Discrete Opamp Open Design - Page 49 - diyAudio
Go Back   Home > Forums > Source & Line > Analog Line Level

Analog Line Level Preamplifiers , Passive Pre-amps, Crossovers, etc.

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 August 2012, 08:19 AM   #481
forr is offline forr  France
diyAudio Member
 
forr's Avatar
 
Join Date: Dec 2004
Location: Next door
Quote:
Originally Posted by ChocoHolic View Post
Scott's posting #135 and the related circuitry is a beauty.
Unfortunately the component count is high.
And most DIY enthusiasts would struggle with a discrete built of such high gain at 20kHz. Slowing down the dominant pole most likely would be necessary to ensure DIYability, unfortunately this impacts performance....

Nevertheless, to me it is the most educative circuitry of this thread IMHO,
thanks for that !!
I missed this one, thanks for having pointed it. I agree with your comments. I like to see Fets of one type only in the differential pair.

Last edited by forr; 29th August 2012 at 08:21 AM.
  Reply With Quote
Old 29th August 2012, 09:38 AM   #482
diyAudio Member
 
Join Date: Sep 2007
Send a message via Yahoo to linuxguru
Quote:
Originally Posted by PaulBysouth View Post
Here is yet another circuit for a discrete opamp. It has the front end of the AD797 opamp, but with a single ended Class A Mosfet output stage. This circuit can be varied considerably. The input devices can be Jfets, Mosfets or BJTs.
I use essentially the same floating current-mirror concept from the AD797 in the all-BJT LF03 discrete opamp. The current mirror is bootstrapped to about one Vbe drop below the output.

LF03 discrete opamp

I do cheat a little bit on the latest revision (lf03c), by using a few 1.25V reference diodes (which are actually monolithic band-gap references) for biasing. But the topology still works ok in simulation with the bandgaps substituted by a pair of 1N4148s or similar.

How does it sound? Very dark, detailed and airy - audibly trumps the OPA627 in the I/V stage of a Marantz CD4000.
  Reply With Quote
Old 29th August 2012, 10:42 AM   #483
gpapag is offline gpapag  Greece
diyAudio Member
 
gpapag's Avatar
 
Join Date: Nov 2002
Location: Athens-Greece
Quote:
Originally Posted by Kindhornman View Post
Steven

Quote:
I had not seen a hand held with that resolution of depth of field.

Steven, I have not seen one either.

These technical capabilities were for dedicated IR imaging devices and not for hand held IR spot thermometers. (And these were for entry level prof. Equipment available since 5 years. Today, with 1-2 zeros added on the check, one can have quantum IR-sterling cooled-sensors, 1us system response times, true radiometric calibration and precision and software with amazing capabilities).

Quote:
You could also drill a hole as you stated and insert the thermocouple into that hole.
By drilling the hole you would appear to have two affects, one being the properties of the cavity depth to width and the second being a much thinner substrate that has a time lag in response to the actual temperature change internal to the device. Am I following what you are saying properly? Thanks again.
The hole works like a trap for IR (heat)radiation. Something like an insulated enclosure with a small hole on top.
The thermal capacity of the walls (material's mass) is much higher than that of the air inside the hole. Thus material temp. in effect dictates cavity’s air temperature (air standing still).
Calibrators for thermocouples -for IR spot thermometers as well- have such cavities (depth to d ratio is minimum 10)

Quote:
I am glad you pointed out the fallacy of increasing the temperature…
It is not exactly fallacy.
It is valid when dealing with DC operating points in such situations:
Static –no signal-conditions
With sub-Hz AC signal
With HF only (say RF) signals.

My point had to do with dynamic conditions at audio frequencies and with audio signals.




Quote:
Originally Posted by jneutron View Post
Actually, that is not the case.***

Interchip dynamics will not change unless the gradient to the outside world is different for each chip. If for example, one die's heat must go past the other to get out.

***the astericks are because the thermal conductivity of silicon is temperature dependent, the eq is:

Kt =286/(T -100) for T =300 to 600K

Cheers, jn
Jneutron
I can not figure out in your post why this is not the case (we are talking “interchip thermal dynamics”)

Per the formula above, Kt decreases with increasing frequency.
What do I miss?

