Question about CCS used in Morgan Jones' book. - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

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 2nd December 2014, 08:56 PM   #1
diyAudio Member
 
Join Date: Apr 2014
Location: US
Default Question about CCS used in Morgan Jones' book.

I have issue with Morgan Jones' book page 135 to 139 talking about the way he use CCS. This is really a question for SS people. But I can't count on the SS people have this book. So I post it here first.

1) More specifically Fig 2.49 and Fig. 2.51. As a former analog IC designer, you cannot use current source as load like as drawn. Current source has very high impedance, you cannot try to balance the plate voltage to the voltage he want by matching the top and bottom CCS like he want. Every slight mismatch of current between the top and bottom CCS will cause the plate voltage either go to the top rail or the lowest point the plate can go. You can only establish the quiescent voltage of the plate by local NFB.

2) in page 139 about choosing BJT. That is questionable. Output impedance of BJT can be found simply by looking at the collector curve and Rout=d(Ic)/d(Vce). Most common BJT will have output impedance over 1Mohm. The ones I use can have output impedance to 10Mohm and I used it in my designs. Yes, he is correct you get even higher output impedance using cascode, but I don't think it's necessary. BUT problem with 1) is the killer. It will not work.

I think the way to get high impedance is using mu or Beta follower instead of this.

Please confirm.

Last edited by Alan0354; 2nd December 2014 at 09:01 PM.
  Reply With Quote
Old 2nd December 2014, 09:02 PM   #2
SY is offline SY  United States
diyAudio Moderator
 
SY's Avatar
 
Join Date: Oct 2002
Location: Chicagoland
Blog Entries: 1
Quote:
Originally Posted by Alan0354 View Post
BUT problem with 1) is the killer. It will not work.


Please confirm.
I confirm that it does work. And extremely well. You may want to look at triode curves and put a horizontal load line on them. Consider as well the stiffness of the cascode vs. the single BJT CCS.
__________________
And while they may not be as strong as apes, don't lock eyes with 'em, don't do it. Puts 'em on edge. They might go into berzerker mode; come at you like a whirling dervish, all fists and elbows.
  Reply With Quote
Old 2nd December 2014, 09:12 PM   #3
diyAudio Member
 
Join Date: Apr 2014
Location: US
Quote:
Originally Posted by SY View Post
I confirm that it does work. And extremely well.
I just don't see it. It is a bad practice to do it like this. As I said, BJT CCS have output impedance into mega ohms. Any slight imbalance between the top and bottom CCS will swing the plate to the rail. You might be able to tweak the CCS to try to balance it, but it's not going to be stable, just temperature change can upset the balance.

This is a very fundamental building block in analog IC design. Any internal circuit of opamp uses plenty of this. It is well documented. Nobody try to balance the voltage by adjusting top and bottom CCS like this without NFB.

EDIT:

From looking at the triode curve, you might be able to make Fig.2.51 work as the cathode uses a zener. But there is no way Fig.2.49 work. The tail current is controlled by the bottom CCS, the triode curve cannot affect the current. Then you have to balance the top and bottom. Even if you can balanced, you still have to concern with balance between the two triodes.

Last edited by Alan0354; 2nd December 2014 at 09:17 PM.
  Reply With Quote
Old 2nd December 2014, 09:19 PM   #4
Merlinb is offline Merlinb  United Kingdom
diyAudio Member
 
Merlinb's Avatar
 
Join Date: May 2006
Location: Lancashire
Quote:
Originally Posted by Alan0354 View Post
1) More specifically Fig 2.49 and Fig. 2.51.
You mean the 3rd edition? It seems to be fig 2.54 in the 4th edition.

Quote:
As a former analog IC designer, you cannot use current source as load like as drawn. Current source has very high impedance,
The idea is the the two anode sources are 'soft current sources'. Their internal impedance is not nearly as high as the tail CCS, so they have enough compliance to work. The internal impadances of these sources is considerably reduced due to the dynamic resistance of the diode/zener voltage sources. (Though I agree, thee CCSs give me the willies, even if two of them are soft!)

Quote:
Most common BJT will have output impedance over 1Mohm. The ones I use can have output impedance to 10Mohm and I used it in my designs.
Not sure how you figure that... The rout of a BJT is not likely to be more than a couple of hundred kilohms at typical valve operating currents.

