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

Equipment & Tools From test equipment to hand tools

My version of the G = 1000 low noise measurement amp (for Ikoflexer).
My version of the G = 1000 low noise measurement amp (for Ikoflexer).
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 13th July 2018, 05:05 PM   #441
Waly is offline Waly
diyAudio Member
 
Join Date: Jan 2011
Location: moderation crapper
Quote:
Originally Posted by gerhard View Post
They have soared to €144 in the mean time and I could not ignore the 400 mOhm.

BTW who would think that adding a ferrite bead to an amplifier input
(Würth from LTspice lib) would f* up the loop gain of an amplifier so completely.
Damping it with the 5 Ohms makes everything nice again.

I had the hope that I could get rid of the gate voltage divider / coupling C for
the cascode/bootstrap, but the depletion MOSFET is not depleted enough. It leaves
only 1 V at the drain of the input JFETs.

The opened loop goes from V5 to the output of U4. R14 is a first order approximation
of the feedback divider.

Reproduced with permission from the author. This one, as shown, won't match your noise requirements (is about 0.3nV/rtHz) but can be fed from an unregulated (or LM78xx regulated) power supply, without an additional noise pip, hence virtually no limit of the number of devices you can parallel. Feel free to use the principle if you find it valuable, only careful with the servo pole-zero distributions, they must more or less match/cancel to avoid a LF gain bump. Without the 1mF/1k LF pass of the servo output, the servo LF noise still creeps in the signal chain.


I've tried it myself and it works a treat. Yes, adding a few ohms in parallel with the input inductors is a good idea.
Attached Images
File Type: jpg LNA.jpg (188.8 KB, 83 views)
  Reply With Quote
Old 13th July 2018, 09:20 PM   #442
scott wurcer is offline scott wurcer  United States
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: dorchester ma
Please tell me R129 is not critical
__________________
"The question of who is right and who is wrong has seemed to me always too small to be worth a moment's thought, while the question of what is right and what is wrong has seemed all-important."
  Reply With Quote
Old 13th July 2018, 10:46 PM   #443
Waly is offline Waly
diyAudio Member
 
Join Date: Jan 2011
Location: moderation crapper
Quote:
Originally Posted by scott wurcer View Post
Please tell me R129 is not critical
Can be obviously omitted for a 0.01% total 99ohm.
  Reply With Quote
Old 14th July 2018, 03:45 PM   #444
1audio is offline 1audio  United States
diyAudio Member
 
Join Date: Mar 2004
Location: SF Bay Area
My version of the G = 1000 low noise measurement amp (for Ikoflexer).
Quote:
Originally Posted by gerhard View Post
I had the hope that I could get rid of the gate voltage divider / coupling C for the cascode/bootstrap, but the depletion MOSFET is not depleted enough. It leaves only 1 V at the drain of the input JFETs.
1 volt actually should be OK for the Jfet and its internal heat will be less. I usually get closer to 3V in that configuration. What current are you running the Jfet at? There is a rapid diminishing return as you increase the current. Going above 5 mA usually doesn't get much return but I have not played with that FET in years.
__________________
Demian Martin
Product Design Services
  Reply With Quote
Old Yesterday, 11:38 PM   #445
gerhard is offline gerhard  Germany
diyAudio Member
 
gerhard's Avatar
 
Join Date: Oct 2005
Location: near Stuttgart, south Germany
Given the price of the Interfet devices, I feel it is justified to let them sweat
a little bit to get as much from them as possible. I know the sqrt(sqrt()) rule :-)
They'll get a nice Thermalloy cooler anyway, if only to provide a thermal low pass.
As long as a pack of Lithium Ion cells works for some hours, that's OK for me.

The junction capacitance explodes @ 1V; half the C would allow twice
the transistors with a resulting sqrt(2) noise advantage.
It is 20% here, and there, and there; I just want to fathom out how far I get
peeling the onion.

Some other depletion FETs from this week's DigiKey box of wonders seem
to fit better. The Infineon devices have much less spread, that would be
nice for a CCS, but the voltage is quite small for a cascode/bootstrap.

That Locky-Z curve tracer delivers questionable results from time to time,
esp. at low voltage / high current. Weird.

