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

Solid State Talk all about solid state amplification.

Looking for a small signal NPN with no Quasi Saturation: MEASURED DATA
Looking for a small signal NPN with no Quasi Saturation: MEASURED DATA
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 24th June 2016, 03:48 PM   #11
Bob Cordell is offline Bob Cordell  United States
diyAudio Member
 
Join Date: Sep 2006
Default Transistors wanted

As some of you know, I'm working on the second edition of my book. One of the things I'd like to do is to expand and bring up to date some of the transistors I am discussing. In perticular, I want really good transistors that can still be purchased.

Here are the kinds of your favorite candidates I want you to suggest:

1. Good IPS transistors in TO-92 packages that can handle at least 70V, have ft > 100 MHz, and have beta >100.

2. Good VAS and predriver transistors, likely in TO-126 packages, that can handle at least 130V, have ft > 100 MHz, and have beta > 100. The 2SC3503 meets these requirements, but I am interested in knowing of others, since availability of good discrete transistors can be spotty.

3. Good driver transistors, likely in TO-220 packages that have similar capabilities to those above, but with higher dissipation and SOA capability.

4. Good power transistors, likely in TO-247, TO-264 and MT-200 packages that have characteristics similar to, or better than, the MJL3281 and MJL1302.

Cheers,
Bob
  Reply With Quote
Old 24th June 2016, 04:46 PM   #12
Mark Johnson is offline Mark Johnson  United States
diyAudio Donor
diyAudio Member
 
Mark Johnson's Avatar
 
Join Date: May 2011
Location: Silicon Valley
Looking for a small signal NPN with no Quasi Saturation: MEASURED DATA
Quote:
Originally Posted by Bob Cordell View Post
It looks like you are assuming an undegenerated VAS transistor fed by a fairly high impedance
I am assuming a cascode current source like the attached, upside down using NPNs. diyAudio member kgrlee points out that, to the extent node "A" and node "X" are zero impedance voltage sources, the two BJTs operate with constant VBE rather than constant Ib. So I made a stab at calculating the worst case (largest possible) value of bypass capacitor from node X to the positive rail, which is necessary to make the impedance sufficiently low.

_
Attached Images
File Type: png Image1.png (54.7 KB, 869 views)
  Reply With Quote
Old 24th June 2016, 05:15 PM   #13
mlloyd1 is offline mlloyd1  United States
diyAudio Member
 
mlloyd1's Avatar
 
Join Date: Feb 2001
Location: Northern Iliinois
Looking for a small signal NPN with no Quasi Saturation: MEASURED DATA
Mark:

Thanks for collecting and sharing this data.
Might you have any such data for 2SA1349 or 2SC3381 monolithic duals that could be shared?
I'm curious about these in 4T Wilsons at the currents being discussed.

Thanks,
mlloyd1
  Reply With Quote
Old 24th June 2016, 06:42 PM   #14
scott wurcer is offline scott wurcer  United States
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: Belmont MA
Quote:
Originally Posted by Mark Johnson View Post
Scott, the MPSA18 may be an unusual case. It's got a low datasheet breakdown voltage (45V) yet tons of quasi-saturation.
Never really explored that device. I just wanted to make the point that the quasi-sat is a necessary evil for many high voltage processes and every circuit position needs to be optimized. One handle is current density so a medium power 120V + device might be the best Vas choice. Did you look at any of those like the 2N3440 (old school)? A TO92 at 10mA and ~60 - 100V is marginal on PD in the first place.
__________________
"The way up and the way down are one in the same"
  Reply With Quote
Old 24th June 2016, 07:24 PM   #15
Mark Johnson is offline Mark Johnson  United States
diyAudio Donor
diyAudio Member
 
Mark Johnson's Avatar
 
Join Date: May 2011
Location: Silicon Valley
Looking for a small signal NPN with no Quasi Saturation: MEASURED DATA
Default more small signal NPNs @ 10mA

I had eight more NPNs in TO-92 packages. Turns out that three of them exhibit quasi saturation. Images attached.

(ZTX851 is the device called out in Art Of Electronics 3rd Edition, for its supremely low rbb (hence supremely low voltage noise) performance).

_
Attached Images
File Type: png 2SC6043.PNG (29.3 KB, 891 views)
File Type: png BC547.PNG (36.1 KB, 879 views)
File Type: png KSC1845E.PNG (37.1 KB, 859 views)
File Type: png PN2369.PNG (36.7 KB, 858 views)
File Type: png ZTX457.PNG (35.2 KB, 462 views)
File Type: png ZTX851.PNG (30.5 KB, 510 views)
File Type: png ZTX853.PNG (28.3 KB, 501 views)
File Type: png ZTX1056A.PNG (31.2 KB, 484 views)
  Reply With Quote
Old 24th June 2016, 07:30 PM   #16
Mark Johnson is offline Mark Johnson  United States
diyAudio Donor
diyAudio Member
 
Mark Johnson's Avatar
 
Join Date: May 2011
Location: Silicon Valley
Looking for a small signal NPN with no Quasi Saturation: MEASURED DATA
Default some medium- and high-current transistors

I measured some medium-power (TO-126) and high-power transistors (TO-220) to see whether they exhibited any quasi saturation. Only two of them did.

