Hawksford based conceptual amplifier

Jorge · 21st August 2005, 11:36 PM

Hello,

This is high feedback conceptual amp, incorporating two of Hawksford ideas - the EC circuit and the balanced EF for the VAS.
Feedback at 20kHz is 66dB, margin is 60 deg.

The LT1115 opamp was used only due to it's specs (FET input, 40MHZ BW). No other reason.

Circuit:

Jorge · 21st August 2005, 11:37 PM

Open loop distortion of the discrete stage (8V PP @ 4 ohms):

Jorge · 21st August 2005, 11:39 PM

Closed loop distortion (as above):

Jorge · 21st August 2005, 11:40 PM

PSR (positive rail):

lumanauw · 22nd August 2005, 12:48 PM

Hi, Jorge,

I dont understand the mathematics of Hawksford EC.
But in your schematic, why dont you just connect output of opamp to the emitors junction of MPSA06/56 (Q12)? I think this way it will provide error correction too, with simpler cct.
And change the above/below transistors (Q7,8,11,14) to CCS.

darkfenriz · 22nd August 2005, 01:23 PM

Hi
I suspect your PSRR results are so good mainly because very high feedback of an op-amp. The fact is such topology is asking for separate regulated power supply for everything but output stage. Voltage dividers from rail voltage and local feedbacks from collector to base of a darlington just won't play good. Keep in mind that simulated caps (like C3 and C13 or C14/16) have zero ESR and ESL, which won't be that nice in reality.
I also think you don't need C7/9, CFPs won't like capcitive loads, better to manipulate with R21/27/30/31/23 and get rid of these caps.
Hawksford EC is just fine, guess R22 should be a pot to adjust bias.
The thing, that makes me wonder are your 'YZ' connections. Probably you intended smart and efficient driving of VAS transistors. But what happens at high frequency, when input capacitances become significant, collector currents of Q8/11 are 90deg. phase shifted with voltage and so on..??? What is your reasonig here?

best regards

Jorge · 22nd August 2005, 03:03 PM

Hello, lumanauw

I will look at your suggestions.

Hello, darkfenriz

Yes, PSR is heavily NFB based, but caps have series resistences - 0.028 for the 3300u, 0.041 for the 1000u and 0.07 for the 470u (LTSpice doesn't shows it in the scheamtic).

C7/C9 balances the Hawksford loop.
Look at the compnsation loop of the VAS - C5, R28, R35 (C6, R29, R36). This network defines the VAS output impedance, and it's not constant through the audio band.
To balance the EC, one needs a similar impedance feeding the EC transistors.
And R30/R31 are high enough in value to keep the CFP stable.

The Y/Z connections are suggested by Hawksford in the Super Cascode paper.
He states it's a minor enhancement, but I thought it would be worth the cost...

Steven · 23rd August 2005, 06:53 PM

Jorge,

Congratulations. You're fast! Within a day after finishing the Hawksford output in the other thread you already have the amp completed. Now the practical part...
The Hawksford balanced EF is nice too; I had forgotten that one.

Steven

aparatusonitus · 23rd August 2005, 10:02 PM

Hi,

I have been busy too...Just a mixbag sch from Pavel Dudek, Steven, Cordell, Mark, Borbely...(you name it). And it works (sort off), to my surprice...Not bad for a newbee

One Q: How to lose a global NFB loop and in the same time keep the same gain and cascode over input diff. pairs?

darkfenriz · 23rd August 2005, 10:14 PM

Hi aparatusonitus
honestly I don't dee how bias is adjusted here, also unlike Dudek's one I don't see how the correction 'corrects' in your amp. Could you give some details about how it is to work?
regards

23rd August 2005, 06:53 PM	#8
Steven diyAudio Member Join Date: Oct 2003 Location: The Netherlands	Jorge, Congratulations. You're fast! Within a day after finishing the Hawksford output in the other thread you already have the amp completed. Now the practical part... The Hawksford balanced EF is nice too; I had forgotten that one. Steven __________________ The Analog Art shows no sign of yielding to the Dodo's fate. The emergence and maturation of monolithic processing finesse has perhaps lagged a bit behind the growth of the Binary Business. But whereas digital precision is forever bounded by bits, there is no limit excepting Universal Hiss to the ultimate accuracy and functional variety of simple analog circuits. - Barry Gilbert, 1973

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Amplifier based on 2N3055	macgyver	Solid State	34	2nd June 2010 10:36 AM
LM3886 based amplifier	russo	Chip Amps	18	4th February 2008 11:27 AM
TL/TQWT conceptual question	aceinc	Subwoofers	7	26th May 2007 02:19 PM
Dr. Amp Tripath based amplifier	pepolman	Class D	8	6th August 2005 06:28 AM
CCS Conceptual Qs, Using Gary Pimm CCS as Example	DrDeville	Tubes / Valves	20	12th May 2004 02:41 PM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	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

22nd August 2005, 12:48 PM	#5
lumanauw diyAudio Moderator Emeritus Join Date: Oct 2002 Location: Bandung	Hi, Jorge, I dont understand the mathematics of Hawksford EC. But in your schematic, why dont you just connect output of opamp to the emitors junction of MPSA06/56 (Q12)? I think this way it will provide error correction too, with simpler cct. And change the above/below transistors (Q7,8,11,14) to CCS.

22nd August 2005, 01:23 PM	#6
darkfenriz diyAudio Member Join Date: Jun 2004 Location: Warsaw	Hi I suspect your PSRR results are so good mainly because very high feedback of an op-amp. The fact is such topology is asking for separate regulated power supply for everything but output stage. Voltage dividers from rail voltage and local feedbacks from collector to base of a darlington just won't play good. Keep in mind that simulated caps (like C3 and C13 or C14/16) have zero ESR and ESL, which won't be that nice in reality. I also think you don't need C7/9, CFPs won't like capcitive loads, better to manipulate with R21/27/30/31/23 and get rid of these caps. Hawksford EC is just fine, guess R22 should be a pot to adjust bias. The thing, that makes me wonder are your 'YZ' connections. Probably you intended smart and efficient driving of VAS transistors. But what happens at high frequency, when input capacitances become significant, collector currents of Q8/11 are 90deg. phase shifted with voltage and so on..??? What is your reasonig here? best regards

22nd August 2005, 03:03 PM	#7
Jorge diyAudio Member Join Date: Apr 2001 Location: Brazil	Hello, lumanauw I will look at your suggestions. Hello, darkfenriz Yes, PSR is heavily NFB based, but caps have series resistences - 0.028 for the 3300u, 0.041 for the 1000u and 0.07 for the 470u (LTSpice doesn't shows it in the scheamtic). C7/C9 balances the Hawksford loop. Look at the compnsation loop of the VAS - C5, R28, R35 (C6, R29, R36). This network defines the VAS output impedance, and it's not constant through the audio band. To balance the EC, one needs a similar impedance feeding the EC transistors. And R30/R31 are high enough in value to keep the CFP stable. The Y/Z connections are suggested by Hawksford in the Super Cascode paper. He states it's a minor enhancement, but I thought it would be worth the cost...

23rd August 2005, 10:14 PM	#10
darkfenriz diyAudio Member Join Date: Jun 2004 Location: Warsaw	Hi aparatusonitus honestly I don't dee how bias is adjusted here, also unlike Dudek's one I don't see how the correction 'corrects' in your amp. Could you give some details about how it is to work? regards