Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Pass Labs This forum is dedicated to Pass Labs discussion.

M2HA
M2HA
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 1st February 2018, 08:42 PM   #21
Hikari1 is offline Hikari1  United States
diyAudio Member
 
Join Date: Jun 2012
Location: Seattle
Quote:
Originally Posted by EUVL View Post
Because he was still using 10k for R15 & R23.
This results in only 5V bias for the Fairchild MOSFETs which is of course not sufficient.
In changing R6 & R24, you must also change the values of R15 and R23 by the same amount.
Unless of course you use Toshiba 2SK2013 / 2SJ313, which require much less bias voltage.

But you can readily test all these in LT Spice.
Then you have a good idea what happens by changing which parameter, before you solder anything.


Patrick
In 6L6's original thread on Wayne's headphone amp he states we can use Fairchild or IRF or Toshiba mosfets with the 10K resistor values.

New PassDIY Headphone Amp (Coming Soon)

Is this because the voltage gain is an opamp with a lower output impedance than the transformer?

I am not asking out the kindness of my heart...I have been trying to adapt Juma's current mirror preamp (attached)which sounds great to Wayne's output stage.Kind of a UGS-like situation in where the current mirror frontend will sit on top of the mosfit output stage. I am starting to put it together (I am listening to just Juma's front end now) but now Iam worrying about the resistors...
Attached Images
File Type: jpg WhammyRev1.jpg (196.4 KB, 277 views)
File Type: png Juma's CM preamp.png (87.3 KB, 262 views)
File Type: jpg IMG_0610 2.jpg (736.2 KB, 269 views)
  Reply With Quote
Old 2nd February 2018, 12:59 AM   #22
EUVL is offline EUVL  Europe
diyAudio Member
 
Join Date: Oct 2003
> Is this because the voltage gain is an opamp with a lower output impedance than the transformer?

Yes.

> but now Iam worrying about the resistors...

So what do you think the output impedance of your frontend is ?
Have you done some simulations (or better calculations) to find out ?

On top of that, why would you need to add a buffer instead of change the 2nd stage to high bias ?
And if you must have a buffer, why not use the 2nd leg to generate the bias voltage, like in the UTHAiM with NFB ?


Patrick
__________________
xen-audio.com
  Reply With Quote
Old 2nd February 2018, 01:18 AM   #23
EUVL is offline EUVL  Europe
diyAudio Member
 
Join Date: Oct 2003
This should be of some help :
The All-FET Line Amp

Especially Fig. 7 & 9.


Patrick
__________________
xen-audio.com
  Reply With Quote
Old 2nd February 2018, 01:42 AM   #24
Hikari1 is offline Hikari1  United States
diyAudio Member
 
Join Date: Jun 2012
Location: Seattle
Quote:
Originally Posted by EUVL View Post
> Is this because the voltage gain is an opamp with a lower output impedance than the transformer?

Yes.

> but now Iam worrying about the resistors...

So what do you think the output impedance of your frontend is ?
Have you done some simulations (or better calculations) to find out ?

On top of that, why would you need to add a buffer instead of change the 2nd stage to high bias ?
And if you must have a buffer, why not use the 2nd leg to generate the bias voltage, like in the UTHAiM with NFB ?


Patrick
The output of Jumaís line amp is 250R or so.

I havenít done simulations...itís quite a learning curve but Iíve tried to use LTsoice. I need some more time.

Iíve seen you UTHAim. I agree itís probably the way to go. But I already have Jumaís current mirror built and it sounds good. Basically, I had most the parts to do Wayneís output stage so why not? Seemed like a good idea at the time. I donít necessarily need the added current drive but I want to try it with headphones. Iíll read Ernoís pdf although he tends to use a lot more toshiba jfets than I have.

I like to build and try a lot of preamps. Much cheaper and less speaker dependent than power amps.
  Reply With Quote
Old 2nd February 2018, 01:49 AM   #25
EUVL is offline EUVL  Europe
diyAudio Member
 
Join Date: Oct 2003
You missed the point.

By adding two resistors to the 2nd leg of your preamp, you can generate the bias voltage for a pair of output devices (MOSFETs or BJTs).
I was not asking you to build a Borbely.
I was asking you to see how Fig. 9 evolved from Fig. 7.
And you can do the same with your preamp and just add 4 resistors and 2 TO220 MOSFETs.


Only my 2 cents,
Patrick
__________________
xen-audio.com
  Reply With Quote
Old 2nd February 2018, 02:05 AM   #26
EUVL is offline EUVL  Europe
diyAudio Member
 
Join Date: Oct 2003
Actually I have to correct myself in post #13.

I overlooked that the secondary of the transformer is connected in series with the primary.
So the impedance of the secondary is not 15k as specified, but 21.6k instead.
The input impedance of the output buffer circuit should therefore not be smaller than 22k.
Hence R6,15,23,24 should be 47k instead.

