Ultra-high performance, yet rather simple - hybrid and more!

[vzaichenko]

pls.I do not have diode 1n5711 and I may replace it with another diode other what kind sir all

You can use 1n4148 here. These diodes are utilized only during heavy clipping, preventing from deep saturation.

 

[vzaichenko]

Thank you Evan!

On Thursday last week I have visited Evan (member evanc) at the coast of New Jersey and saw and heard his build of TubSuMo (mids and highs) in combination with SlewMasters (bass section). Cool hand-made wooden compartments, excellent speakers with REAL woofers, the room is just the right size. Everything fits The whole thing sounds delicious - extremely natural and detailed, with sound sources all over the room.
Audio paradise I wish every system would sound like this one

Congratulations Evan and thank you for the cold Hoegaarden, all your hospitality and a quick excursion to the beach

Cheers,
Valery

 

[still4given]

Wow, good job Evan!

 

[evanc]

Valery...You are too kind. The pleasure was mine. It is a treat for me to meet one of the very talented and kind people that make up this forum. Any of my successes here are due to the help of folks like you.

I liked hearing "when you are with Russians you never drink alone." People after my heart.
Cheers, Evan

 

[ostripper]

On Thursday last week I have visited Evan (member evanc) at the coast of New Jersey and saw and heard his build of TubSuMo (mids and highs) in combination with SlewMasters (bass section). Cool hand-made wooden compartments, excellent speakers with REAL woofers, the room is just the right size. Everything fits The whole thing sounds delicious - extremely natural and detailed, with sound sources all over the room.
Audio paradise I wish every system would sound like this one

Congratulations Evan and thank you for the cold Hoegaarden, all your hospitality and a quick excursion to the beach

Cheers,
Valery

You met Terry and Evan ... passed right over me ??? (they are on both sides
of me).... Oh well - not close enough to a coast.
PS - I would of got you high on moonshine (and other things) .
Evan - those are masterpieces. I might of chose wood over chinese aluminum !!
OS

 

[vzaichenko]

You met Terry and Evan ... passed right over me ??? (they are on both sides
of me).... Oh well - not close enough to a coast.
PS - I would of got you high on moonshine (and other things) .
Evan - those are masterpieces. I might of chose wood over chinese aluminum !!
OS

Man, next time I go to the US - I will plan for a longer stay to visit you in Smoky Mountains and the other friend in Austin, Texas (this time she came to NY for two days just to see me ) It will be really cool!

 

[still4given]

You met Terry and Evan ... passed right over me ??? (they are on both sides
of me).... Oh well - not close enough to a coast.
PS - I would of got you high on moonshine (and other things) .
Evan - those are masterpieces. I might of chose wood over chinese aluminum !!
OS

Your new avatar, Pink Floyd?

 

[evanc]

OS..Masterpieces is a bit of a stretch....more like plywood boxes. The real art is in your guys excellent designs. Me thinks you and I may have a few things to share if our paths ever cross.
Evan

 

[ostripper]

OS..Masterpieces is a bit of a stretch....more like plywood boxes. The real art is in your guys excellent designs. Me thinks you and I may have a few things to share if our paths ever cross.
Evan

What I meant was there are so few craftsmen left. Everything is cheap and
mass produced.
I built hundreds of ported isobaric sub boxes decades ago , the learning
curve on that is one step beyond just simulation and layout of an
amp circuit.

With the amps , me and Val get it right pretty much all the time. Making
speakers and cabinetry needs extended skills. Some think just a pure
technical creation requires a superior skillset . A physical creation requires
different but equally sophisticated "gifts".

PS- It was MUCH harder for me to build an amp than to design one !
I was literally "pulling hair" .

OS

 

[rhythmsandy]

Here are the spectrums together with measurements in the following order:

1) THD 1KHz 4.2V RMS - NO OPS
2) THD 1KHz 4.2V RMS - with OPS

3) THD 10KHz 4.2V RMS - NO OPS
4) THD 10KHz 4.2V RMS - with OPS

5) THD 20KHz 4.2V RMS - NO OPS
6) THD 20KHz 4.2V RMS - with OPS

7) IMD 14KHz + 15 KHz 4V RMS - with OPS

Listening - delicious, high resolution, dynamic, natural sound. Excellent elastic low-distortion bass. In fact, very close to the other OPS with 3-5 pairs of IRFPs and EF drivers, based on SlewMonster - well, pretty much ensured by IPS.

Waiting for ksc1845/ksa992 transistors for testing the new IPS...
The new one has got extra 6db of loop gain, plus at least twice as much of VAS current. Will be interesting

Cheers,
Valery

Is this review from you from Jfet input or Valve input stage?

 

[vzaichenko]

Is this review from you from Jfet input or Valve input stage?

Those are measured with the Valve at the input.

 

[rhythmsandy]

Those are measured with the Valve at the input.

Have you tried using Jfets?
In the other thread Ostripper mentioned clearly that this kind of VAS stage has stability issues and I have read your complete thread but why is it oscillating with some boards and some are working fine is there a small parasitic element in connecting wire is disturbing the stability?

waiting for your Jfet input since the board is same..

Can I use 6SN7 at input stage what values to be changed?

 

[wahab]

Nice design, based on the Sansui diamant topology apparently..

On the Bipsumo, and for whatever bipolar IPS, R6 is in the wrong side of R5, it should be connected directly to the transistor base to get mimimal DC offset, this has been perhaps already been pointed in the thread.

 

[vzaichenko]

Adapted to +/-50V rails and measured at 50W (20V RMS @ 8 ohm)

Square waves are measured at 40V p-p amplitude.

 

[kacernator]

OK, some measurements on IPS alone (~20Kohm resistive load).
I deliberately ran the square waves at high swing - that's where some artifacts normally appear.

Good points - only 3.6 degrees phase shift @ 20KHz, nice clipping, very good square wave response, THD / IMD - no comments.
Pretty fast VFA. Well, OPS will slow it down - we'll see, to what extent...

Checked with a pair of blue LEDs in VAS offset circuit as well as with the double-red pairs. No difference in noise floor. Just not at all. Nothing. Left double-reds for now, but either can be used.

So far so good... Next test - with OPS attached.

Cheers,
Valery

How does it compare to JFETs input? Is tube really that good compared to JFETs?

Thanks.

 

[vzaichenko]

How does it compare to JFETs input? Is tube really that good compared to JFETs?

Thanks.

I have actually never tested the jFET option in the real prototype.

However, the tube one is very good - see the performance with non-switching OPS in the other thread:
http://www.diyaudio.com/forums/soli...e-old-ideas-1970s-ips-ops-35.html#post4554858

There are also a few amendments described in the following post in that thread.

 

[still4given]

Hi Valery,

I'm at the point where I want to finish off some amps with cases. This is one of the ones I want to do. Are there some changes that I should incorporate in this final build? I plan to do the Vertical with the Non-switching OPS so if you can would you please post a final as-built here for the Tubsumo IPS and OPS?

Thanks, Terry

 

[vzaichenko]

Hi Terry,

Are you reading my thoughts?
Your post came in right at the time, when I was re-measuring TubSuMo with original LatFET OPS after some final amendments (to the OPS, but anyway)

Original combination is still excellent - so I publish the whole thing here for those who may be interested.

However, TubSuMo IPS + NS-OPS is really the top performer - great candidate for the case build. In this case, just see the IPS part.

Key changes, comparing to the initial version (the one you've got, I believe):
R6, R11 = 10K
R27, R28, R30, R31 = 100R
R29 = 150R
D14, D15 - only one Red LED on each side (instead of two)
D8, D9 - 63V Zeners are removed

Still no need to cut traces, just solder the jumpers over two removed Red LEDs.

I run the whole thing with +/-52V rails now. More than enough power (up to 120W @ 8 ohm) at lower temperatures. Q10, Q13 have smaller heatsinks now - barely warm.

This one came to the state of a very mature design now - in terms of quality, stability and repeat-ability

Despite its rather high-impedance feedback network, this is still a CFA-fast VFA

Spectrums presented are taken at 50W power @ 8 ohm load.
Wquare waves - around 20V p-p @ 8 ohm load.

Cheers,
Valery

 

[vzaichenko]

TubSuMo with LatFETs

Here are the spectrums for the same configuration, but with 10W output @ 8ohm load.

 

[jwilhelm]

Are the original values still good for 70V operation with the NS-OPS?