_
Attached Images
File Type: png 2SC3503F.PNG (33.6 KB, 545 views)
File Type: png 2SC3601E.PNG (28.3 KB, 461 views)
File Type: png 2SC4793.PNG (30.1 KB, 441 views)
File Type: png 2SC5171.PNG (27.8 KB, 431 views)
File Type: png 2SC5200.PNG (26.8 KB, 430 views)
File Type: png 2SC6144.PNG (32.4 KB, 420 views)
File Type: png 2SD882.PNG (31.6 KB, 408 views)
  Reply With Quote
Old 24th June 2016, 07:31 PM   #17
Mark Johnson is offline Mark Johnson  United States
diyAudio Donor
diyAudio Member
 
Mark Johnson's Avatar
 
Join Date: May 2011
Location: Silicon Valley
Looking for a small signal NPN with no Quasi Saturation: MEASURED DATA
Default the rest of the medium- and high-current NPNs

here
Attached Images
File Type: png BD139.PNG (28.2 KB, 453 views)
File Type: png D44H11.PNG (34.5 KB, 410 views)
File Type: png KSC2690AY.PNG (30.6 KB, 395 views)
File Type: png KSD1691G.PNG (28.1 KB, 391 views)
File Type: png MJE200.PNG (29.0 KB, 380 views)
File Type: png MJL3281A.PNG (27.2 KB, 392 views)
  Reply With Quote
Old 24th June 2016, 07:57 PM   #18
scott wurcer is offline scott wurcer  United States
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: Belmont MA
Quote:
Originally Posted by Mark Johnson View Post

(ZTX851 is the device called out in Art Of Electronics 3rd Edition, for its supremely low rbb (hence supremely low voltage noise) performance).
This all continues to make sense, design for extra low rbb ends up being more die area, sort of like a bunch of paralleled devices (see for instance the folklore around the LM3XX? matched pair). The current density is then low and Vcesat, etc. all benefit at a loss in speed due to higher parasitic capacitances and the fact that you operate away from peak ft in many cases.
__________________
"The way up and the way down are one in the same"
  Reply With Quote
Old 24th June 2016, 09:41 PM   #19
kgrlee is offline kgrlee
diyAudio Member
 
Join Date: Nov 2011
Location: Cooktown, Oz
Quote:
Originally Posted by Mark Johnson View Post
I am assuming a cascode current source like the attached, upside down using NPNs. diyAudio member kgrlee points out that, to the extent node "A" and node "X" are zero impedance voltage sources, the two BJTs operate with constant VBE rather than constant Ib. So I made a stab at calculating the worst case (largest possible) value of bypass capacitor from node X to the positive rail, which is necessary to make the impedance sufficiently low.
If the application is the CCS for the LTP i/p of a power amp, I don't think you will gain much with fancy CCS versions.

There's at most a couple of volts swing at the LTP so the CCS never gets near sat. More important to look at small (and not wonky) Ccb of the i/p transistors.

Besides if the CCS tail isn't 'perfect', it'll only introduce some nice 2nd harm. and there's so much feedback at very LF that I doubt you'll even see that

There's more reason in the CCS for a VAS but even then, the more complex CCSs take up another volt or so and you have very little return for the greater complexity. If you're lucky, you might get an extra 1dB in your power spec for supa dupa THD.

But 2sc3503f looks worrying with quasi-sat below Vce 10V, especially when used for the VAS
_____________________

On the subject of whether the VAS is fed by a current or a voltage, IIRC, I made a comprehensive study in my tpc-vs-tmc-vs-pure-cherry thread.

The answer is as expected, "it depends".

What does come out of that is driving the VAS with an emitter follower, ala Self, does feed it with a 'voltage'.

Apart from the expected extra gain, an important benefit is that the 'voltage' drive makes the wonky Ccb of the VAS much less important so a cascode VAS becomes unnecessary.
  Reply With Quote
Old 24th June 2016, 11:23 PM   #20
Mark Johnson is offline Mark Johnson  United States
diyAudio Donor
diyAudio Member
 
Mark Johnson's Avatar
 
Join Date: May 2011
Location: Silicon Valley
Looking for a small signal NPN with no Quasi Saturation: MEASURED DATA
Quote:
Originally Posted by scott wurcer View Post
... design for extra low rbb ends up being more die area, sort of like a bunch of paralleled devices ... The current density is then low and Vcesat, etc. all benefit
Is this because you devote more top surface area to base contacts and their surrounds? Area which is stolen from emitter area? The ratio (total emitter area)/(total die area) goes down, and this is a rough approximation of current density?

Perhaps you also lay out the emitter fingers at their minimum possible width (min contact + 2*min_surround)? Then you lay out base contacts the full finger length, on both sides of each finger, to reduce the distance between the "active base" vertical region, and the base contact?

Base-emitter-base subassembly fingers would be stacked side by side like cigarettes in a box, with collector sinkers laid out parallel to the BEB fingers and placed every "N" fingers for the largest value of N you dared? An N which balances layout efficiency vs. collector resistance?

I can't imagine that a checkerboard layout of emitters and bases, with basket weave diagonal metallization of the base contacts and emitter contacts, would be as efficient. The 2D array of emitters would have less total area than the 1D array in the cigarette/finger approach.
  Reply With Quote

Reply


Looking for a small signal NPN with no Quasi Saturation: MEASURED DATAHide 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
Measured data on the J105 Nchannel JFET Mark Johnson Solid State 22 12th June 2017 07:56 PM
Sanken 2SC4883 and 2SA1859 SPICE models with quasi-saturation keantoken Software Tools 7 8th August 2014 08:13 AM
Small signal NPN transistor 2SC2458GR or 2SC1740S-R srinath Parts 12 18th April 2012 01:25 PM
Need an all NPN (Quasi) ops power amp JojoD818 Solid State 18 25th December 2003 03:04 PM
Quasi-Saturation rarya Chip Amps 0 27th November 2003 06:17 AM


New To Site? Need Help?

All times are GMT. The time now is 10:35 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Copyright ©1999-2020 diyAudio
Wiki