ampage Tube Amps / Music Electronics	For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Re: Building a 'clean' PI

6/5/2000 9:55 AM
Chris Harden	Re: Building a 'clean' PI Ray, Thanks for giving me the correct name for that circuit. I found it as an example circuit in a tube data book, the actual circuit was shown with a 12AX7, so I changed some of the values to use with a 12AT7. Would it be possible for you to run the sim with a 12AU7, and possibly tell me any resistor value changes that would be needed. I could really do with a copy of that program, I have a simple demo SPICE prog, but it's very difficult to get meaningful info. Any ideas on how much a copy is? probably too expensive for me, but worth a look. Another quick question on this circuit, can you still use the spare grid as an aux input, for FX returns etc? What are the best values to use for the 'balancing potential divider'network? Thanks for all your help. Chris

6/5/2000 9:55 AM

Chris Harden

Re: Building a 'clean' PI

Ray,

Thanks for giving me the correct name for that circuit. I found it as an example circuit in a tube data book, the actual circuit was shown with a 12AX7, so I changed some of the values to use with a 12AT7.
Would it be possible for you to run the sim with a 12AU7, and possibly tell me any resistor value changes that would be needed. I could really do with a copy of that program, I have a simple demo SPICE prog, but it's very difficult to get meaningful info. Any ideas on how much a copy is? probably too expensive for me, but worth a look.

Another quick question on this circuit, can you still use the spare grid as an aux input, for FX returns etc? What are the best values to use for the 'balancing potential divider'network?

Thanks for all your help.

Chris

6/5/2000 2:44 PM
Ray Ivers	Chris, TubeCAD is available from glass-ware.com for $49.00, and TubeCAD and Audio Gadgets (a handy program for miscellaneous amp-related calculations of all kinds, from RC filters to power supplies) together are available for $79.00. I can't say enough good things about TubeCAD; it's not perfect, but I can't imagine designing without it now. I've been hassling John Broskie to write a push-pull output stage TubeCAD add-in, but he says he just ain't got the time right now (evidently the pentode modeling's a bear). Here are the 12AU7 numbers for 300 and 550 volts: 300 V: Plate 27K, cathode 560 ohms, 10.2 mA -105/+134 P-P AC, 795 mW plate dissipation 550 V: Plate 47K, cathode 1500 ohms, 10 mA -220/+224 P-P AC, 1.5 watts PD The Ampeg B15S (Groove Tubes book) uses an 820K/1M voltage divider for a 12AU7 driver with 47K plate resistors, 1.5K cathode, 405 V B+. I think I would probably use 1M and 910K resistors with an additional 250K variable resistance in series with the 910K for AC balance adjustment; that way, you have all sorts of balance/imbalance options at your disposal. I should think you could use the second grid (I don't think I would call it a 'spare') as another input; the common unbypassed cathode resistor should insure the signal would be seen equally at both PI plates. I don't think I can recommend this, though; even with a 12AX7 driver the gain just isn't there for anything other than a super-hot effects return. Ray

6/5/2000 2:44 PM

Ray Ivers

Chris,

TubeCAD is available from glass-ware.com for $49.00, and TubeCAD and Audio Gadgets (a handy program for miscellaneous amp-related calculations of all kinds, from RC filters to power supplies) together are available for $79.00. I can't say enough good things about TubeCAD; it's not perfect, but I can't imagine designing without it now. I've been hassling John Broskie to write a push-pull output stage TubeCAD add-in, but he says he just ain't got the time right now (evidently the pentode modeling's a bear).

Here are the 12AU7 numbers for 300 and 550 volts:

300 V: Plate 27K, cathode 560 ohms, 10.2 mA
-105/+134 P-P AC, 795 mW plate dissipation

550 V: Plate 47K, cathode 1500 ohms, 10 mA
-220/+224 P-P AC, 1.5 watts PD

The Ampeg B15S (Groove Tubes book) uses an 820K/1M voltage divider for a 12AU7 driver with 47K plate resistors, 1.5K cathode, 405 V B+. I think I would probably use 1M and 910K resistors with an additional 250K variable resistance in series with the 910K for AC balance adjustment; that way, you have all sorts of balance/imbalance options at your disposal. I should think you could use the second grid (I don't think I would call it a 'spare') as another input; the common unbypassed cathode resistor should insure the signal would be seen equally at both PI plates. I don't think I can recommend this, though; even with a 12AX7 driver the gain just isn't there for anything other than a super-hot effects return.