And even then, the JFETs see a load of 600R, which is quite a lot.


Sorry for overlooking,
Patrick

.
Attached Images
File Type: jpg M2-HPA EUVL V2.jpg (118.1 KB, 245 views)
__________________
xen-audio.com
  Reply With Quote
Old 2nd February 2018, 04:28 PM   #27
Hikari1 is offline Hikari1  United States
diyAudio Member
 
Join Date: Jun 2012
Location: Seattle
Quote:
Originally Posted by EUVL View Post
You missed the point.

By adding two resistors to the 2nd leg of your preamp, you can generate the bias voltage for a pair of output devices (MOSFETs or BJTs).
I was not asking you to build a Borbely.
I was asking you to see how Fig. 9 evolved from Fig. 7.
And you can do the same with your preamp and just add 4 resistors and 2 TO220 MOSFETs.


Only my 2 cents,
Patrick
Thanks Patrick.

After reading barbell's article I think I understand. R15/17 for gates. R22/21 for bias. But Borbely is using laterals...for verticals I would need some source resistors too which is not a problem.

Any hints on the value of R22 and R21 for my application?
Attached Images
File Type: png Screen Shot 2018-02-02 at 9.12.10 AM.png (540.2 KB, 215 views)
File Type: jpg Screen Shot 2018-02-02 at 9.12.20 AM.jpg (139.5 KB, 89 views)
File Type: png Screen Shot 2018-02-02 at 9.28.12 AM.png (84.0 KB, 77 views)
  Reply With Quote
Old 2nd February 2018, 05:19 PM   #28
EUVL is offline EUVL  Europe
diyAudio Member
 
Join Date: Oct 2003
Open a new thread, and learn to use Spice.


Patrick
__________________
xen-audio.com
  Reply With Quote
Old 3rd February 2018, 04:04 PM   #29
fernanwaka is offline fernanwaka  Europe
diyAudio Member
 
Join Date: Mar 2011
Hello, I'm Fernando, a friend of Jose's. who also is making this amplifier.

I want to thank you for your help. In my case, I was able to solve the problems that I had with my M2HA.

The problem in my case, was on the PSU, I had badly oriented transistor. I fixed and I could set the both rails to +/- 15V. I don't need modify the amplifier, both channels are with 10K in R6, R15. R23, R24, and 3R3 in R16, R22. The DC offset was stable and under control using the 1K trimpot RV1.
Click the image to open in full size.
I read all posts, and follow the Patrick recomendation and bypass the C2 and now the sound is more neutral, clear and less distosion, I liked how C2 looked on the amplifier, but sound better without it. Thank you Patrick. I don't understand why in this amplifier is not necesary and in original M2 is need it. I'm clumsy with these things.

In the other hand, If I have free time I want change R6, R15. R23 and R24 to 47K, I'm not sure if R6 and R23 should be a trimpots. I want decrease the bias in the MOSFETs increasing the resistors values R16 and R22, now are 3R3 and the temparture in the heatsink is 51 ļC, we are winter, in summer the tempearature could be increase until 70ļC, I don't want a heat system in the room in summer. Maybe isn't good idea change the bias.

Apologies for my bad english I try to do it the better that I can.
Attached Images
File Type: jpg IMG-20180128-WA0013.jpg (159.4 KB, 102 views)
File Type: jpg IMG-20180202-WA0030.jpg (166.8 KB, 94 views)
  Reply With Quote
Old 3rd February 2018, 05:02 PM   #30
EUVL is offline EUVL  Europe
diyAudio Member
 
Join Date: Oct 2003
> I'm not sure if R6 and R23 should be a trimpots.

Start with 2x 100k 10-turn trimpots near mid-position.
Adjust R6 and R23 so that you get 0V at the transformer output AND the amplifier output.
(Make sure you ground the input first.)
After you are happy the offsets are close to zero and stable, replace them with fixed resistors.

> R16 and R22 now are 3R3 and the temparture in the heatsink is 51 ļC, we are winter,
> in summer the tempearature could be increase until 70ļC, I don't want a heat system in the room in summer.
> Maybe isn't good idea change the bias.

It depends on the headphone you use.
If you use 300R headphones, then 70mA is plenty (7.5R).
If you use 30R headphones then you do want to have 160mA bias.

To reduce temperature, use a more efficient heatsink with vertical fins.
For example Fischer SK47 75mm high.
Fischer Elektronik Profile heatsinks, 1.05 K/W, Aluminium, black anodised SK 47 75 SA | Burklin Elektronik


Patrick
__________________
xen-audio.com
  Reply With Quote

Reply


M2HAHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


New To Site? Need Help?

All times are GMT. The time now is 12:44 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki