DIYHiFi.org

For the sake of audio
It is currently Wed Aug 23, 2017 9:17 pm

All times are UTC




Post new topic Reply to topic  [ 130 posts ]  Go to page Previous  1 ... 5, 6, 7, 8, 9  Next
Author Message
PostPosted: Tue Jun 12, 2012 1:04 pm 
Offline
Cow
User avatar

Joined: Tue Dec 14, 2004 9:32 pm
Posts: 446
Location: Swiss Mountains
CG wrote:
2SK214??[...]


Typo i think.
Schematic says 2SK246, which is a JFET and would be in line with the symbol used.


Top
 Profile  
Reply with quote  
PostPosted: Tue Jun 12, 2012 4:59 pm 
Offline
Muriel
User avatar

Joined: Thu Jul 26, 2007 5:45 pm
Posts: 1111
Hi,

zinsula wrote:
CG wrote:
2SK214??[...]
Typo i think.
Schematic says 2SK246, which is a JFET and would be in line with the symbol used.


An earlier version showed 2SK214/2SJ77 as driver for the Bipolars. These have these levels of capacitance (2Sk214 = 90pF Ciss).

For the 2SK246 in the Compound pair it goes like this.

Crss = 2.5pF
Ciss = 9.0pF

The 2SK246 has a gain of 2, so it's reverse transfer capacitance (Crss) is doubled to 5pF. The input capacitance (Ciss) is bootstrapped a lot, so it basically disappears.

As for operating at zero tempco, in a compound pair the bias/current is dominated by the first device.

So if this device has a zero Tempco operating point (Fet's and J-Fets have a variable tempco with current, BJT's have a constant tempco) any temperature related changes are minimised (in theory cancelled). As we also reduce the absolute current and thus dissipation to 20% of just using a single device any changes due to changes in dissipation with signal.

If the input devices parameters change with temperature (which they do for BJT's) we have an error that remains outside the global feedback loop and cannot be improved using global feedback. This can be improved by heavy degeneration, compound pairs, cascoding and operating the device that dominates at zero tempco. You may observe my proposed design uses all these measures in quite some measure...

Ciao T

_________________
"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." Richard Feynman


Top
 Profile  
Reply with quote  
PostPosted: Tue Jun 12, 2012 10:58 pm 
Offline
Hen

Joined: Mon Feb 27, 2006 4:21 pm
Posts: 9
Location: Fort Collins, CO
If the 2sk214 has a gain of 2, shouldn't cgd be tripled?

Too bad the input impedance of a bjt driver stage is in the single to double kohm range. Otherwise, you could do a npn-pnp compound pair to substitute for a hard-to-get and expensive 2sk214/2sj77 and still do your selective local feedback thingy.....


Top
 Profile  
Reply with quote  
PostPosted: Wed Jun 13, 2012 3:14 am 
Offline
Muriel
User avatar

Joined: Thu Jul 26, 2007 5:45 pm
Posts: 1111
Hi,

rlim701 wrote:
If the 2sk214 has a gain of 2, shouldn't cgd be tripled?


The 2SK214 in the Driver stage has a gain of slightly below 1.

The gain of around 2 happens in the input stage where there is a 2SK246 & 2SA970 compound pair.

rlim701 wrote:
Too bad the input impedance of a bjt driver stage is in the single to double kohm range. Otherwise, you could do a npn-pnp compound pair to substitute for a hard-to-get and expensive 2sk214/2sj77 and still do your selective local feedback thingy.....


The latest version changed to 2SK1056/2SJ162 (or Exicon etc.) Mosfets as drivers.

I would recommend only some form of lateral mosfets as drivers.

They have several distinct advantages over other options. Their capacitance varies very little with signal (unlike vertical Fets) and is quite low, plus their input impedance is almost purely dominated by the capacitance. At the same time they are very fast, so they do not add appreciable phase-shift of their own to the output stage, making the circuit by far more stable, especially if compared to a EF3 circuit.

Their downside is a poor linearity with speaker loads and relatively high output impedance, which are very much mitigated against in use as drivers (as compared to use as output devices).

Ciao T

_________________
"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." Richard Feynman


Top
 Profile  
Reply with quote  
PostPosted: Sun Jul 08, 2012 9:08 pm 
Offline
Sheep

Joined: Mon Jun 02, 2008 8:17 pm
Posts: 11
:grin: my second Goldmund Mimesis 9.2 clone amplifier it's ready and play music wery well , I do not need to much theoretical stuff ......


You do not have the required permissions to view the files attached to this post.


Top
 Profile  
Reply with quote  
PostPosted: Fri Jul 13, 2012 11:16 pm 
Offline
Hen

Joined: Mon Feb 27, 2006 4:21 pm
Posts: 9
Location: Fort Collins, CO
Hi T,
How did you analyze the shunt regulator? Did you attach a load R where R is the regulated voltage divided by whatever current your load needs?
Thanks.


Top
 Profile  
Reply with quote  
PostPosted: Sat Jul 14, 2012 6:21 am 
Offline
Muriel
User avatar

Joined: Thu Jul 26, 2007 5:45 pm
Posts: 1111
Hi,

rlim701 wrote:
How did you analyze the shunt regulator? Did you attach a load R where R is the regulated voltage divided by whatever current your load needs?