Cheers, Gerhard

(please feel free to correct me if I invent ungrammatical constructions.
It won't get better otherwise!)
Attached Images
File Type: png ixys_cpc3703.png (377.1 KB, 13 views)
File Type: png dn3525_bsp149.png (339.3 KB, 13 views)
__________________
"That's right!" shouted Vroomfondel, "we demand rigidly defined areas of doubt and uncertainty!" "We don't demand solid facts! What we demand is a total absence of solid facts. I demand that I may or may not be Vroomfondel!"

Last edited by gerhard; Today at 12:02 AM.
  Reply With Quote
Old Yesterday, 11:57 PM   #446
gerhard is offline gerhard  Germany
diyAudio Member
 
gerhard's Avatar
 
Join Date: Oct 2005
Location: near Stuttgart, south Germany
Quote:
Originally Posted by Waly View Post
Yes, adding a few ohms in parallel with the input inductors is a good idea.
Yes, it limits the damage done by the on-board ferrite bead, but an external
undampened inductance still will make the amplifier oscillate. The negative
impedance, when looking into the gate, is still there.
__________________
"That's right!" shouted Vroomfondel, "we demand rigidly defined areas of doubt and uncertainty!" "We don't demand solid facts! What we demand is a total absence of solid facts. I demand that I may or may not be Vroomfondel!"
  Reply With Quote
Old Today, 12:21 AM   #447
dimitri is offline dimitri  United States
diyAudio Member
 
dimitri's Avatar
 
Join Date: Oct 2002
Location: retired
Send a message via ICQ to dimitri
Quote:
The junction capacitance explodes @ 1V
Gerhard, would you care to elaborate?
  Reply With Quote
Old Today, 02:07 AM   #448
gerhard is offline gerhard  Germany
diyAudio Member
 
gerhard's Avatar
 
Join Date: Oct 2005
Location: near Stuttgart, south Germany
That is just normal varactor behavior.
See for example the data sheet of the process that makes the IF3601 / IF3602.
The gate/drain capacitance is unlikely to behave much better.

Cheers, Gerhard
Attached Images
File Type: png xxxxx.png (47.8 KB, 5 views)
__________________
"That's right!" shouted Vroomfondel, "we demand rigidly defined areas of doubt and uncertainty!" "We don't demand solid facts! What we demand is a total absence of solid facts. I demand that I may or may not be Vroomfondel!"

Last edited by gerhard; Today at 02:11 AM.
  Reply With Quote
Old Today, 02:16 AM   #449
1audio is offline 1audio  United States
diyAudio Member
 
Join Date: Mar 2004
Location: SF Bay Area
My version of the G = 1000 low noise measurement amp (for Ikoflexer).
Here is some serious work on modeling the capacitance of gate source and gate drain vs voltage. http://www.iaeng.org/publication/WCE..._pp771-775.pdf

The change is significant (on all semis) and C goes down with voltage. Its also a distortion generating parameter that can't be removed with simple feedback. Cascode is your friend but with 1V and big JFET gates the C is higher. I usually get about 3V with lower current and in this case higher noise. However even though theoretically the noise goes down as you approach 0Vds with the high current fets it doesn't quite hold. One of the standard texts explained the phenomenon but I long ago lost track of the text. A Quantech is really helpful in exploring this. (I need to get mine back from Constellation. . .) Cooling also helps. The big JFET was created for the infrared mapping satellite I was told by Interfet. In that application it was cooled to a very low temperature. But it all may be marketing BS. They also told me they use a 2" semi line. That allowed me to buy a complete waver and have them make custom pairs to my spec. I don't know it they are that accommodating any more.
__________________
Demian Martin
Product Design Services
  Reply With Quote

Reply


My version of the G = 1000 low noise measurement amp (for Ikoflexer).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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Filters For Low-Noise Measurement Preamplifier bryman79 Solid State 63 25th July 2011 06:24 PM
preamp supply LOW power LOW noise LOW cost drdagor Power Supplies 1 13th September 2010 11:39 PM
Ultra low noise measurement amp tschrama Solid State 20 22nd September 2008 04:28 PM
New version of Audio Tester has T/S Measurement..... venki Multi-Way 1 18th September 2005 03:09 PM
Fs Measurement - Baffle, low ohm amp? kimbo Multi-Way 27 30th January 2005 03:01 AM


New To Site? Need Help?

All times are GMT. The time now is 04:38 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