• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

page Eli Duttman's Simple p-p with 5965 MQ EXO-173 and 6AQ5

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello Eli,

I was wondering if you or anyone else has experiemeted with the MQ EXO-173 phase splitter in your simple p-p amplifier design?

My thought was to substitute the 12AT7 driver with a 5965 and splitting phase with EXO-173 to a pair of 6005's.

Any ideas or help would be appreciated?

Norris
 
hey-Hey!!!,
That choke( or more exactly half of it ) is a bit more load than I think a 5965 would be happy with. You have some options given your output tube choice. You can drive the choke with one section as a cathode follower. The 6AQ5's are not going to need a lot of swing and the cathode follower should be far more comfortable with the choke.

There is one more consideration, that choke seems to do best with a large coupling cap( ~5 uF ), so you may wish to avoid the use of such large caps and do phase splitting with a tube circuit. Again, the small drive requirements are working in your favour, even a split-load will work here( even with Class A B+ levels ).
cheers,
Douglas
 
You're referring to the highish Rp of the 5965 (compared to, say, a WE 417a)? If so, I had my EXO-173s done with nickle with the understanding that this will open up use of drivers with comparatively higher Rp. So that might be an option for Norris. Another option would be to use the driver specified in the SETH design. IIRC the SETH uses a 3uf cap, not 5, but the point about a big cap in there is well taken. I will say, however, I built a SETH variation using a 417a --> 3uf --> driving 6550's and the amp received great reviews. I felt it was a very musical and powerful sound, it was my first venture into using a transformer phase splitter (well, autoformer) and it worked very nicely. Elegantly simple.
 
Thanks Douglas, Eli and Steve,

Douglas:

I agree that a large value capacitor in the signal path would be a compromise, even 1uF seems large.

Looking at the Electra-Print EL84 p-p circuit using their SMA-2 splitter choke and 6J6 driver. I see that it only require a 1uF coupling capacitor. I was curious why they can get away with using a driver tube that has an Rp of 7.1k ohms, versus 1.8k of the 5842?
Where the 5965 has an Rp of 7k.

http://www.electra-print.com/el84psapp.php

Would using a quality coupling transformer like the Lundahl LL1660S for push pull with a differential driver stage be a better solution over the choke, or vacuum tube phase splitter with a SE driver to p-p?

If so, is there a circuit available that I might be able to easily adapt to use with the 6005?

Steve:

If using the MQ EXO-173 with nickle, what other driver tubes from your experience might work well with the 6005, but require a much smaller value of coupling capacitor than 3uF?

Thoughts from Eli about how to make the 5965 work with the EXO-173:

"Buffer the 5965 common cathode voltage amplifiers with DC coupled IRFBC20 MOSFET source followers. Build a reasonably robust B+ supply and run the FETs with 10 mA. of drain current. Believe me, the O/P impedance will be LOW."

And finally, would anyone of you have suggestions of the best way to split phase, either with iron, vacuum tubes, differential, or SE to p-p?

Any and all suggestion will be appreciated.

The simpler the better for my knowledge level.

Thanks

Norris
 
Norris,
The inductance of that half of the coil sets the coupling cap size. Then the tube has to live with that low( numeric) load at LF. I think the ability of triodes to live with eliptical, reactive load lines is a bit overstated...they just sound better when the line is straight.

I use 1.5 uF coupling caps in my big amp and it has a balanced input to the grid choke. That choke is ~700 Hy and has a gap so that the inductance does not vary too much with signal. Each phase works into ~350 Hy. My driver/LTP is a 6H6 cascode with 15k plate loads. I'd rather go for 10k but that would put my OPT even further over the edge in terms of idle current through the primary( it is E-Linear rigged ).
cheers,
Douglas
 
Thanks Douglas,

I will study your amplifier arrangement and see if I can figure out how to impliment it to my needs.

Bas,

Yes, I thought about that after I posted, but did not have the edit option available. Or, did not figure out how to execute it.

Eli Duttman's amplifier was the inspiration for this thread.

Could you change the title for me, or tell me how to edit it myself?

