X Preamp

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
diyAudio Retiree
Joined 2002
Oh what the hell ..........

Actually I do have an X preamp in the design phase. Not shown is the Fet follower to buffer the circuit from the volume control. The balanced part need to see equal impedance. i have a nifty servo that uses only 3 transistors and acts on the common mode offset and ignores the audio signal this should operate with lower distortion and 1 less cap than a servo for each balanced output as well as a greatly simplified circuit. This is a work in progress but someone could probably build one from this design with AC coupled outputs. The bais points and feedback are based on a pretty extesive study of Borbely's Jfet and Pass's mosfet preamp designs with several nested feedback loops. The Borpass preamp doesn't sound like a very enticing name and the Nerno preamp is absolutey out of the question!

There is about 10 hours of Spice analysis to get this far. The circuit is counter intuitive concerning output impedance. The preamp has to drive two (or one for nonbalanced amps and that was the problem) single ended loads with respect to ground at the amp input [ INSERT STEVE EDDY TRANSFORMER SPEECH HERE ] and the differential and common mode output impedances are different which was a real dilemma in the early design phase. I have never been forced to fight my gut level instincts conflicting with the Spice results like this one! Made me nuts for two days trying the resolve the difference.

It also lead to a very interesting dialog with Mr. Pass on an op amp circuit that does not work and was posted on the forum. That hasn't stopped the op amp vendors from publishing it in some data sheets and app notes. Seems the original circuit required some coupling caps the op amp guys left out. I think even Mr. Pass was scratching his head on that one.
 

Attachments

  • x15pre.jpg
    x15pre.jpg
    28.4 KB · Views: 2,761
Hey Fred,

Good one,

I had a look at the Borbely pages last night after readng your post. I also down loaded Mr Pass's article on diy opamps, its excellent and I really like the references to CCS which I will study carefully over the weekend. Borbely also has an updated CCS using just two fets and one resister!!

I will have a closer look at your design when I clear the bench of the X Aleph and my new re built BOZOZ.

Ian
 
diyAudio Editor
Joined 2001
Paid Member
Hey Fred, That's awfully nice of you to share. Interesting (tempting?)

"There is about 10 hours of Spice analysis to get this far."
You have to admit that shows developing stuff isn't a matter of just plugging into a modeler ;)

Ian : Is that mudcake stuff a strine(sp?) expression or are you referring to the washing? I think they use an ultrasonic transducer hooked to the Aleph to do the wash up there.

I did once hint to Cyclotron Guy that maybe I could drop by to take the tour of the Pass labs Research and Production world headquarters. He responded that the complete tour would take about 30 seconds, but didn't say they would refuse to let me enter ;) I think they suspected that I was hoping for a mini Aleph on exiting, the way Wonder Bread gave the second grade class mini loaves of bread after the bakery tour. I'm afraid you are right Fred- I seem to want everything handed to me :devilr:
 
By the way this is a bit off topic but its still progess, I tried to X the BOSOZ last night.

Gee what a change. Has anyone else done the conversion recently?

At the moment I'm evaluating dual CCS (Aleph style) in normal BOSOZ mode..


The short answer is if it ain't broken don't fix it.

Ian:)
 
Na Na,

I put it in the X Aleph Thread like a good little Boy.......well worth a try...Muahhahahahahh.

As mentioned earlier too early to say about X BOSOZ ..but .My neighbours can hear the the dymamics I'm sure.

What happens when you (X) x (X) = Rub the Magic......L.........Big Balllllaaaboooong............., I mean Boommmmm....... Like Bruce Wllis said I know Boom.

Time for Peanut Butter & Chocolate...yummy.

Ian:lickface:
 
Ian:

How did the Aleph CCS's work on the BOSOZ? This is one of my favorite combinations.

I am not convinced that we know how to draw an X at this point, plus, as the master said; "not all amplifiers should be Xed."


____________________________________________________
The laws of physics always apply.

The real question is;
which ones this time?

Tmac
 
O'kay,

I'll post back here for now sorry to confuse you all.

The results of my Sunday school readings...

It took a while to arrange the CCS and re jig the pre for X BOSOZ.

The setup.....

X BOSOZ exactlyLike Hendrik, 10K on inputs, 39K feedback - in to + out, + in to - out. The - in is taken to earth at the node of 10K and the attentuator.(Thats another story 23 position switch from some special!!)

The CCS is irf 610 on good heatsink (like mini Aleph bloody hot mate) + 220R on gate, 9.1 v zener bias 10K to earth and about 60R on the source to about -58V The CCS is common to both active Fets (610). This gives about 75ma total current.

So what happened..... Okay BIG BulllaBooooooom

Running the Aleph 2..........Woooooooooooooooooow.

I like the results, accuracy, natural and very listenable. thats a hard nut to crack!

XBOSOZ with active ccs......10/10
The spatial delineation is very impressive, there is a nice holograph, and the silence between the notes on solo Piano is outstanding.

With the X BOSOZ resistive ccs
Nice width/depth and drive, but there is some Smoke on the Water, the sound seems to be fighting itself, aiming for purity but not quite. I find it distracting.


The BOSOZ with resistive ccs.
Well we all know that sound, just like your aunts favourite plum pudding....yummy, engaging, warmth and soothing.

The analysis

There would appear to be close correlation between the use of active CCS and X topology , and the the use of std Fet diff pair and resister tail.