George
__________________
["Second Law is a bitch." - SY] ["The Road To Heaven:Specify the performance & accept the design. The Road To Hell:Specify the design & accept the performance"-Bruno Putzeys]
  Reply With Quote
Old 29th August 2012, 01:12 PM   #484
diyAudio Member
 
jneutron's Avatar
 
Join Date: Jan 2004
Location: away
Quote:
Originally Posted by gpapag View Post
Jneutron
I can not figure out in your post why this is not the case (we are talking “interchip thermal dynamics”)

Per the formula above, Kt decreases with increasing frequency.
What do I miss?

George
You spoke of hotter temperatures changing the thermal response/deviation of the circuit because of the lower ability to transfer heat outside the system. This of course impacts the absolute temp of the junctions, but not the differential between chips, which is the discussion.

I did point out that silicon's thermal resistance/conductivity is a function of temperature, as is BeO's. Other than that effect, the thermal response of the system is not dependent on the sink temp, because the heat sources are not.

jn
  Reply With Quote
Old 29th August 2012, 01:31 PM   #485
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: cambridge ma
Quote:
Originally Posted by hitsware View Post
How much emitter resistance is needed
to degenerate the gain of a 2n4401 down to
an (enhancement-version) 2sk170 ?
(since the enlightened know that bi-polars are voltage controlled)
I have very limited models for discrete devices so consider the 4401/4403 a place holder. What you mention above is not really necessary, the bi-polars are close to a simple 2:1 current mirror so the left and right sides of the input transconductance contribute equally to the output current in true push pull fashion. The voltage gain is developed only at the output. This is a simple difference from the usual use I have seen where there is only a large-ish resistor on the left side and the degeneration on the Vas is much smaller yielding two voltage gains in series.

I also think the Cdg of the JFET's will be much less of a problem without large voltage gain at the drain. In addition all the junctions balance so this stage is very thermally stable.

Sorry my symbol library does not even have a PFET symbol, luckily the model assignment fixes that. Also it was late, this is intended to drive the output stage of your choice to make a complete amplifier. I was simply saying that it at 20k even open loop there was low distortion and lots of gain left.
__________________
Silence is so accurate.

Last edited by scott wurcer; 29th August 2012 at 01:38 PM.
  Reply With Quote
Old 29th August 2012, 02:12 PM   #486
bcarso is offline bcarso  United States
diyAudio Member
 
Join Date: May 2005
Location: Canoga Park, California
Quote:
Originally Posted by jneutron View Post
You spoke of hotter temperatures changing the thermal response/deviation of the circuit because of the lower ability to transfer heat outside the system. This of course impacts the absolute temp of the junctions, but not the differential between chips, which is the discussion.

I did point out that silicon's thermal resistance/conductivity is a function of temperature, as is BeO's. Other than that effect, the thermal response of the system is not dependent on the sink temp, because the heat sources are not.

jn
Indeed, it is a seductive but misleading notion per se to suppose that running warmer will reduce the importance of fluctuations. I see it stated without proof periodically, as if it were intuitively obvious, and it is not. IF the temperature coefficient itself were a strong function of absolute temperature then you would see an effect, whether beneficial of not clearly depends on the sign.

Another notion is the concern with dissipation that changes a chip temperature from, say, 25 C to 75 C, and then people start worrying about the increase in thermal noise. Since (for true thermal noise sources) the voltage or current noise is proportional to the square root of the absolute temperature, a change of that magnitude is only an increase in noise spectral density of about eight percent. Not welcome, but not horrible.

However, for other processes generating noise, like the shot noise in drain-gate leakage, the temperature dependence is strong, roughly doubling the leakage current every delta 10-12 K, and with that the midband current noise proportional to the square root of the current. When JFET opamps began to appear it was dicey to evaluate the actual bias currents at the inputs if you weren't sure how warm you'd be operating. Some manufacturers pointed out that some bipolar amps with teeny input stage currents and super-beta devices could beat the JFET amps at higher temperatures!

At the operating temperature of the photodiode array preamps (one per video line), near dry ice, the net input currents were about 1 and 3 fA. The devices were 2N4416s, run at about 6V Vdg, with Vgs about -100mV.
  Reply With Quote
Old 29th August 2012, 02:55 PM   #487
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: cambridge ma
Quote:
Originally Posted by scott wurcer View Post
I also think the Cdg of the JFET's will be much less of a problem without large voltage gain at the drain. In addition all the junctions balance so this stage is very thermally stable.
In fact the voltage gain from input to drain is ~-1 just as in a cascode with like devices but here the current mirror configured as it is makes this voltage common mode at the drains so with balanced source impedance we will see little effect from Cdg. In essence there is not much gained by cascoding the input FET's as long as the DC gate current is not too compromised by Vdg. In lower gain voltage follower applications the common mode effects are more important but ballancing the source impedances for AC cmrr is still important.
__________________
Silence is so accurate.

