MQ passive preamp impressions (Ingot)

johnsonad · 18024

0 Members and 1 Guest are viewing this topic.

Offline johnsonad

  • Hero Member
  • *****
    • Posts: 1670
on: May 17, 2012, 12:43:10 PM
This wasn't a planned project but one that turned out surprisingly well.  Initially I was looking to aquire one of Mike's P&G pots for a future preamp project (maybe the 300B?).  Reading a few posts from the AA MQ forum it seemed that these were a great match for Mike's 10k/10k B7 iron.  So a deal was cut and Mike sent me a pair of the 10k/10k HN trannies and a P&G pot.  After a little soldering I had a home brewed Ingot passive preamp for the system.

I'm currently suffering from too much gain and the thought was a passive might work well at this point.  All of my sources have a low output impedance (minus the Eros) which work well with the Ingot and the little old Quad 303 of which I just rebuild has a 220k input impedance.  Speakers are Altec Model 19's.  They are very sensitive but I've found they are not the easiest to control well.  For some reason the 303 does a very good job of paring with these speakers which is great as it was cheap.

As Xavier had posted a year or so ago, I'm happy but at a loss as to how this passive can sound so good.  In my system and to my ears there is an improved clarity, better resolution and detail and improved imaging.  This passive is going to stay in the system for a while until I get a good feel for the sound (or lack of) of it.  So far I'm happy with the results.  If you happen to have a pair of these transformers in your stash and a 10k volume pot, I recommend you give this design a try if your system impedances can handle it and you are suffering from too much gain as I am. 

Aaron Johnson


Offline mqracing

  • Moderator
  • Jr. Member
  • *****
    • Posts: 26
Reply #1 on: May 18, 2012, 04:00:10 PM
Hi Aaron.  Thanks for the post.

For a handy reference--- and some ideas of circuit possibilities check out this little
blurb Six Moons published back in 2005.

http://www.6moons.com/industryfeatures/mq/mq.html

I too really like the sound quality of this simple passive. But it is also possible to
add a buffer (tube or solid state) after the input trans\attenuator combination if,
for example, you wanted to drive very long cables. 

here's another mention of the Ingot in Six Moons...
 
"...readers should also know about Michael LeFavre's Magnequest Ingot designed
in collaboration with Steve Eddy. Their iteration of a minimalist preamp circuit uses
a small input transformer to inductively stabilize impedance seen by the source
while a 10K pot across the secondary handles passive attenuation. Another Ingot
version adds a FET buffer."

the "magic" is in using a good quality 1:1 transformer--- I especially like (at the text
suggests) using inductive loads and making them massive impedance wise--- and then
using a simple resistive element to do the job we actually want to do in the most
straightforward manner possible.   It's the right approach--- and it sounds right as well.

For experimenters---- here's a neat little "trick" to try out---- try it and listen to it--- don't
ask for me an elegant technical explanation of it's funtioning--- just try it.   Since the B7
10K to 10K is a 1:1 turns ratio--- take and parallel the two windings to each other---- hook
up dot to dot and non-dotted terminal to non-dotted terminal.   Now you have a simple choke...

you may have seen (though rare) choke loaded output stages--- now try a choke loaded input
stage--- put your P&G RF-15 across the paralleled windings as you "normally" would.  Remember
the dot is hot.  Give it a listen and report back.

I'd be interested in learning of anyone's "ears on" experience especially in contrasting the sound
btwn "choke" and "two winding trans" implementation.

Of course you will lose the isolation btwn source and load if you implement the B7 10K:10K as a
choke--- but other than that I really liked it when I gave it a try.

MSL







Offline Paul Joppa

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 5844
Reply #2 on: May 18, 2012, 04:58:36 PM
Here's another one to try. IF you need some gain, and IF you need to drive long or low impedance lines, or maybe balanced lines, see if you can borrow a Smack to follow the Ingot. Keep the interconnect between them short and high quality, and perhaps disconnect the Smack's gain pot (assuming you Ingot has a higher quality attenuator). I expect it would sound very good, but it's not an experiment I'm going to get around to any time soon - if anyone does, I'd sure be interested to know what you hear.  :^)

Paul Joppa


Offline johnsonad

  • Hero Member
  • *****
    • Posts: 1670
Reply #3 on: May 18, 2012, 10:24:28 PM
Thanks Mike and Paul.  Fun stuff for sure!