Duh!

Thanks

Norris
 
Norris,
One more thing, the balanced drive to the grid choke does better without an interleaved choke. The additional capacitance or perhaps the coupling itself works against you. Seemed counter intuitive( matter of fact so counter I went to the trouble of doing the interleaved experiments and found the negative result ). The 173 iron phase splitter has not enough inductance in half its coil to be a pleasant load just in case you needed another issue to overcome.
cheers,
Douglas
 
The thing is. Once you start using something like the MQ EXO-173 it won't be an El Cheapo anymore. Or does he have another design with the EXO-173 somewhere? Anyone have a link?

But ofcourse the EXO-173 is one way of keeping the phase splitting simple.

And Douglas is correct that you'd need something strong to drive the choke or a transformer for that matter. I noticed John Atwood's Artemis Labs DP-2 amp using a KT-88 for that. Since I saw that I've been thinking of driving my EXO-173's with a KT-88 as well. :)
 
My experience w/EXO-173 has primarily been limited to driving it with a 5842 (417a) at (IIRC) 12 or so ma. Did an outstanding job for me (again, mine is nickle core). However, the opportunity to drive it with a monster like a KT88 (triode I assume) would be fun. Maybe we should explore the Sakuma style of similar tubes throughout - say a 2A3 driving PP 2A3?
 
Bas Horneman said:
The thing is. Once you start using something like the MQ EXO-173 it won't be an El Cheapo anymore. Or does he have another design with the EXO-173 somewhere? Anyone have a link?

But ofcourse the EXO-173 is one way of keeping the phase splitting simple.

:)


Hi Bas:

I agree with your statement about "won't be an el cheapo anymore" with the proposals that have been floating about. It would be an excellent starting (constructing) point just as published.

What discussion of the EXO-173 that has been discussed here has been fraught with misinformation. The EXO-173 is very simple to implement and yields excellent performance as a phase splitter if used properly. Orvelle Labs ran a lot of tests of the EXO-173 on the Audio Precision testing system--- this unit does well with source impedances up to 5 kohms. We recommend that the r sub p of the tube driving it be btwn approx 2K and 5K.

with regard to the blocking cap size---- this will vary depending on the core material with which the 173 is built and what low frequency cut-off point you choose.

But---- once again--- I see the capacitor boogey man appearing in full dress---- yes, you must use a blocking cap with the 173--- it's size will range perhaps from 1 or 1.5 uf to 4 to 6 uf--- again depending on core material and low frequency that your specifying.

But--- keep in mind that a conventional two winding phase splitting interstage---- whose power supply is in series with the primary of the interstage---- that you will generally have a much larger cap in the signal path (your last power supply cap or decoupling cap) than your likely to use with the 173 phase splitter. And that last power supply cap can also resonate with the L of the primary in the series fed arrangement.

almost always--- parafeed implementations use smaller caps than any of the common series feed alternatives. And use of a LCR calculator can provide a good first approximation of what cap size might be required in the parafeed arrangement.

In terms of freq response and phase response---- the EXO-173 is actually quite good. Amplitude response is within a few tenths of one db out to 20 khz and phase response differential at 20 khz is in the worse case two degrees. John Levreault tested a UTC phase splitter alongside the 173 and reported that the 173 had better freq response as well as better phase response than the standard bearing UTC phase splitter.

the best way to learn about the 173 and some of it's applications would be to visit the MQ forum and do a search or pose your queries there.

As Bas and others have pointed out--- iron phase splitters aren't the economic way to do things.

Eli's circuit uses and LTP---- LTP's can be quite, quite good. And if you want to use iron--- LTP's loan themselves well to LC coupling btwn stages---- here is another example to look at;

[http://www.audioasylum.com/forums/magnequest/messages/7472.html[/url]


now--- if we are getting into L as in horsepower races---- the BCP-16 stacked with the appropiate nickel will have far greater than 4,000 henries each. So it is extremely easy to drive and has extremely low winding C (less than 25 puffs).

and you can check out CT grid choke scenario at the following url;

[http://www.audioasylum.com/forums/magnequest/messages/8025.html[/url]

bearing in mind that if you used triodes (like the 5965) you would not need the screen taps shown in the CT grid choke.

there are a thousand ways to use inductors--- and a thousand more ways to eliminate their use---- and a gizzillion ways to implement a multi-stage amp. The trick or beauty is in learning how to opimize for and within the particular topology which is chosen.