Last edited by scott wurcer; 29th August 2012 at 03:06 PM.
  Reply With Quote
Old 29th August 2012, 02:56 PM   #488
diyAudio Member
 
Join Date: Mar 2002
Quote:
Originally Posted by scott wurcer View Post
I have very limited models for discrete devices so consider the 4401/4403 a place holder. What you mention above is not really necessary, the bi-polars are close to a simple 2:1 current mirror so the left and right sides of the input transconductance contribute equally to the output current in true push pull fashion. The voltage gain is developed only at the output. This is a simple difference from the usual use I have seen where there is only a large-ish resistor on the left side and the degeneration on the Vas is much smaller yielding two voltage gains in series.

I also think the Cdg of the JFET's will be much less of a problem without large voltage gain at the drain. In addition all the junctions balance so this stage is very thermally stable.

Sorry my symbol library does not even have a PFET symbol, luckily the model assignment fixes that. Also it was late, this is intended to drive the output stage of your choice to make a complete amplifier. I was simply saying that it at 20k even open loop there was low distortion and lots of gain left.
Scott,
Thank You for the answer, but you misunderstood my post
It had nothing to do with your circuit.
I would simply like to know how much emitter degeneration
it would take to make a 2n4401 have ~ the same gain as a
2sk170 ? I.E. Use degenerated bipolar complements to replace
the disappearing jfet complements.
  Reply With Quote
Old 29th August 2012, 03:04 PM   #489
diyAudio Member
 
Join Date: Mar 2002
Quote:
Originally Posted by abraxalito View Post
But you have been educated on this subject - you yourself said that 'the enlightened know transistors are voltage controlled' That Hfe isn't well defined is a corollary of this I believe.
I was being facetious. (I thought)
What little education I have
(USN Electronics Technician and Occupational electronics
(~ 30 years ago)) both used hfe as the main parameter
with bipolars. AND that xsistors are current controlled as
opposed to tubes being voltage controlled.
(I'm not arguing the point, only stating what I was taught)

hfe = delta Ic / delta Ib (IIRC)
re ~ 22 Ohm (small signal device)
input Z = re + (emitter resistor * hfe)

Last edited by hitsware; 29th August 2012 at 03:10 PM.
  Reply With Quote
Old 29th August 2012, 03:21 PM   #490
ief is offline ief  Netherlands
diyAudio Member
 
ief's Avatar
 
Join Date: Jun 2011
-noob allert-

Today got a new pre-amp and that always makes me highly interested in what exactly I have but knowledge is almost non existant I'm afraid. Still I hope it would be ok for me to post this here since this thread seems a good way to see if I 'get' some of the stuff discussed. (perhaps some credit for going through the whole thread? )

If this is not wanted and/or to OT just ignore please.

Anyway, attached a sniped of the pre-amp I acquired today (chuffed ) and some questions.

a. The part in the red square is the discrete opamp (?)
b. The circeled parts are the DC coupling caps you all are wanting to avoid (?) (and thus this design would not fall in the wanted design category)
c. This must mean some sort of DC servo feedback stuff talked about a couple of pages back is non existant and the follow up question would be...
d. Why would the designer chooses for the DC coupling caps (quit a few in the whole design) is cost the only reason or are there more benefits of this design?
e. anything else you like to share, love to learn more!
Attached Images
File Type: jpg discr-opa.jpg (226.5 KB, 330 views)

Last edited by ief; 29th August 2012 at 03:24 PM.
  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
Discrete OPAMP audio-gd Vendor's Bazaar 27 20th September 2012 04:02 PM
discrete opamp help blackpowderaudio Parts 0 16th December 2009 03:46 PM
THAT transistor headphone amp (250ma discrete opamp) design sanity check. Russ White Headphone Systems 19 13th December 2007 12:52 PM


New To Site? Need Help?

All times are GMT. The time now is 01:17 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