Go Back   Home > Forums > Member Areas > The Lounge

The Lounge A place to talk about almost anything but politics and religion.

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 29th October 2009, 02:23 AM   #2021
diyAudio Member
 
Wavebourn's Avatar
 
Join Date: Sep 2006
Location: Pleasant Hill, CA
Send a message via Skype™ to Wavebourn
Quote:
Originally Posted by Capt Grogg View Post
leds respond to light, will generate noise from any varying light i.e. flouresent or monitor, so do diodes and transistors in glass packages
Yes, they respond to light, but the response is shunted by their own very low dynamic resistance because they are directly biased . Don't worry.
__________________
"Our youth [...] have bad manners, contempt for authority; they show disrespect for their elders, contradict their parents, [...] and tyrannize their teachers. -- Plato, 447-367 BCE
  Reply With Quote
Old 29th October 2009, 03:06 AM   #2022
SY is offline SY  United States
diyAudio Moderator
 
SY's Avatar
 
Join Date: Oct 2002
Location: Chicagoland
Blog Entries: 1
Quote:
Originally Posted by Capt Grogg View Post
leds respond to light, will generate noise from any varying light i.e. flouresent or monitor, so do diodes and transistors in glass packages
Only reverse biased. I think I posted some measurements about this a few years ago; stuck a fluorescent light right next to a lit LED and saw absolutely no effect.

edit: Just saw Wavebourn's post. What he said.
__________________
"The way to deal with superstition is not to be polite to it, but to tackle it with all arms, and so rout it, cripple it, and make it forever infamous and ridiculous."- H. L. Mencken
  Reply With Quote
Old 29th October 2009, 04:25 AM   #2023
anatech is offline anatech  Canada
diyAudio Moderator
 
anatech's Avatar
 
Join Date: Jun 2004
Location: Georgetown, On
Hi Capt Grogg,
I think you must be referring to reverse biased glass diodes only. Those get a case of the nasties when exposed to florescent light. Too bad because we see glass signal diodes everywhere (1N4148, 1N914) in reverse biased applications.

I have not seen this effect from any forward biased LEDs, and you don't really want to reverse bias LEDs. Some have a low reverse breakdown voltage. I have not seen damage from this, but what's the point of doing it?

Anatoliy's explanation is good enough for me. Never thought about this case to be honest with you.

Hi Syn08,
Very good of you do do this, thank you. I'm feeling better about my way of building a current source already! Red LED (cheap) and a BJT.

Now, if you are going to test some J-Fets for this, may I suggest a couple that might be popular as well? In no way do I want to overload you.

J202 - Along with J201. Same family but higher IDSS for the J202.
2N5484 - Recommended to me.
2SK30A - Old number, but recently available
2SK117 - Alternate for 2SK170 in this application, might work better anyway.

The last two are dependent on your having one to try, general interest. The J201 / J202 are JAN low noise parts. Probably not the best for this application.

There was a line of depletion mode MosFets, I'm thinking ND2012L (TO-92), ND2020, ND2410 and ND2406 from Vishay (these days). I'm not suggesting you try all, but possibly if one is available? **** actually, doesn't look like it these days.

Let's try looking at Supertex products some more, but lower power parts like the DN2470, DN2625 or LND150N3G. I don't know if any others exist, but the cool looking Siliconix line of depletion mode mosfets didn't survive the test of time. Darn.

Just thinking out loud, typing as I go. If any of you has an idea, please sing out. That's really why I posted my thought process as I searched for parts. Of these last parts listed, the LND150N3G looks interesting.

There is a good part selector sheet here on the Supertex site. Some interesting application notes are linked from the right, lower part of that page.

-Chris
__________________
"Just because you can, doesn't mean you should" my Wife
  Reply With Quote
Old 29th October 2009, 04:59 AM   #2024
iko is offline iko  Canada
diyAudio Moderator
 
iko's Avatar
 
Join Date: May 2008
Location: Toronto
There's something that doesn't add up here. Several sources point out that best jfets for ccs would have low output conductance. For instance, Vishay app note AN103:

"The best FETs for current sources are those having long
gates and consequently very low goss. The Siliconix
2N4340, J202, and SST202 exhibit typical goss = 2 uS at
VDS = 20 V. These devices in the circuit of Figure 4 will
provide a current source adjustable from 5 uA to 0.8 mA
with internal impedance greater than 2 Mohm at 0.2 mA.
Other Siliconix part types such as the 2N4392, J112, and
SST112 can provide 10 mA or higher current."

Later in the app note they mention the j201 as well.

