Aleph-X builder's thread. - Page 224 - diyAudio
Go Back   Home > Forums > Amplifiers > Pass Labs

Pass Labs This forum is dedicated to Pass Labs discussion.

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 5th February 2013, 09:31 PM   #2231
gl is offline gl  United States
diyAudio Member
 
Join Date: Dec 2003
Location: Sierra Foothills - California
Hi William,

Very cool. Would you recommend the change?

I remember the the discussion about improving the front end CCS performance by substituting three 3V zeners for the 9V unit. I bought the 3V zeners but never did get around to installing them though.

Graeme
  Reply With Quote
Old 5th February 2013, 11:58 PM   #2232
Eric is offline Eric  United States
diyAudio Member
 
Eric's Avatar
 
Join Date: Nov 2000
Location: Central PA, USA
Thanks for sharing your results, William.

I am curious to hear how offset tracks with temperature now... I am planning to attach the KTY81's directly to my main heat sink right between two of the output mosfets - though I haven't worked out the mechanics just yet.

By the way: what caps are you using for your input? Currently, I'm using cheap Dayton caps from PartsExpress.com, but was looking to replace them with Intertechnik audyn true copper cap. I keep putting it off due to their cost...
__________________
My DIY Basement Theater, Aleph-X Amplifers and Avro Open Baffle Speaker construction pages.

Last edited by Eric; 6th February 2013 at 12:03 AM.
  Reply With Quote
Old 6th February 2013, 06:51 AM   #2233
Salomon is offline Salomon  Germany
diyAudio Member
 
Salomon's Avatar
 
Join Date: Jun 2009
Default Fixing all Aleph-X Offsets

I am sure with with the right combination of PTC it is possible to completely compensate the temperature drift.
It is my opinion the type of zener diodes does not make such a big difference.
I only can speak for the normal version of the Aleph-X circuit with IRF9610.
For all who use a Aleph-X with JFETs I am sure William will find a good part combination.

I use 2 pieces of the KTY81 PTC. I chose this device because it has a very linear characteristic.
With three pieces and a trimmer in series of the third part it is possible to compensate for the temperature drift completely.
Please see my attachement
The new trimmer is for right adjusting between over compensation and under compensation.
Try a 2K trimmer and begin the adjustment with adjusting the biggest value.
First you have to adjust he normal offset trimmer and to judge the remaining offset drift from cold to warm.
If the PTCs don`t compensate completely the temperature drift, you have to decrease the adjustment of the additional 2k trimmer to a smaller value.
The right adjustment is a combination of the adjustment of both trimmers.
If changing the adjustment of the 2K trimmer it is important immediately to re-adjust the normal offset trimmer to zero offset.
A good thermal coupling of the PTCs to the temperature of the output FETs (heat sink) is very important !
*
We are speaking here about the offset that we measure from each output pin to ground. This one you can compensate with the PTCs.
The offset measured between both output pins is depending of the matching of the IRF9610, its Drain Resistors and the matching of the parts of the output stages.
If you want to easy get a little remaining offset to zero it is possible to solder a resistor over the Drain resistor (390R) who you can measure the bigger voltage of both IRF9610.
Try to determine the right value with first solder a 10K trimmer over the Drain resistor.
Begin to trim from highest value (10K or more) to a lower value.
If you measure 0mV over both output pins you have found the right value. Solder out the trimmer and measure the value. Now you can change it against a normal resistor and solder it over the normal Drain resistor. This choice is better for long time offset stability.
Now the offset between the output pins allways will be stable at 0mV or less than 10mV . Cold or warm amp will not make any difference.
*
Sorry for my bad English, but it is important for me to make this information accessible to all persons in the world who love the Aleph-X like I love it.
*
Here you find some pictures of my Alep-X amp.
Class A monster: Aleph-X by Horst Wolf Audio Tweak
Attached Images
File Type: jpg Compensation.jpg (20.2 KB, 227 views)
  Reply With Quote
Old 6th February 2013, 06:53 AM   #2234
diyAudio Member
 
wuffwaff's Avatar
 
Join Date: Aug 2001
Location: Ingolstadt Germany
Hi,

had the amps on quite long last night and measured around -2V of offset. Measured again this morning (cold) and had around 3V in the left channel and 1.5V on the right channel. So the voltage swing is around 5V left and 3,5V right (remember there are different R24 in both amps)

This means, that I have to raise the influence of the PTC a bit more, lower
R24 and raise R26.

Before the swing was around 8V (R46/47 being 22k). Changing R46/47 to 27k would have changed this to around 10V.

