[MrCurwen]

I'm an open loop guy all the way.

There's no way to know at the moment, but my amp probably has LESS of an "SS sound" than The Little Dot, since my amp is open loop. Despite the 20 or so FETs that reside in it.

Sorry off topic. Couldn't resist.

 

[Maxx134]

I'm an open loop guy all the way.

There's no way to know at the moment, but my amp probably has LESS of an "SS sound" than The Little Dot, since my amp is open loop. Despite the 20 or so FETs that reside in it.

Sorry off topic. Couldn't resist.

Is it the one you pictured?
how many tubes and what are they?
The past picture shows a straight line of tubes..
Just curious to know the layout

 

[coinmaster]

With 1µS timestep and improved accuracy math on it takes a couple of minutes to do a 100mS sim of a complex circuit. And my computer is made of potatoes from 2004.

Even my overclocked 4770k takes a while to even begin the simulation.
I think LTSpice hates all the integrator/comparator feedback loops I'm using. Feedback loops seem to be the enemy of load times.
 

Spice has inbuilt THD measument, usable by one line;

.four 1000 9 V(n002)

Okay I loaded that up and this is the MK6 stock, sorry about the scale, not much I can do about it. 

The THD is 0.136160% stock
 
This is the MK6 with impedance mod

THD is 0.062541% modded
Which is weird because it looks like it actually has more distortion.
 
Still trying to get my design to load

 

[MrCurwen]

Is it the one you pictured?
how many tubes and what are they?
The past picture shows a straight line of tubes..
Just curious to know the layout

Since you asked, I'll answer despite off topic!

The pictured one was my flagship amp. Forced balanced from input to output.

The two rightmost tubes are input stages; 6SN7 russian eq.

The four left side tubes are 4P1L, each grid driven by a source follower buffer.

The power buffer on top is 4 source followers with CCS loads driving parafeed OT's. Each follower has about 15mA bias current.

It has 4 PSU:s, each regulated with a simple 2 FET regulator;

1) +290V for the input stage and the grid buffers
2) +356V for the output tubes
3) +305V for the power buffer
4) -14V for the input stage tails.

The signal parts have 28 FETs, the regs have 8. Also 4 low rdson FETs regulate each of the 4P1L filament supplies. The amp has a pitch black background despite having DH tubes (not a given by any means).


My current (not in any casing, just breadboard) workhorse amp is similar topology. 12AX7 input stage, grid buffers feeding 6E5P output tubes. Single B+ of 310V with a two FET reg, no output buffer.

This one has some problems, some background noise. The super high transconductanse tubes such as 6S45P and 6E5P can be a pain to deal with. They do provide excellent dynamics though, even without a FET power output buffer. Not as good, but pretty great.

Anyway it has good sound quality, so I'm too lazy to work on it. I'll build another amp for myself later, identical to the flagship amp I made earlier. I already have the chassis worked and painted and also most of the woodwork done.

I've made a number of variations on the theme with different tubes. Next I'll make a fully forced balanced 6SL7 -> grid buffer -> 2C22 -> power buffer, for one person I know.

2C22 is in my experience the most linear and sturdy 6SN7 variant there is. It should be, since it was made to be just that.

The amp I'm thinking about sending out to the gentleman in Denmark and possibly further on from there would be fully forced balanced 6SL7 -> grid buffer -> 6SN7 -> very small power buffer, kind of a stripped version of the previous model. I must make it yet smaller or it's not going to be possible...

 

[Redge78]

For the record, I have (slightly) changed the tittle of this thread, and made an addition in the first post.
 
For the last 20 pages or so, the discussions have severely drifted from their original purpose.
Personally, I find that interesting, even if I would have preferred to have some practical suggestions to work on.
But for a newcomer it may be only useless technical garbage.
 
And our initial aim was to help the newcomers, remember.

 

[coinmaster]

My initial aim was just to make an awesome amplifier 

 
We've hit a wall on simple mods anyway so this is the next step to more mods.
Plus this is valuable information.
 

even if I would have preferred to have some practical suggestions to work on.

CCS/gyrator/grid drive/power supply/LED/battery mods are all practical.
It depends on how far you wana go. Like I said I'm ordering PCBs for them for myself anyway so you can just take the file and order your own. Not difficult to screw them onto an extension box with standoffs and the implementation to the board is simple.

 

[Maxx134]

We've hit a wall on simple mods anyway so this is the next step to more mods.

Not me,
I have to push out:
1- bias adjustment refinement for MK8 powertubes
2- impedance switch mod
3- driver tube change & refinement
4-DE-coupling caps refinement
5- parasitic red caps upgrade
6- power tube rolling with corresponding bias adjustment.

Future mods:
7- Extend bottom chasis
8- PSU upgrades

 

[coinmaster]

Future mods:
7- Extend bottom chasis

That should be your top priority lol.

 

[Redge78]

  We've hit a wall on simple mods anyway so this is the next step to more mods.
 

 
You may have but I don't. I have enough new mods to keep me busy for the rest of the year, not speaking of the mod that I have already implemented and are in test right now.
 

  Plus this is valuable information.

 
I don't deny that, even if the term 'information" may be a bit strong.
There is a large community of (european ?) DIYers that would be less than enthousiast on those CCS and FET schematics.
So until  I have actually listened to those SS things, I will remain cautious. 
 
Anyway, my concern is more about "the new guy" whose main hobby is to listen to music and not to understand the electronic finesses. Let say he has 25 hours (?) to read/understand/order/solder/test our mods and he will go back to listening ... not sure those last pages of thread will be of a great help.
 
 
 

 
CCS/gyrator/grid drive/power supply/LED/battery mods are all practical.
It depends on how far you wana go. Like I said I'm ordering PCBs for them for myself anyway so you can just take the file and order your own. Not difficult to screw them onto an extension box with standoffs and the implementation to the board is simple.

 
Don't get me wrong, all that may be very good.
But until you can give us an actual "MK6" listening feedback (corroborated by someone else would be a must), with a list of components, some nice pictures and an "How to" ... this is not "practical" in my book.
 

 
It depends on how far you wana go. Like I said I'm ordering PCBs for them for myself anyway so you can just take the file and order your own. Not difficult to screw them onto an extension box with standoffs and the implementation to the board is simple.
 

 
You're right, my ambition never was to rip the amp off and to modify 90% of it. 
I want to keep a sensible "time spent" to "SQ increase" ratio.
Even if the thread has impacted badly this ratio ... it also brought us some new joy.

 

[Redge78]

  That should be your top priority lol.

 
Tell him that he would have more room to new decorating ideas !
 

 

[coinmaster]

There is a large community of (european ?) DIYers that would be less than enthousiast on those CCS and FET schematics.

I've done a lot of reading over the last 6 months, I've heard both sides of the story and the "tube purist" ideology has been debunked a thousand times. That's putting aside the science, data, experiences, and overall design logic behind it. Even if all SS devices were evil they aren't always in the "signal path" like the gyrator for example, depending on how picky you are going to be on your idea of "signal path".
 
I'll be the first to say that all SS amps I've tried are garbage, even the big heavy expensive ones but you might as well ditch the opamp in the MK6 since it's SS right?
Way too many claims from people that laugh at audiophoolery and design tube amps for decades that SS can sound good with proper implementation for me to dismiss it, coupled with the design knowledge I've gained and data I've seen I doubt it's untrue. In the same way I doubted high end caps can't make a difference even though so many people claimed that was BS, and here I am after blowing 800 bucks on duelund caps and heard an amazing difference.
 
I think ignorance and stubbornness is the enemy of innovation and progression.
 

 

Anyway, my concern is more about "the new guy" whose main hobby is to listen to music and not to understand the electronic finesses. Let say he has 25 hours (?) to read/understand/order/solder/test our mods and he will go back to listening ... not sure those last pages of thread will be of a great help..
 

No they won't, that's why I've been saying for months now you should put a disclaimer saying the front page is where the tried and tested mods and instructions are at, that way we can discuss and progress in the thread without resorting to inefficient PM secrecy that only a select few can get in on and learn/contribute. Much better this way for a lot of reasons.
 
Although I do agree it was good that most of our earlier conversations were kept in pms. At least now we can discuss things intelligently knowing what the design is and not having me be a clueless newby to electronics pestering you for information 

 

Don't get me wrong, all that may be very good.
But until you can give us an actual "MK6" listening feedback (corroborated by someone else would be a must), with a list of components, some nice pictures and an "How to" ... this is not "practical" in my book.

I'm workin on it. If anyone wants the PCB design I can finish up the last details and upload the file now, there's no reason it should not work the design is functional and the PCB layout has minimal inductance but other then that it will probably be july before I do my own tests at this rate.
 

I want to keep a sensible "time spent" to "SQ increase" ratio.

Indeed, but from a technical point of view, the mods are such an upgrade that it is likely they will be worth it. At the very least you can rest knowing you have the best 

 That's kind of my goal.

Tell him that he would have more room to new decorating ideas !
 

 

Indeed he would. I can only imagine..

 

[Maxx134]

 
That should be your top priority lol.


...Indeed he would. I can only imagine...

Tell him that he would have more room to new decorating ideas !

Haha!

I already had body / chassis idea ready for next amp,
Which will be never made at this rate haha .

but you're right looks like it's time to extend my bottom especially since I can use that space for switch mods haha

** edit:
at this rate looks like my Amp may not be ready for the upcoming New York headfi meet,
but at I least I will bring my newly-modded headphones..
Hd800
HE1K
PM2
All modded.
Finally I went a bit off-topic

 

[coinmaster]

Finally I went a bit off-topic

How dare you
 

at this rate looks like my Amp may not be ready for the upcoming New York headfi meet

Oh right...forgot about that. I don't think I will be going, the last thing I need is to relapse on good sound while my amp is still in progress.

 

[Maxx134]

http://www.head-fi.org/t/788652/official-new-york-spring-meet-saturday-june-11-2016/0_20

 

[SonicTrance]

I've been reading recent posts regarding biasing the driver stage but did not find any useful info.
 
What I'd like to try is to bias the ECC32 more optimally in the MK6 by altering the Ra and Rk.
 

According to the above chart, a 47K Ra and 1K2 Rk would give me 1.9mA. That'd be a significant decrease in Ra though compared to stock 220K.
 
Thoughts? @MrCurwen