I used the nominal load as current sink (because the AC impedance of the circuit it will feed will be quite high and used the various build in instruments in Tina Ti which includes a resistance meter that also works AC.

Independently I analysed the Circuit using Pen & Paper and the transistor etc. Datasheets, which is how I got the ten fold excessive optimism of the simulator re. output impedance, compared to my calculations

Ciao T

_________________
"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." Richard Feynman


Top
 Profile  
Reply with quote  
PostPosted: Sun Jun 18, 2017 3:10 am 
Offline
Goat

Joined: Mon Nov 28, 2005 1:10 pm
Posts: 514
Kuei Yang Wang wrote:
The front-end regulation uses a pair of 317/337 as CCS (and with a circuit that does not have the circuit flaw that caused the 337 CCS in Walt Jung's test to have very poor HF behaviour)
I take it that the cap from output pin to gnd will improve the 317 ccs as well? I have not measured it but I have an excellent sonic result with this simple 317 ccs (with Thorsten´s suggested cap). It is used before a shunt regulator powering a balanced common gate I/V stage, single supply. Thanks Thorsten for the ccs tidbit.
Attachment:
lm317 ccs.png


You do not have the required permissions to view the files attached to this post.


Top
 Profile  
Reply with quote  
PostPosted: Sun Jun 18, 2017 12:18 pm 
Offline
Hen

Joined: Thu Dec 30, 2004 3:50 pm
Posts: 9
What cap value?


Top
 Profile  
Reply with quote  
PostPosted: Sun Jun 18, 2017 8:08 pm 
Offline
Goat

Joined: Mon Nov 28, 2005 1:10 pm
Posts: 514
As I recall the value was 10uF 100V on Thorsten´s schematic. I used 100uF 50V with a 3.5mm lead spacing as this is important for the HF impedance of the cap.


Top
 Profile  
Reply with quote  
PostPosted: Mon Jun 19, 2017 4:54 pm 
Offline
Hen

Joined: Thu Dec 30, 2004 3:50 pm
Posts: 9
alexandre wrote:
As I recall the value was 10uF 100V on Thorsten´s schematic. I used 100uF 50V with a 3.5mm lead spacing as this is important for the HF impedance of the cap.


Which schematic?


Top
 Profile  
Reply with quote  
PostPosted: Tue Jul 11, 2017 1:20 am 
Offline
Goat

Joined: Mon Nov 28, 2005 1:10 pm
Posts: 514
I can now hear big differences in sound with different regulators and bypasses in my open-loop I/V converter. This point of the audio chain presents an opportunity to tune the sound to my liking and the changes are even more evident than tweaks in the power amp (which is high feedback).
Changing one bypass cap can make or break the sound. This may be obvious to most here but I wasn´t fully convinced of it until now. I had two small Lelon electrolytics where the rubber was bulging. The sound was fine but the rubber was stuffed because of the anti-corrosive spray that I recently used. I changed them to equivalent Epcos units and I wanted it to be an improvement but it just wasn´t. I have "measured" this in medium term listening satisfaction rather than first impression. It went on from there and I have tried a few different regs and bypasses. TL431 from TI is clearly better than onsemi (both at unity gain for AC). Again I wanted the onsemi part to be better (someone mentioned lower noise over at DIYA). Now I am listening to this discrete shunt built from whatever parts I could find in my drawer. Is motorola MJE340 any good?
Posted here because it is based on Thorsten´s shunt reg from this thread. V1 is BC548 connected as zener - emitter is positive, base is negative. I1 is the CCS from this thread and I2 is the load, the open loop I/V (single supply).
Attachment:
shunt sch.png


You do not have the required permissions to view the files attached to this post.


Top
 Profile  
Reply with quote  
PostPosted: Tue Jul 11, 2017 1:25 am 
Offline
Goat

Joined: Mon Nov 28, 2005 1:10 pm
Posts: 514
Attachment:
shunt ac.png


You do not have the required permissions to view the files attached to this post.


Top
 Profile  
Reply with quote  
PostPosted: Wed Jul 12, 2017 5:04 pm 
Offline
God

Joined: Thu Dec 30, 2004 10:50 pm
Posts: 989
Location: Minnesota
You have entered the point of no return zone.

keep at it.


Top
 Profile  
Reply with quote  
PostPosted: Wed Jul 12, 2017 7:10 pm 
Offline
Goat

Joined: Mon Nov 28, 2005 1:10 pm
Posts: 514
Thanks Mike, point of no return indeed!
Now that I have some perspective I can say those Lelon caps sounded really nice. I would describe them as having natural and fluid human voices. Spoken word sounded great and was easy to follow. Excellent depth of field too. I gotta procure more.
Out of curiosity I will measure |Z| and phase angle of some electrolytics at 10Hz, 100, 1K, 10K, 100K.
Anything on your bench Mike? And where is everybody?


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 130 posts ]  Go to page Previous  1 ... 5, 6, 7, 8, 9  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group