@Eric: I coupled the KTY to the current source fet as I found out this had by far the biggest influence on absolute DC Offset. Cooling and heating the outputs, diff pair and some more had almost no influence.
On the input are a pair of black gate F? (the red ones, bipolar)

@Graeme: Yes I would because it means you can raise the value of McMillan resistors which (I think) has a positive influence on sound quality.
I did install the 3 zeners but somehow that didnīt work as it should have (giving a negative tempco)

William
__________________
een ooievaar is geen konijn want zijn oren zijn te klein!
  Reply With Quote
Old 6th February 2013, 07:43 AM   #2235
diyAudio Member
 
wuffwaff's Avatar
 
Join Date: Aug 2001
Location: Ingolstadt Germany
Hi Horst,

we somehow posted at the same time

Out of the results of the two amps I can calculate a third setting to get the drift to (almost zero). Will be something like R24=680R plus KTY and R24 = 820R. Will try this evening.

In my amp the temperature influence of the output fets is almost zero. I've put one side of the amp (current source and output) on one heatsink and the other side on the other heatsink. This way Vgs changes over temp will compensate each other.

William
__________________
een ooievaar is geen konijn want zijn oren zijn te klein!
  Reply With Quote
Old 6th February 2013, 08:47 AM   #2236
Salomon is offline Salomon  Germany
diyAudio Member
 
Salomon's Avatar
 
Join Date: Jun 2009
Hi William,

i am sure both stages have a temp offset drift
It is clear that the drift of the input current source who changes the Drain voltage of the IRF9610 has more effect because this changing of the drain voltage will be multiplied by the gain of the output stage.
It may be a good idea to be heated driver or heated input current source stage taking as reference temperature for the PTC
For me it was mechanically easier to take the temperature output stage as the reference.
I think with both variants you get a good offset compensation.
  Reply With Quote
Old 7th February 2013, 06:43 AM   #2237
diyAudio Member
 
wuffwaff's Avatar
 
Join Date: Aug 2001
Location: Ingolstadt Germany
Hi,

tried another combination yesterday. R24 750R, R26 820R. Voltage swing is under 1 Volt now but..........

As I discovered a few years ago (and forgot in the meantime) the absolute DC Offset depends on the mains voltage (at least in my amps). If the voltage goes up, the absolute offset goes negative and vv. This also explains the little jumps (0,5-1V) I see when observing the abs.dc offset.

This means more negative late at night and more positive early at night. I will see if I can measure it with a variac from 220-240V.

I will also try to up the McMillans from 27k to 33k.

William
__________________
een ooievaar is geen konijn want zijn oren zijn te klein!
  Reply With Quote
Old 7th February 2013, 10:09 AM   #2238
Salomon is offline Salomon  Germany
diyAudio Member
 
Salomon's Avatar
 
Join Date: Jun 2009
Hi William,

very interesting!
I am sure you will find the best values for a perfect compensation.
Maybe its better to use only one of piece of the PTCs because you a using a JFET stage if you get over compensation.
Using the normal IRF9610 circuit it is not possible to get over compensation.
If users with IRF9610 want to make a full compensation, the have to take three pieces of the PTCs and solder the second in series with a trimmer. See my other post.
I am also interrested how it sounds with the bigger values of the McMillans.

Thank you for all and good luck.

Horst
  Reply With Quote
Old 7th February 2013, 10:23 AM   #2239
diyAudio Member
 
wuffwaff's Avatar
 
Join Date: Aug 2001
Location: Ingolstadt Germany
Hi,

just found something I wrote in 2006.........

QUOTE=wuffwaff;867384]Hi,

it does contribute since abs. dc offset is set with the neg. rail as a reference. When this changes, abs. dc changes.
In my amp ca. 3V of mains (232-229) change, changes offset by 1.5V

William[/QUOTE]

Will still try if this is also true for the Jfet input with larger drain resistors.

@Horst, the bigger McMillans do sound better.

William
__________________
een ooievaar is geen konijn want zijn oren zijn te klein!
  Reply With Quote
Old 7th February 2013, 03:48 PM   #2240
diyAudio Member
 
wuffwaff's Avatar
 
Join Date: Aug 2001
Location: Ingolstadt Germany
Hi,

just measured the right channel with 39k McMillan resistors:

230V 0V
225V +3,5V
235V -3,5V

with 27k:
0V
+2V
-2V

and the left channel with 33k:
0V
+3V
-3V

DC behavior over temp is very good now with almost no drift if I keep the mains voltage constant (with a variac)

So the absolute offset follows the supply voltage changes quite well depending on the value of the McMillans.
To compensate for this the current source should deliver a bit less current when the supply voltage goes up. Maybe someone has a nice idea how to do that?
I'm thinking but didn't come very far yet.

I will leave it like that for a while (left 33k, right 39k) and observe a bit, choose one off both or go back to 27k.

william
__________________
een ooievaar is geen konijn want zijn oren zijn te klein!
  Reply With Quote

Reply


Hide 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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



New To Site? Need Help?

All times are GMT. The time now is 01:12 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright Đ1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2