Last edited by Merlinb; 2nd December 2014 at 09:29 PM.
  Reply With Quote
Old 2nd December 2014, 09:25 PM   #5
Merlinb is offline Merlinb  United Kingdom
diyAudio Member
 
Merlinb's Avatar
 
Join Date: May 2006
Location: Lancashire
Quote:
Originally Posted by Alan0354 View Post
From looking at the triode curve, you might be able to make Fig.2.51 work as the cathode uses a zener.
Yes, fig. 2.51 (3rd ed) certainly works; it's a very ordinary application of the CCS. It would not work reliably if the triode was replaced with a pentode or transistor, which may be what is misleading you. Triodes have very low internal resistance.
  Reply With Quote
Old 2nd December 2014, 09:30 PM   #6
diyAudio Member
 
Join Date: May 2011
Location: Silicon Valley
I have the fourth edition; in it the page numbers are 143-146 and the Figure numbers are 2.54 and 2.56. The fourth edition includes this "smoking gun" on p. 144:
At first sight, the circuit has two constant current sources, both trying to define the current in the same wire. The trick is that the cathode CCS is deliberately superior to the anode constant current sources, enabling it to enforce its behaviour upon them.
  Reply With Quote
Old 2nd December 2014, 09:33 PM   #7
diyAudio Member
 
Join Date: Apr 2014
Location: US
Quote:
Originally Posted by Merlinb View Post
Yes, fig. 2.51 (3rd ed) certainly works; it's a very ordinary application of the CCS. It would not work reliably if the triode was replaced with a pentode or transistor, which may be what is misleading you. Triodes have very low internal resistance.
Yes, I am using the 3rd edition.

Yes, that's what I miss is the triode has low internal impedance for Fig. 2.51 as the plate voltage can modulate the plate current. But that still present problem in Fig. 2.49. In Fig. 2.49, the current is set by the bottom CCS, not by the triode.

But the output impedance of BJT looking into the collector is very high. The collector curve is almost flat because the early voltage is very high. Rout= d(Ic)/d(Vce). I did the calculation before with one transistor, it is way over 1M.
  Reply With Quote
Old 2nd December 2014, 09:38 PM   #8
Merlinb is offline Merlinb  United Kingdom
diyAudio Member
 
Merlinb's Avatar
 
Join Date: May 2006
Location: Lancashire
Quote:
Originally Posted by Alan0354 View Post
But the output impedance of BJT looking into the collector is very high. The collector curve is almost flat because the early voltage is very high.
The Early voltage is not that high, typically 50V to 200V. For example, when operating at a collector current of 1mA you probably wouldn't get better than 200V/1mA = 200k ohms. And that's optimistic.
  Reply With Quote
Old 2nd December 2014, 10:06 PM   #9
diyAudio Member
 
Join Date: Oct 2004
Location: Adelaide South Oz
Alan,
My experience with the Baby Huey design was that the cascode CCS gave noticable AUDIBLE improvement over the single transistor CCS when used in the "tail" of a differential amp.
As Merlin says above, any single transistor based CCS will have its impedance limited by Early Effect to not more than a few hundred KOhms at best. I recall I calculated (from circuit measurements) the effective impedance of the LED referenced single transistor CCS that I initially used and got 130 KOhms. Changing to a cascode gave superior smoothness in the top end.
Once you get the impedance of any CCS above about 100K then you need to start worrying about device (and stray) capacitance. The cascode will help with that too, AND it is the reason I don't like LM317 and similar based CCS.
Cheers,
Ian
  Reply With Quote
Old 2nd December 2014, 10:19 PM   #10
diyAudio Member
 
Join Date: Apr 2014
Location: US
Quote:
Originally Posted by Merlinb View Post
The Early voltage is not that high, typically 50V to 200V. For example, when operating at a collector current of 1mA you probably wouldn't get better than 200V/1mA = 200k ohms. And that's optimistic.
Yes, I double checked, I use 50K emitter resistor in the CCS in my design, that's the reason I got such a high output impedance. Yes, you are right, with lower emitter resistor like 1K, you only get about 200K or so.

The question still remain in Fig. 2.49, That you have CCS in both top and bottom. How can you determine the quiescent voltage of the plate. The bottom CCS control the current, not the triode. This is different from Fig. 2.51 that the triode can move the plate voltage to have plate current matching the CCS on top.

Thanks
  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


Similar Threads
Thread Thread Starter Forum Replies Last Post
Morgan Jones' CCS's Kay Pirinha Construction Tips 10 15th June 2012 07:41 AM
Morgan Jones BALANCED? vvs_75 Tubes / Valves 3 3rd April 2007 03:26 PM
Morgan Jones book to swap pingfloid Tubes / Valves 3 8th May 2004 05:19 AM
MY Morgan Jones headamp. digi01 Tubes / Valves 9 25th April 2003 02:19 AM


New To Site? Need Help?

All times are GMT. The time now is 12:09 AM.


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