I have a few from the last of Grey's group buy and they're all pretty "spot on" the same, as per his standard test.
I had a stab at maching myself could you please post a link to Gray standard test method I am shure I got a lot to learn from there.
Tanks
Al
From memory, I think Gray tested them using Nelson's technique, it's in one of his papers. I'll see if I can find it - might take me awhile - been a couple years.
As per Patrick's advice, the ones from Gray did need to be further tested at the 250mA current for this amp - mine were pretty close, at the higher temp (50*C) my amp runs at. It does make quite a difference.
As per Patrick's advice, the ones from Gray did need to be further tested at the 250mA current for this amp - mine were pretty close, at the higher temp (50*C) my amp runs at. It does make quite a difference.
one can measure a single device at a single current value.
That is easy and very unrepeatable. Temperature affects the results so much, the result is little better than worthless.
If you can measure a pair of devices at equal temperature to each other, at a single current. Then you can compare the two devices. That's the bit that Euvl alluded to with his heatsink.
If you can compare a pair of devices at equal temperature and at a few different currents then even better.
That is easy and very unrepeatable. Temperature affects the results so much, the result is little better than worthless.
If you can measure a pair of devices at equal temperature to each other, at a single current. Then you can compare the two devices. That's the bit that Euvl alluded to with his heatsink.
If you can compare a pair of devices at equal temperature and at a few different currents then even better.
Qusp, how many do you need - I have a few from the last of Grey's group buy and they're all pretty "spot on" the same, as per his standard test.
Well I'm planning on making both daox and 2 x circlotron 2 stage x (plus some devices fory mate Tom who's helping me put with the pcb so he can have one too. . So that would make 12, I bought some last night from deep surplus they are starting to run out in the lower values, so bought 20 x 1v 10 x .89 8 x .92 and 16 x 1.1. But will buy some of yours as well if you don't mind. I bought that many just in case they really do run out, as I would like to have a go with a simple xd iv once I have made a couple of established designs even thought of the idea of a circlotron portable amp for iems
So yeah that would be fantastic, even if you just have 5 or 10 local shipping is pretty sweet too. I'm literally in the middle of moving, so if you pm me or email, I may not get back straight away, as I'm on my iPhone now and it's a pita posting. Will be back online once I move the mac out there Thursday, as net is offline here
What's with the hot blood here? I think euvls post On matching, may have just been meant for me, no need to personalize it. Guess you guys have history or something. Because that was pretty out there.
Thanks for the matching tips everyone. I Like the idea of using half the real circuit, have seen the pass method in zv8-9, but wanted more specific lower current method as headphone Amps aren't really parallel to the 1 and 2a operating point talked about there.
Last edited:
Tanks jameshillj
Your help is very much apreciated.
I am making same progress on testing as Papa method and as it was mentioned as Grey standard I would like to have a go to test as he does shurely if it is remembered after a couple of years is well woth having a look.
I am curentley using a 100000 count meter and a digital termometer capable of 0.1C is not lab stuff or sometink I can borrow from work but should be near enough to what the majority of as have.
I can get quite close to the 3 decimal place once the temperature setles but the 4th decimal keep vandering around.
Don't get me wrong 3 decimal is good enough I think especialy as I will then test the lot once built as Papa say "listen to it", and shuld help getting the worst ones out of the way again not all of has can count on trained ears.
Any way to make the point clear Gray test has been labeled as standard (maybe right maybe not ) you certanly remember about it from 2 years ago this is good enough for me to ask for a pointer.
If this help to get rid of 10 or 20 % of the mossfet to test once built is surely worth looking at.
Furthermore I am thinking that other members may find this usefull especialy as we are a collection of peoples whith different background and pokets and not all of have a record in the top 10 or a curve tracer.
Tanks
AL
Your help is very much apreciated.
I am making same progress on testing as Papa method and as it was mentioned as Grey standard I would like to have a go to test as he does shurely if it is remembered after a couple of years is well woth having a look.
I am curentley using a 100000 count meter and a digital termometer capable of 0.1C is not lab stuff or sometink I can borrow from work but should be near enough to what the majority of as have.
I can get quite close to the 3 decimal place once the temperature setles but the 4th decimal keep vandering around.
Don't get me wrong 3 decimal is good enough I think especialy as I will then test the lot once built as Papa say "listen to it", and shuld help getting the worst ones out of the way again not all of has can count on trained ears.
Any way to make the point clear Gray test has been labeled as standard (maybe right maybe not ) you certanly remember about it from 2 years ago this is good enough for me to ask for a pointer.
If this help to get rid of 10 or 20 % of the mossfet to test once built is surely worth looking at.
Furthermore I am thinking that other members may find this usefull especialy as we are a collection of peoples whith different background and pokets and not all of have a record in the top 10 or a curve tracer.
Tanks
AL
Hi qusp
Tanks for the chance to clear the air and put things in prospective.
First of all sorry for hijacking this tread I have seen the matching bit mentioned and asked a question regarding it.
I regard to the bad blod just have a look at evvery single tread or post I make 2 this morning one here one on the balanced F5.
I think that some one has a score to setle and due to my "bad temper" I got draghed in to it.
Then have a look at any other tread again and again I ask is it raining?
And get told that water is damp.
Just guess who?
Need a clue Quote "is well known to the moderators but he stay in the lmits"
this is getting realy anooing and I probably said to much already.
Andrewt is already in my ignore list.
In regard to Patrick I hope that eventualy we will learn how to live togheter, altrought I am not the first one to have same similar problems with him.
To all the rest of you sorry again for taking this tread over.
