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 13th May 2013, 01:13 AM   #11
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
OK, the line amp is flawed. In short, the input is connected right to a gain summing node at the gate of Q7A and Q7B. It's a fairly crude summing node/virtual ground, as the gain around Q7 is low (compared to an op amp), even with the current source loading presented by Q6A/B. In short, the open loop gain/bandwidth of the preamp will depend on the source impedance driving into the Q7 gate. I got an indication of this early on, as the first gain/phase plots I made of the line amp were with the input pots adjusted short of full. With the high driving impedance represented by the input pot at less than full tilt, the gain-phase plots showed none of the high frequency peaking evidenced by the plots shown here. Subsequent measurements were done with the pots turned up to 100%.

The gain-phase analyzer has an intermediate source impedance (600 ohms?), and causes problems at high frequency. Driving the inputs with the 50 ohm impedance square wave output of a function generator gave me 50 ns edges on the output waveform, with no evidence of peaking or other antisocial behavior.

Clearly, you can't have a line amp that that is so heavily dependent on the driving impedance. The circuit attached here isolates the summing nude and fixes the problem (in simulation), at the cost of some more parts. I'll be checking out the real deal as soon as I have time to rig the extra components on my current revision board.
Attached Images
File Type: gif Lineamp_gain.gif (18.4 KB, 533 views)
  Reply With Quote
Old 15th May 2013, 11:42 PM   #12
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
Other folks (SY in particular) have pointed out that with the output of most signal sources, a line amp with gain is usually not necessary. Possible exceptions might designs such as the SIT and "Fake SIT" SE amp designs posted elsewhere on DIYA, and some of the less sensitive Pass amps. I'm trying my hand at several line amp w/gain designs, as it's a nice option to have in one's back pocket, and getting a good clean gain of 5-10X without resorting to something that looks rather suspiciously like an op amp is a challenge.

Having said that, I may settle down to a unity gain line amp, and the circuit shown here is a possibility. Some compensation was needed, as the circuit oscillated like an SOB without it in initial simulation. Distortion as simulated is hilariously low (0.00007%). When it gets down that low, I tend not to quite trust either the simulated THD or harmonic distribution. At any rate, I used some current mirrors to cut down the number of expensive depletion mode mosfets to just one (which would go in place of the current source I used in the simulation). There are jfets that would also work, though with lower output impedance. Using the current mirror also allowed me to remove a level shifting LED from the source of the P-channel mosfet.

At any rate, food for thought for now, as I want to get the stuff I have working before I try something else.
Attached Images
File Type: gif Line_UGain.GIF (17.2 KB, 503 views)
  Reply With Quote
Old 16th May 2013, 09:32 AM   #13
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
A hypothetical amplifier with a gain of 28times producing a maximum output signal of 28Vac (100W into 8ohms and/or 200W into 4ohms) needs 1Vac as Maximum input signal.
An F5 with a gain of 6times produces a maximum output signal of 14.1Vac and needs 2.4Vac as a Maximum input signal.

Both these amplifiers will work well with very many sources if a gain block with switchable gains of +0dB (buffer), or +6dB (2times pre-amp), or +12dB (4times preamp).

The F5 & +12dB requires a Maximum signal of 600mVac

Are there many/any Sources that an F5 would need 10times gain (+20dB) giving a sensitivity of 240mVac to get up to clipping level?
__________________
regards Andrew T.
  Reply With Quote
Old 16th May 2013, 11:08 AM   #14
Bonsai is offline Bonsai  Taiwan
diyAudio Member
 
Bonsai's Avatar
 
Join Date: Jul 2003
Location: Taipei, Taiwan
Conventional wisdom has it that non digital source inputs output 150mV and power amps need 1V to drive them to full power.

So, attenuate you CD by 20 dB before it goes into the line stage
Set you line stage gain to 16 dB

and there you have it - the standard hi-fi signal chain gain structure
__________________
bonsai
Amplifier Design and Construction for MUSIC! http://hifisonix.com/
  Reply With Quote
Old 16th May 2013, 11:16 AM   #15
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Quote:
Originally Posted by Bonsai View Post
Conventional wisdom has it that non digital source inputs output 150mV and power amps need 1V to drive them to full power.
..................
Are there any or many that are that low?

If there are some and one wants to cater for them, then add a gain stage for that Source ONLY.

Do not attenuate all the other Sources down to this lowest level and then boost them back up to the sensitivity level required by the next stage of amplification.
__________________
regards Andrew T.
  Reply With Quote
Old 16th May 2013, 03:33 PM   #16
Bonsai is offline Bonsai  Taiwan
diyAudio Member
 
Bonsai's Avatar
 
Join Date: Jul 2003
Location: Taipei, Taiwan
This is quite the normal way of doing things Andrew. And I works well noise wise.

I have looked at switching the gain on the line stage on the low level inputs, but, for 2V digital source signals, you still have to attenuate by 6 dB to get to the 1 V for the amp input sensitivity. If you are aiming for a nominal 10k input impedance, you are left with 2.5k source, vs attenuation at c. 900 ohms. The other option is to buffer the digital source and then divide down with low value resistors. Works well, but there's added complexity and it is thus inelegant.
__________________
bonsai
Amplifier Design and Construction for MUSIC! http://hifisonix.com/

Last edited by Bonsai; 16th May 2013 at 03:44 PM.
  Reply With Quote
Old 16th May 2013, 07:49 PM   #17
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
You guys can argue as you like, it's not going to affect what I do. This is a description of a minimal setup with std PCB size to test gain blocks in my system as I devise them, not design-to-order.
So, no switchable gain, no MC input. I have absolutely no incentive to do either.
  Reply With Quote
Old 20th May 2013, 09:06 PM   #18
scholl is offline scholl  United States
diyAudio Member
 
Join Date: Dec 2012
Hi,

I'm working along the same lines as you are.

How would your design compare to a simpler circuit around the LND150? I'm thinking plain common source direct coupled to a source follower with LND150 CCS in the source circuit. All biased 8-12ma. That would be 3 LND150s and 4 resistors plus output cap with high value resistor to ground.

If a non inverting design is required a grounded grid stage direct coupled to the source follower...
  Reply With Quote
Old 20th May 2013, 10:12 PM   #19
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
I would use a different part than the LND for a current source. The on-resistance on those parts is pretty high, and the current capability rather low for my tastes. I generally use the DN2530N3 or N5 unless I want >100mA, in which case I use an IXYS part.
A simple current-source loaded follower will work OK, though the transconductance will be lower and the distortion somewhat higher. The purpose of the extra parts in my circuit is to get higher transconductance/lower output impedance by means of a short and direct feedback loop. One could also run a mosfet or a darlington at a high bias current to get higher transconductance/lower output impedance, at the cost of more power dissipation.
  Reply With Quote
Old 21st May 2013, 04:10 PM   #20
scholl is offline scholl  United States
diyAudio Member
 
Join Date: Dec 2012
Good points!!

The LND has a nice curve and maybe best as common source voltage amp with different devices for current amps.

Thanks,

Scott
  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
Is this preamp circuit OK? Caleb Chia Solid State 21 16th April 2013 06:04 PM
Which opamp in this preamp circuit? lordearl Analog Line Level 5 23rd January 2012 07:14 AM
Need help with this preamp circuit b0nk3rs Solid State 11 17th January 2007 08:55 AM
Mic Preamp from X preamp circuit? e.lectronick Pass Labs 2 13th November 2002 02:01 PM
Preamp Circuit cm961 Everything Else 9 15th June 2001 02:12 AM


New To Site? Need Help?

All times are GMT. The time now is 07:34 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