Mike,  I may have wired it odd regardless.  I have hot to dot in and hot to dot out on the transformers.  Is this the choke loaded fashion or standard fashion? 

Paul, I have WAY too much gain in the system but will consider that in the future if I need more gain.  I'm at about 9 o'clock on a passive with room filling volume ;)

Regards,

Aaron
« Last Edit: May 18, 2012, 10:26:19 PM by johnsonad »

Aaron Johnson


Offline mqracing

  • Moderator
  • Jr. Member
  • *****
    • Posts: 26
Reply #4 on: May 18, 2012, 10:46:31 PM

Mike,  I may have wired it odd regardless.  I have hot to dot in and hot to dot out on the transformers.  Is this the choke loaded fashion or standard fashion?  



I'm not sure that I understand from the above how you have it hooked up.

The standard hookup goes as follows. One end of the coil will have the product ID.  Call that the primary.  It will
have two solder lugs.  One with a dot, one with no dot. The terminal WITHOUT the dot is going to be your ground.
The dotted terminal will be the "high AC"  or the hot terminal.  

The other end of the coil--- with no product ID---- call that the secondary.  Same thing.  The non-dotted terminal
is going to be your ground.   The dotted terminal will be hot.   You will put your potentiometer\atttenuator across
the secondary side of the transformer.  


To "convert" it to a choke you would use a small jumper cable and connect the two dotted terminals together and
then (separately) use another piece of jumper cable and connect the two non-dotted terminals together.  You have
now paralleled the two physical windings---- and electrically you now only have one active winding--- hence, like a
traditional choke.   You still want to use the convention of the dotted terminal is AC HOT and the non-dotted terminal
is your ground and put your pot\attenuator across this single winding in the proper orientation.

hope this helps,

MSL
« Last Edit: May 19, 2012, 07:28:57 PM by mqracing »



Offline johnsonad

  • Hero Member
  • *****
    • Posts: 1670
Reply #5 on: May 18, 2012, 11:13:16 PM
Thanks for the reply Mike.  I have it wired correctly :)  I'll put a couple jumpers on it sometime in the future once things have settled and I get a good feel for the sound.  Today I removed the switch and wired it for single source in and out. 

Recording after recording I'm impressed by the preamp.  It is ruthlessly detailed and a there is no masking a bad recording.  A good recording sounds especially good now.  I was considering finding a home for my Altec's until this combo came along.  I'm continually surprised at how these speakers can scale up or down with the gear they are paired with.

Mike, these are HN transformers.  Does the lamination material make a difference in a 1:1 transformer?  If so, what would the difference be between the HN and say cobalt or M3/M4 sound wise?

« Last Edit: May 19, 2012, 06:26:45 PM by johnsonad »

Aaron Johnson


Offline mqracing

  • Moderator
  • Jr. Member
  • *****
    • Posts: 26
Reply #6 on: May 20, 2012, 08:41:33 AM

Mike, these are HN transformers.  Does the lamination material make a difference in a 1:1 transformer?  If so, what would the difference be between the HN and say cobalt or M3/M4 sound wise?



Does the lamination material make a difference? From a technical POV, yes. The higher perm materials will produce a greater
amount of L at any given drive level.  Hi perm materials also tend to have less core losses.

But.... if you recall a long time ago I had put up in the specs (inductance in this case) for various grid chokes--- accdg to the
core materials employed in each one--- but rather than list a quantity I stated the L "qualitatively"--- using language akin to
"good", "better", "best".  I forget the exact terms I used for the chokes but the point was that even the least amount of L of
any of the grid chokes was more than sufficient from a technical POV for the absolute greatest majority of applications.

In one sense--- depending on the specific application--- what is often required and of the greatest import is having or maintaining
a minimum ( a healthy min--- not a starved min) of L----  once you get to the good quality level of acheiving sufficient L---- the cases
where L is then multiplied manifold by a higher perm material may be of limited technical and practical benefit.