Ray

6/5/2000 7:27 PM
Chris Harden	Sounds like a 12AU7 would be OK. Would the higher current of the 12AU7 compared with the 12AX7 help if the output valves started drawing current? I'm using equal values for the Potential divider at the moment, I thought theoreticly that would give the best balance. Time to stop thinking and get the scope out and play. Incidentally, my Celestion silver series arrived today, no more crappy 6in car-boot sale speakers! All I need now are those OT's! Chris

6/5/2000 7:27 PM

Chris Harden

Sounds like a 12AU7 would be OK. Would the higher current of the 12AU7 compared with the 12AX7 help if the output valves started drawing current? I'm using equal values for the Potential divider at the moment, I thought theoreticly that would give the best balance. Time to stop thinking and get the scope out and play.

Incidentally, my Celestion silver series arrived today, no more crappy 6in car-boot sale speakers! All I need now are those OT's!

Chris

6/5/2000 8:56 PM
Chris Harden	I've just modded the PI to use a 12AU7 with 27k plate loads and 470R common cathode. It still seems to be running a little cold, but it sounds far better. I think I've now got the O/P tubes overdriving before the PI does, which was one of the main objectives. The combination of those changes and the new speaker have created a almost descent sounding amp. More fidling needed, but I'm on the right track. Thanks again Chris

6/5/2000 8:56 PM

Chris Harden

I've just modded the PI to use a 12AU7 with 27k plate loads and 470R common cathode. It still seems to be running a little cold, but it sounds far better. I think I've now got the O/P tubes overdriving before the PI does, which was one of the main objectives. The combination of those changes and the new speaker have created a almost descent sounding amp. More fidling needed, but I'm on the right track.

Thanks again

Chris

6/5/2000 11:04 PM
Ray Ivers	Chris, I'm replying to both your posts here. You really can't use equal values for the potential divider, Chris; you have to have some signal present at the second stage's grid, and this signal is present only if the two resistors are slightly unequal. Does the scope show identical output at each plate? The 12AU7 will be much better than a 12AX7 at driving the output tubes into the grid current range; unfortunately, small-value capacitors generally don't excel at passing AC current, so your gains will be limited compared to transformer or direct-coupled cathode follower grid drive. The 12AU7 is pulling an acceptable amount of current, I think - more than 20 percent of max, which is what TubeCAD considers the minimum for good performance. Ray

6/5/2000 11:04 PM

Ray Ivers

Chris,

I'm replying to both your posts here. You really can't use equal values for the potential divider, Chris; you have to have some signal present at the second stage's grid, and this signal is present only if the two resistors are slightly unequal. Does the scope show identical output at each plate?

The 12AU7 will be much better than a 12AX7 at driving the output tubes into the grid current range; unfortunately, small-value capacitors generally don't excel at passing AC current, so your gains will be limited compared to transformer or direct-coupled cathode follower grid drive. The 12AU7 is pulling an acceptable amount of current, I think - more than 20 percent of max, which is what TubeCAD considers the minimum for good performance.

Ray

6/6/2000 5:37 PM

Chris Harden

I've finally moved the scope and sig. gen. from the bedroom (old workshop) to the garage (new workshop), in order to test my PI. I haven't changed any components since the previous posts, but the O/P tubes are now EL34's.

The balance with equal value resistors is pretty good. I'd say the lower one is no less than 80% of the higher. I don't think the difference is enough to worry about, It might just be enough to give a little bit of second harmonic distortion.

quote:
"The 12AU7 will be much better than a 12AX7 at driving the output tubes into the grid current range; unfortunately, small-value capacitors generally don't excel at passing AC current, so your gains will be limited compared to transformer or direct-coupled cathode follower grid drive. "

Would making the caps larger at all make a difference, or would I leave myself open to blocking distortion?

Chris

<<First Page

<Prev

Page 2 of 2