Yet, this does not align well with syn08's measurement. In particular, it seems that the LEDs are all of a sudden a winner. But what psrr does a LED based ccs offer? This matters too, I think. Any thoughts?
  Reply With Quote
Old 29th October 2009, 06:06 AM   #2025
diyAudio Member
 
john curl's Avatar
 
Join Date: Jul 2003
Location: berkeley ca
Ikoflexer, you have make an important point. It is in the SELECTION of parts and whether you filter or not, that makes the difference.
For example, IF I use a 2SK246Gr, or Bl, then the load resistor can drop to between 3K-5K rather than 15K, used by Syn08. For the record, I have never used more that 5K in my life for a Norton equivalent current source, it just was not necessary. For example a 2SK246 Gr operating at 6ma would give you 2.5 times lower noise, because the Gm would only double from the 1.3ma device and the load resistor would drop to 3K or so. This is from the data sheet. Look it up, everyone. Then 50nV/rt Hz is the typical limit, except that it is easy and usually necessary to use a cap to lower the noise, in any case. Then even the original example with a 15K load is not really a problem, because the cap will effectively filter the noise. In fact, it would filter the noise more effectively than any higher current fet. That pretty much leaves out the 10+ ma parts for all practical purposes.

Last edited by john curl; 29th October 2009 at 06:08 AM.
  Reply With Quote
Old 29th October 2009, 09:16 AM   #2026
syn08 is offline syn08  Canada
Account disabled at member's request
 
Join Date: Aug 2005
Location: Toronto
Quote:
Originally Posted by john curl View Post
For example a 2SK246 Gr operating at 6ma would give you 2.5 times lower noise, because the Gm would only double from the 1.3ma device and the load resistor would drop to 3K or so.
After somebody is spending time and energy, explaining thoretically and confirming experimentally that, for a fixed output voltage, Gm has no impact on the output noise, you are back and pushing the same bull chip.

You haven't read above or you don't get it. One has to remain calm and for the n-th time in the last years conclude: this is a waste of time.
  Reply With Quote
Old 29th October 2009, 09:35 AM   #2027
Bonsai is offline Bonsai  Taiwan
diyAudio Member
 
Bonsai's Avatar
 
Join Date: Jul 2003
Location: Taipei, Taiwan
Syn08, on your LED noise plot I see a pek at 60hz (ok, its more than 100dB down, so its pretty low) - is that wiring noise or what?

BTW, 50nV per root Hz is about 7.2uV of 20Hz to 20Khz wideband noise if my calculation is right.
__________________
bonsai
Amplifier Design and Construction for MUSIC! http://hifisonix.com/
  Reply With Quote
Old 29th October 2009, 09:49 AM   #2028
syn08 is offline syn08  Canada
Account disabled at member's request
 
Join Date: Aug 2005
Location: Toronto
Quote:
Originally Posted by Bonsai View Post
Syn08, on your LED noise plot I see a pek at 60hz (ok, its more than 100dB down, so its pretty low) - is that wiring noise or what?

BTW, 50nV per root Hz is about 7.2uV of 20Hz to 20Khz wideband noise if my calculation is right.
Wiring and relatively high impedance. LM326 has very low impedance (under 1ohm), that's why the 60Hz peak is missing. I could kill all those artifacts by using balanced mode/cabling, but this little experiment is not worth the trouble.

You calculation is correct, and that's about 5% of my Ortofon 0.12mV cartridge. That's how much noise voltage is injected in the gain stage, and it's only the gain stage PSRR to protect against.
  Reply With Quote
Old 29th October 2009, 09:59 AM   #2029
Salas is online now Salas  Greece
diyAudio Chief Moderator
 
Salas's Avatar
 
Join Date: Oct 2002
Location: Athens-Greece
Quote:
Originally Posted by Bonsai View Post
Syn08, on your LED noise plot I see a pek at 60hz (ok, its more than 100dB down, so its pretty low) - is that wiring noise or what?

BTW, 50nV per root Hz is about 7.2uV of 20Hz to 20Khz wideband noise if my calculation is right.
-125dB I think. The Led is going to get 5-20mA CCS range from various Jfets. This one is just from 2k7. There are sequel graphs to see I mean, when syn feels like it.
  Reply With Quote
Old 29th October 2009, 10:12 AM   #2030
syn08 is offline syn08  Canada
Account disabled at member's request
 
Join Date: Aug 2005
Location: Toronto
Quote:
Originally Posted by Salas View Post
-125dB I think.
-103dB
  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 06:14 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2