Remember that--- in many cases where your using a choke (or a transformer's pri L) you will have a policeman on the job--- meaning
that there will be a R component in parallel with the L.  What you want is for the R to be dominant. This is acheived by making the L
very, very large compared to the R.  

From this one technical POV even our M3\M4 10K:10K input trans has more than sufficient L to do it's job well. The rest of the core
materials are mostly "icing on the cake" from this consideration.

Now--- onto the subjective--- the issues of "how do they sound"---- this is the "art" part of amp building.  Each will sound (in an
appropriately high resolution circuit) different--- but ranking these different "qualities" of sound is subjective to some greater or lesser
extent.  We have customers and OEM's who vary from preference for say all M4, to others who like pinstriped cores, to yet others who
prefer an all nickel core.  Unfortunately--- very few listeners will ever have the chance to "audition" in the same circuit multiples of the
same trans with differing core materials to form their own opinions.

And remember that "core materials"--- as sexy as they are--- are only a part of the overall "goodness" in a transformer---
other materials (insulating mats, magnet wire used, lead wires, terminal arrangements, etc) also have an influence on "goodness"---
as doesn't build quaity (how well is the coil wound, how well is the core stacked, etc) of the trans.  It's a trinity of virtues
that must be present--- virtues in all aspects of the design, materials, and workmanship.

And when you throw in the "buzz word" mentality--- that core mats are "sexy"--- you can get steered in the wrong direction by
just focusing on one of the three virtues that makes for a "good" transformer.  An all nickel core in other words is NOT a guarantee
of a good quality transformer overall.  I'd rather a great Peerless output trans design built on pure M6 core than an inferior design built on
an all nickel core.

I'd love to see some comments on different core mats--- especially from users who have listened to the same trans otherwise.


MSL







    

« Last Edit: May 20, 2012, 11:18:41 AM by mqracing »



Offline Jim R.

  • Hero Member
  • *****
    • Posts: 2194
  • Blind Bottlehead
Reply #7 on: May 28, 2012, 07:39:24 AM
Hi Aaron,

So you got it finished -- excellent.  Want to bring it over for a test listen? :-)

Just curious if you've used other passives in the past and how this compares.

I used to be a die-hard passive guy but after many years decided I liked the more typically robust and weighty presentation of an active pre -- which is where I am now, though still keeping an open mind.

-- Jim

Jim Rebman -- recovering audiophile

Equitech balanced power; uRendu, USB processor -> Musette DAC -> 5670 tube buffer -> Finale Audio F138 FFX -> Cain and Cain Abbys near-field).

s.e.x. 2.1 under construction.  Want list: Stereomour II

All ICs homemade (speaker and power next)


Offline johnsonad

  • Hero Member
  • *****
    • Posts: 1670
Reply #8 on: May 28, 2012, 09:36:21 AM
Hi Jim,

I have had only one other passive, a home brewed Pass DIY B1 buffer.  It had dual mono volume control with PEC pots and I quickly got rid of it.  My system has so much gain that the volume control needs to be percise and well balanced.  I hate fussing with two pots...

I'm not sure if I'm a passive guy yet, but this has opening my eyes.  My active pre is off for servicing so it will stick around for a while.  If I could put my finger on any difference it would be that the active pre is more level all the way around.  Good recordings sound good but bad recordings don't sound as bad as they do with the passive if that makes sense.  It somehow levels the field some and this may be its saving grace in the long run. 

Once the active pre returns I may be willing to send it your way for a demo :)  I wouldn't mind hearing your DAC HD at some point too.

Regards,

Aaron

Aaron Johnson


Offline pboser

  • Full Member
  • ***
    • Posts: 119
Reply #9 on: June 03, 2012, 04:44:36 AM
I don't know if I have dumb questions, or just ignorant ones!
I'm thinking of making a "passive preamp" like the Ingot with a pair of Western Electric 111C repeat coil transformers I have.  They are 600:600 ohms, and I simply don't know what the effect of using them in the hifi area would be.  (I guess I need to do some reading about impedence - any suggestions?).
Some Japanese audiophiles tout these transformers as helping CD sound, and it seems they just wire them in and use them as 1:1 transformers between the CD player and the preamp.
Assuming I'm using standard sources (CD player, Squeezebox, tube or SS phono preamp), how can I best implement this?  What impedence volume control should I use?  Do I need a buffer, and if so, would it be before or after the "preamp"?
I know I'm showing my ignorance here, and sometimes that's annoying to list members, but if you could indulge me to point me in the right direction to learn, I promise I'll try to learn something!
Regards from a beautiful day in NJ,
Pete