The BOSOZ to me and the resistive ccs of 1500R tail has a nice synergy with the end sound, ie they mate well.

So does the X BOSOZ with the active CCS desribed.

But X BOSOZ with the resistive ccs tail is less together in totallity of balance....like smoothy Peanut Butter but taste a bit strange.

The resistive ccs seems to add a nice skin to the sound, the X is cancelling everthing but the skin is still there, so that is confusing the mind.

The active ccs removes the skin and the X cancels everthing and it sounds complete.

With std BOSOZ, the skin is there and the warmth, I think they complement each other.

Hope all that made sense.

Sunday school is over.

Ian
 

Attachments

  • sundayschoolover.jpg
    sundayschoolover.jpg
    53.8 KB · Views: 2,112
Disabled Account
Joined 2002
macka said:

There would appear to be close correlation between the use of active CCS and X topology , and the the use of std Fet diff pair and resister tail.

The BOSOZ to me and the resistive ccs of 1500R tail has a nice synergy with the end sound, ie they mate well.

So does the X BOSOZ with the active CCS desribed.

But X BOSOZ with the resistive ccs tail is less together in totallity of balance....like smoothy Peanut Butter but taste a bit strange.

Great information. Thanks.

Intersted, dazed and confused.
Your opinion is quite different from Henrik's.
You praise X BOSOZ with the active CCS while Henrik does with the resistive CCS.
The difference I see is that yours is X BOSOZ + Non-X Power while Henrik's is X BOSOZ + X Power.

And, confused with the threads on mixed subjects...

JH
 
Yeah,

Well nothing is set in stone with any of this stuff:

Intersted, dazed and confused.
Your opinion is quite different from Henrik's.
You praise X BOSOZ with the active CCS while Henrik does with the resistive CCS.

The difference I see is that yours is X BOSOZ + Non-X Power while Henrik's is X BOSOZ + X Power.

Correct, Henriks set up is different as you say he has X pre + X power.

These are just my observations, when I get the X Power running, hopefully soon, they may change.

But while I hear differences, there must be reasons, I am just trying to apply some logic and rationality to draw a conclusion in my scenario.

But that can be consusing when you are standing up side down (here in the southern hemisphere!!) so forgive me

Ian:bigeyes: :confused: :cubist:
 
Ian:

Good mourning to y ‘al down under.

Thanks for a nice clear report on the BOSOZ preamp. I am looking forward to your comments on the rest of your tests with the Xed Aleph 2.

Also, enjoyed the group picture.

THE THREAD:

I would like to see a Zen Variations article from Nelson on Xing an amplifier. I learned a lot from that series.
 
Hi all

This discussion should have bin under the X-BSOZ or X SOZ thread.
Any way, I encoutered, after may hours of listening sessions, only very subtle differences with and without CCS, certainly no "Smoke on the Water" with resistive currentsources.
But I believe the differences can be related to the set up and specially the parts used.
 
Re: Re: some gratitude please

uli said:


:devilr:
Working unit:

X-Modul

Uli
:nod: :nod: :nod:

I remember your thread, it was remarkable how little attention it got. Personally, I find your circuits very elegant with the only drawback to diffusion is that the great majority of people around here cannot understand them.
However, your point is well taken that fred was scooped a long time ago with this x-preamp. I wouldn't have expected this from someone that can always wip out a pdf document in reference to anything audio.
 
understanding

grataku said:

However, your point is well taken that fred was scooped a long time ago with this x-preamp. I wouldn't have expected this from someone that can always wip out a pdf document in reference to anything audio.


Hi Grataku,
As I am not a native English speaker, I don´t completely
understand your post.
Please could you explain?

Thanks
Uli:nod: :nod: :nod:
 
scooped and scoped

I am happy to finish 3rd (or have I msised another one too?) I did not show the voulme control buffer. The circuit needs to be driven from equal source impedances on each half for the best performance. Even a 10K pot will imbalance this impedance and as function of the volume setting. The 100K input will load the wiper of the pot and change it's log taper. I have heard some express concerns about signak current through the wiper of the pot as well. Forget about a 100K pot without a buffer.

I have a few questions:

Why not use mosfets in place of the BJTs?

Since your topology uses an inverted cascode and one stage of gain, instead of two as the Xamp uses, how did you get enough open loop gain?

If the preamp is being by a source impedance less than the 100 ohm input resistor, isn't the close loop gain 220K divided by 100 or 2200 assuming an open loop gain much greater than thie closed loop gain?

For a 100k load, isn't the open loop gain about 50K divided by 50
or 500 which is less than the closed loop gain?

How much cable capacitance can it drive without stability or bandwidth limtation issues?

What is the output impedance?

How do you deal with DC offset on the output?

I am not trying to nit pick or engage in armchair engineering without looking closley at the circuit. I am very curious about these parameters which took a lot of thought when I was designing mine.

Thanks,

Fred
 
when a problem comes along you must whip it

grataku said:


I remember your thread, it was remarkable how little attention it got. Personally, I find your circuits very elegant with the only drawback to diffusion is that the great majority of people around here cannot understand them.
However, your point is well taken that fred was scooped a long time ago with this x-preamp. I wouldn't have expected this from someone that can always wip out a pdf document in reference to anything audio.

Whipping out out a PDF technical refence is a lot easier the whipping out a design.... I had to think for awhile on this one!
 

Attachments

  • 76whip_it.jpg
    76whip_it.jpg
    6.5 KB · Views: 974
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.