Al
PS r ealy enjoided the Accademy joke had it in mind to have a gigle comment....
Suppose I had some to learn from you as well.
Tanks for the chance to clear the air and put things in prospective.
First of all sorry for hijacking this tread I have seen the matching bit mentioned and asked a question regarding it.
I regard to the bad blod just have a look at evvery single tread or post I make 2 this morning one here one on the balanced F5.
I think that some one has a score to setle and due to my "bad temper" I got draghed in to it.
Then have a look at any other tread again and again I ask is it raining?
And get told that water is damp.
Just guess who?
Need a clue Quote "is well known to the moderators but he stay in the lmits"
this is getting realy anooing and I probably said to much already.
Andrewt is already in my ignore list.
In regard to Patrick I hope that eventualy we will learn how to live togheter, altrought I am not the first one to have same similar problems with him.
To all the rest of you sorry again for taking this tread over.
Al
PS r ealy enjoided the Accademy joke had it in mind to have a gigle comment....
Suppose I had some to learn from you as well.
taylor dao distortion
Just came across this thread and have a question about the taylored dao. Has anyone who's built this buffer run a distortion test on it. I'm not interested so much in the thd, as I am in the distortion spectrum at about 1V p-p into a 50 ohm load. Would anyone be willing to do that?
I would build it and do the test but don't have the parts on hand.
Just came across this thread and have a question about the taylored dao. Has anyone who's built this buffer run a distortion test on it. I'm not interested so much in the thd, as I am in the distortion spectrum at about 1V p-p into a 50 ohm load. Would anyone be willing to do that?
I would build it and do the test but don't have the parts on hand.
i ve bring my non-taylored version to schoool for some quick tests and i remember it measured quite well generally, especially bandwith: almost perfect squarewaves from at least 3hz to couple hundreds khz with almost no ringing. It did have a good high frequency overshoot though with a resistive 30 ohm load (only load i tested it with). It also had a slight 2nd harmonic distortion but that s it. As for the answer to your question, i m afraid i cant provide any answer. But i woudnt be surprise if Patrick already ran this particular test. I will however probably do some tests again in january when i ll have the taylored built.
edit: oops. now i remember the ringing was mostly from the generator![Smile :) :)](data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7)
edit: oops. now i remember the ringing was mostly from the generator
Last edited:
Thanks, but yes, I was mostly interested the size of 2nd, 3rd, 5th, and 7th harmonic distortion at 1V p-p into 50 ohm. Can be easily tested with the sound card and the ARTA software:
ARTA Download
ARTA Download
i ll look into that if i have the time. but if it helps, from memory there was no harmonic distorsion whatsoever i could see except a tiny 2nd, all this for the non-taylored version. i dont know how the taylored can differ from it.
anyway, IMO it s worth a listen![Wink ;) ;)](data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7)
edit: i just saw patrick tested the taylored at 2.6v input at post 19. so my testing confirms it measures nicely but brings nothing new...except in my case input voltage was around 2v i think, not sure though.
anyway, IMO it s worth a listen
edit: i just saw patrick tested the taylored at 2.6v input at post 19. so my testing confirms it measures nicely but brings nothing new...except in my case input voltage was around 2v i think, not sure though.
Last edited:
i ve ordered 150 2sj103 from zhoufang here:
http://www.diyaudio.com/forums/vend...you-always-wanted-but-hard-get-they-here.html
if you want some and if you are kind
maybe i can give you a couple when i ll match them, dont know when though...
ps: 2sk170 7(+-1)ma idss is ok instead of j511. see earlier posts for details
http://www.diyaudio.com/forums/vend...you-always-wanted-but-hard-get-they-here.html
if you want some and if you are kind
ps: 2sk170 7(+-1)ma idss is ok instead of j511. see earlier posts for details
Last edited:
2SK170 has too much Yfs. It can work, but you need to redesign the DAO circuit quite a bit.
Google "Tubecad Taylor Current Source" to get some basic information.
You can also follow their approach and use a BJT with the added complcation of bias.
ZhouFang should have 2SJ103.
http://www.diyaudio.com/forums/vend...ted-but-hard-get-they-here-3.html#post2275358
Patrick
Google "Tubecad Taylor Current Source" to get some basic information.
You can also follow their approach and use a BJT with the added complcation of bias.
ZhouFang should have 2SJ103.
http://www.diyaudio.com/forums/vend...ted-but-hard-get-they-here-3.html#post2275358
Patrick
@pidesd, thanks for the offer, but I'll buy some from Shou Fang if it turns out I can't do it any other way.
Thanks Patrick, found the tubecad page.
Sorry to bug you again but when you say 2.6V output, is that p-p, peak, or RMS?
Also, did you measure this on a real implementation of the buffer? The reason I ask is the -206dB; I'm having a hard time coming up with something that would measure such low value. What I mean is, -206dB down from 2.6V is 130pV.
The distortion at 2.6V output into 60 ohm load (AKG 701) is compared to the original DAO Follower :
DAO vs w Taylor
Gain 0.926 0.959
2nd -89dB -67dB
3rd -87db -110dB
4th -99db -120dB
5th -121db -143dB
6th -151db -174dB
7th -188db -206dB
Thanks Patrick, found the tubecad page.
Sorry to bug you again but when you say 2.6V output, is that p-p, peak, or RMS?
Also, did you measure this on a real implementation of the buffer? The reason I ask is the -206dB; I'm having a hard time coming up with something that would measure such low value. What I mean is, -206dB down from 2.6V is 130pV.
- Home
- Amplifiers
- Pass Labs
- Some other Source Follower Configurations