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

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  

Paraphase inverter questions


 :
11/5/2004 1:48 AM
Greg Simon Paraphase inverter questions
Hi All. I have read and heard that the paraphase inverter that was used in some of the early to mid 50's fenders, and other amps, doesn't do a clean sound very well and doesn't give much headroom because the signals out of it aren't balanced very well. What I am wondering is if it would be a good match for an  
EF86 > cathode follower > EQ > marshall 2nd gain stage preamp, or if I should try something with a cleaner signal for the PI?  
 
I don't mind gain, because this is supposed to be a blues amp, but I would like some useable clean headroom. I don't want it to be like a Dual Recto or anything liek that for gain levels.  
 
The output section is a couple cathode biased 7868's, and I was planning on leaving the PI stock as it came. (amp was a Conn organ amp from '64) The PI stock is similar to the first Fender paraphase inverter, but has some different values for the resistors and caps, and has seperate cathode resistors.  
 
Thanks!  
Greg
 
Today
ampage
archive
And now, a word from our sponsors:


 
11/5/2004 8:25 AM
Dutch

Greg--  
That EF86 will likely have enough gain to drive a paraphase phase inverter in front of 7868s without the extra gain stage. In my" target="_blank">http://www.nyquist-plot.com/?content=stereo_tube_amp'>my stereo amp I have a single triode gain stage for each channel's preamp and I have plenty of gain with 12AX7s in the pre and PI to drive the amp to full power with a Strat (relatively low output pickups) and get output tube overdrive. 7868s (like their octal-based twin brother, the 7591) don't require a whole lot of grid drive. With the extra gain from the EF86, you might even have enough drive with a tone stack following it, if it isn't a very lossy one.  
 
If it were me, and I was contemplating going with your arrangement, I'd use a 12AU7 for the cathode follower and subsequent gain stage, and I might not even bypass the cathode on the subsequent gain stage. With a 12AX7 you'll have a ton of gain and will likely find that the thing does clean with the volume barely cracked open, and then goes into grind beyond that. If you decide to go this way, a master volume between the second gain stage and the PI or at the 7868 grids might not be a bad idea.  
 
HTH!  
 
C ya,  
Dutch
 
11/6/2004 12:41 AM
Greg Simon
Hey Dutch, thanks very much for the detailed reply! I was kind of woried after doing some reading that I was going to get too much gain there. The 7868's with the paraphase inverter were stock on the Conn, and I'm adding preamps to it. I've already wired up the socket in the preamp chassis for the EF86, so I'd rather stick with that if I can. Currently, I'm using a Vox Top Boost AC30 tone stack, though that may change. I don't have the room to add a master volume, so its either a 12au7 as you suggest, or go with a different PI. What is another PI that has a little cleaner signal that would be good to use there? Also, should I change the resistor values if I use a 12au7?  
 
The paraphase I have in this is a little different from the typical Fender, but I don't know what the differences make in the sound department. Where Fender has a 1M grid leak resistor coming into the inverter, this one has a 250K, and then a 33k series grid resistor also. The cap coming out of the plate is a .039uf. Then it goes through a voltage divider with a 470k at the top and a 12k to ground. Then into the other grid and out through another .039uf cap. The grid leak on this 7868 is a 470k, and there isn't one on the other 7868 because it is using the divider. The plate resistors on each side of the paraphase inverter are 220k's, and the cathode resistor on the first half of the PI is a 1.6k, and there is feedback injected there also. The other cathode is a 1.5k with no feedback. The PI is a 12ax7. So I'm not sure if this is better balanced than the older Fenders, but its different for sure! I appreciate your help.  
Greg
 
11/6/2004 7:19 PM
Dutch

quote:
"What is another PI that has a little cleaner signal that would be good to use there? "
 
 
Greg--  
A Cathodyne has a gain of 1, as opposed to the 25 or so you're probably getting from each side of the paraphase. If you opted for a gain stage and a cathodyne in its place, you'd gain (pardon the pun) a bit of wiggle room for trying out different preamp tubes, as the PI balance wouldn't be dependent on the PI tube's gain being in a certain range, as the paraphase is. Of course, the paraphase sound may be cool, depending on what you want out of the amp. I'd go with a 12AU7 for the cathode follower driving the tone stack and the gain stage thereafter, and leave it unbypassed, to cut the gain down a bit.  
 
quote:
"Also, should I change the resistor values if I use a 12au7?"
 
 
In the paraphase PI, I'd stick a 50k pot in series with the 12k resistor, between it and the 470k, and move the grid wire for the second half of the paraphase up to the pot wiper, so you can dial in the amount of signal needed to balance (or slightly unbalance!) the PI outputs to your liking.... The plate and cathode resistors sound like they'd work ok, although you might want to parallel both of the 220k plate loads with 100k resistors if you find it doesn't have enough headroom. 7868s require so little drive that you'll probably get away without it, though....  
 
HTH!  
 
C ya,  
Dutch
 
11/8/2004 10:48 PM
Anonymously Anonymous Question Dutch
Hey Dutch, I was wondering where it is that shows you how much drive different tubes need? Like you mentioned the 7868 doesn't need much...where do I find that spec? I've also heard the EL84 doesn't need much, but I have no idea where to find that.  
Thanks!  
Greg
 
11/9/2004 4:17 AM
kg
look at the effective bias voltage (ie, the voltage from cathode to g1), and there's your peak drive voltage.  
 
hth  
ken
 
11/9/2004 1:00 PM
Dutch

quote:
"look at the effective bias voltage (ie, the voltage from cathode to g1), and there's your peak drive voltage."
 
 
Thanks for the pinch-hit Ken! :^)  
 
C ya,  
Dutch
 

  Page 1 of 3 Next> Last Page>>