Peter Boser


Offline Paul Joppa

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 5844
Reply #10 on: June 03, 2012, 06:36:43 AM
600 ohms is a low impedance for consumer audio; most devices are designed to drive 10K or greater. This is especially true of tube electronics - for example, Seduction prefers 50K or greater, though it can drive as little as 15K with acceptable results.

Some - not all - solid state electronics can drive 600 ohms, even if they are not designed with that in mind. In that case, the transformer would provide some ultrasonic filtering, augmenting the "anti-aliasing filter" function of the DAC in your CD, Squeezebox, etc.

Paul Joppa


Offline pboser

  • Full Member
  • ***
    • Posts: 119
Reply #11 on: June 03, 2012, 08:54:58 AM
Thanks for the reply, Paul.
I get what you've said here, but I was wondering?- is there any impact on the impedance of the primary that comes from plugging the secondary into a higher impedance?  (is this reflected impedance?)
Or is it still essentially a very low impedance load?
Now it's getting cloudy (in the sky and in my brain),
Pete

Peter Boser


Offline mqracing

  • Moderator
  • Jr. Member
  • *****
    • Posts: 26
Reply #12 on: June 03, 2012, 10:02:14 AM

....is there any impact on the impedance of the primary that comes from plugging the secondary into a higher impedance?  (is this reflected impedance?)
Or is it still essentially a very low impedance load?

The reflected impedance on a primary is the product of the load impedance across the secondary multiplied
by the turns ratio squared.

The important point to remember is that the "effective primary impedance" is the reflected impedance in parallel with
the inductive reactance of the primary.  If your inductive reactance (at suitably low frequencies) is not sufficiently large (i.e., much greater in magnitude than the reflected impedance) then the "effective impedance" will actually be less than the reflected impedance.

So you cannot "make" say a 600 ohm primary "look" like a 10 K impedance in a typical 1:1 turns ratio trans by loading the secondary with say a 10K ohm resistor.  Yes--- the reflected impedance back to the primary may be 10K but it will be reduced by the inductive reactance of the primary--- which in the greatest likelihood will not be nearly sufficient to support the 10K reflected impedance.  

MSL







  

« Last Edit: June 03, 2012, 10:05:34 AM by mqracing »



Offline pboser

  • Full Member
  • ***
    • Posts: 119
Reply #13 on: June 04, 2012, 05:42:43 AM
Thank you for the thoughtful and informative reply, Mike. 

This is beyond my knowledge, so I'm reading and re-reading, but what you hit is exactly what I don't know enough about to figure out the problem.  You even anticipated my thoughts about loading the secondary!
So my plan would only work if I was in the 600 ohm range for inputs and outputs, because the 600:600 transformer won't have sufficient incuctive reactance, but a 10K:10K transformer won't have this problem - got it!

Mike, while I've got you, is the advantage of the Ingot over just a pot the stuff Paul mentioned?  "...the transformer would provide some ultrasonic filtering, augmenting the "anti-aliasing filter" function of the DAC in your CD, Squeezebox, etc."  Are there other advantages?
(Not trying to pry into your intellectual property!  Anyway, as you can see, I wouldn't know what to do with it!)

Thanks again, from a rainy Jersey shore,
Pete

Peter Boser


Offline johnsonad

  • Hero Member
  • *****
    • Posts: 1670
Reply #14 on: June 04, 2012, 11:26:45 AM
Today is the first day I've tried the Ingot with the Eros.  They are an impedance miss match at 4k into 10k but you wouldn't know it by the sound.  The way I understand it there should be some bass roll off but if there is, it's not noticable to comment on.  This my one piece of gear that I miss the gain of an active stage as there is an increase in tube rush.  Maybe this will get me off my duffer and put an active stage between the Otari and Eros to boost the signal a little. 

I never thought I would be a passive guy but I haven't found a reason to remove it yet :)

Aaron Johnson