Eli's implementation has a lot going for it--- good support from Eli---
it's a design that is fully developed and refined by the designer.
Taking EIi's circuit--- and modifying it--- as Bas says--- it won't be Eli's circuit any longer.


MSL
 
smbrown said:
Mike, can't the EXO-173 be used as a center tapped grid choke (sans UL taps)? And/or, when will ct grid chokes be available (and price)??


Hi Steve:

In addition to use as a phase splitting autoformer--- the 173 can be used as a center tapped grid choke---- the EXO-173 is actually quite flexible and can also be used as a two terminal grid choke (i.e., single ended). It can also be used as a 1:2 step-up and a 2:1 step-down. All the while excluding DC.

I can in many cases accomadate and build a custom center tapped grid choke--- if I know your application details.

Personally, I like to use the BCP-16 grid choke after a LTP wherein the phase splitting has already been done---- and now you can source and use two of the aforementioned grid chokes--- without them interacting with each other--- and in two small packages you can realize a very high load impedance gained from each of the chokes being in the kilohenry range inductance wise. And the cost of two BCP-16's is cheaper than say a single EXO-173 (if core materials are kept constant).

Out of curiosity--- do you have your schematic drawn up? If it
is not proprietary--- I and perhaps others would enjoy seeing how you implemented the 173.

MSL
 
MQracing said:



there are a thousand ways to use inductors--- and a thousand more ways to eliminate their use---- and a gizzillion ways to implement a multi-stage amp. The trick or beauty is in learning how to opimize for and within the particular topology which is chosen.



to illustrate the above---- here is an example of the LTP being followed by a 1 to 1 center tapped two winding series fed interstage transformer. And of course if your LTP consists of triode tubes you would forego the screen taps on the primary side of the push-pull IT.

http://www.audioasylum.com/forums/magnequest/messages/7999.html

On first blush it appears that there are no caps in the signal path--- but look closely at the ac path to ground and notice the 10 uf cap shunted to ground to provide the ac signal a "low impedance" path to ground. This 10uf cap is IN the signal path.

none-the-less this is yet another schema that is possible. And it would support some circuit arrangements very well--- for instance if you wanted to use fixed bias in your output stage. And the primary of this two winding interstage could be thought of functionally as a ct grid choke in terms of it's interaction with the plates of the tubes driving it.

Different iron for different purposes and ends.


MSL
 
Guys,

Was talking to Mikey this morning and he asked me to chime in.

Engineer is the diminishing of tradeoffs of a sort. Sure you are using a larger cap here but what are you gaining?

Well first off say you used something like 220K grid resistors and 0.1uF caps and a tube phase splitter. Well as the signal approached 60% that 220k resistor is going to start to draw current and rebias the tube.

Something like the 173 removes the need for the phase splitter and also has a low dcr which means you will get considerably more power out of the design, because when the grid starts to draw current the tube won't rebias.

In my use of this part I would suggest using a SRPP too drive it. First the El84 needs very little input voltage for full output. I would use either the 6SN7 or the 9 pin version 6FQ7/6CG7 type tube.

Another thing to remember about the EL84 is it has tons of gain. Noise is always a factor with this tube.

Thanks
Gordon
 
MQracing said:




What discussion of the EXO-173 that has been discussed here has been fraught with misinformation.

now--- if we are getting into L as in horsepower races---- the BCP-16 stacked with the appropiate nickel will have far greater than 4,000 henries each. So it is extremely easy to drive and has extremely low winding C (less than 25 puffs).

MSL

Where is all this misinformation?

Keep in mind that your claimed L and C_winding will yeild a self resonant point around 500 cps. Above that it behaves like a capacitor.
cheers,
Douglas
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.