knowledgeable could help you out. I think both Tre' and I were assuming you were talking about a conventional triode gain stage. That'll teach me to give advice, without getting all the facts straight. Good luck and have fun.
twystd
Post a Followup:
| FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
||||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
