Critics needed

kimschips · 22nd March 2005, 05:29 PM

Hey

Here is my 25Cents for a circuit with folded Back cascode coupled transistors.

All transistors are 2SC2240(npn) and 2SA970(pnp)
except Q17-18 are 2SC4159 and 2SA1606.

The supply (+-10V comming from a LCAudio shuntreg)

The circuit is intended for a Highend preamplifier.

The gain is 2times (6dB)

Please make some comments about my circuits and your opinien to make it to the top.

Thanks

Best regards

darkfenriz · 22nd March 2005, 07:25 PM

first impresions:
1. I don't know how lm339 behaves, but check if its output reaches close to rail voltage (10...0...-10 I guess)

2. I haven't calculated the poles, but am a little suspicious about C1 and stability issues- do you really need it?

3. It is global feedback free, so check if distortion is small enough

4. It's demon fast- think how to avoid RF interference, consider input C to the ground or paralel to R8, just several picoFarads

5. If above are pointless, then... GOOD WORK! Listen ihow it sounds and enjoy.

cheers

kimschips · 22nd March 2005, 08:22 PM

Thanks

Does someone know how to connect two DC servos, if i want to make it balanced?
Imagine I also take out off the left Collektor in both differentiale amplifiers, so i have a + and a - output.

1. its working fine as it is..

2. C1 prevents High freqency issues in the circuits.
Maybe it commes from Q15-16 when it´s try to hold down Q17-18? noted it with scope.

3. The distortion is 0.08% with 10Vpp with RL =100R and clipping with 15Vpp.

4. Good point there-> I think a resistor in serie with a capacitor about 56pF in parallel with R9 (and R10) should do the job, so you prevent RF in the circuits?

Best regards

anatech · 23rd March 2005, 01:54 AM

Hi kimschips,
The nifty thing about a diamond buffer is ... if you thermally couple Q7,12,17 and 18 the bias will track and remain extremely stable. You can delete Q15 & Q16. Your low distortion is expected from this buffer.
-Chris

kimschips · 23rd March 2005, 06:57 AM

Hey Anatech

My idea with Q15 and 16 was to make a CCS with Q17-18, then I could use 2SC2240 2SA970 (for Q15-16) and did not have to couple Q15-18 together with Q17-18.
Q15-16 do also make a shortcircuit protectection.
Butt it´s not needed now with diamond buffer??

Would it be better to lower R20,21 to 100R and R22-23 to 5ohm?

I forgot to say that i measured the distortion fromm 100Hz to 20Khz (0.08%) with this:
PC
Soundcard Soundblaster Audigy 2
Software : Oscillometer 5.07
Low distortion generator.

How about the resistors R3-R8? does anyone have a better solution to values?

Best regards

Eva · 23rd March 2005, 11:56 AM

The LM339 IC isn't an op-amp, it's a comparator thus it's designed with response times in mind and ignoring stability issues. There are no guarantees about comparator stability but be sure that it won't be unity-gain stable, it will oscillate like crazy when negative feedback is applied

peranders · 23rd March 2005, 12:04 PM

The LM339 isn't unity gain stable unless you make a very slow output stage but remember also that the servo is injecting a small amount of audio signals back into the amp, therefore the servo have to have good audio properties to 100-1000 Hz depending where you place your target.

You must also take into account the supply voltage of the servo and desired output voltage.

Why don't you let the servo go to the inverting input like the rest of us would do?

kimschips · 23rd March 2005, 04:37 PM

Hey.

(Peranders) I won´t send the servo back to the inventing input, because the circuit are also intended to be used balanced and in my cd player, which gives a balanced voltage out from the dac.

About slewrate it´s not so fast as Peranders, butt I measure it tonight again.

Anyone having ideas to make it faster and to lower the distortion?

How good is a capacitor in 100uF to decouple the supply for freqencies higher than 100Khz?

I´m asking because the supply is comming from a Shunt reg which impedance and Ft will be about 0.01R from DC to 400Khz.
And I try to make the circuit so good (Powersupply) so the sound will be free as possible from "capacitorsound"

Thanks for all
Kim

sam9 · 23rd March 2005, 04:58 PM

My limited experience with a comparator, LM339, says that the output will be binary, i.e. a square wave with the same period as the input signa. In this case "low" is ground/0V and "high" will be just a smodgen below the rail voltage (or whatever feeds Vcc). If the feedback cap is big enough (1,000uF ?) the square wave will become a sine again but with a DC offset.

Is this what's intended here?

sam9 · 23rd March 2005, 05:17 PM

I have a broader question/concern about DC servos:

Some musical signals may naturally have brief (a few cycles of) DC offset in them even if the average offset over a longer period of time averages out to zero offset. In such a case, doesn't a DC servo introduce distortion (in the broad sense of: Output .NE. Input*A, not just harmonics)? Following that logic, is a simple adjustment that achieves quiecent zero offset not only all that is needed but all that is desired?

These are not Socratic questions -- I'm really puzzeled on this metter.

22nd March 2005, 08:22 PM	#3
kimschips diyAudio Member Join Date: Dec 2004 Location: Denmark	Balanced out? Thanks Does someone know how to connect two DC servos, if i want to make it balanced? Imagine I also take out off the left Collektor in both differentiale amplifiers, so i have a + and a - output. 1. its working fine as it is.. 2. C1 prevents High freqency issues in the circuits. Maybe it commes from Q15-16 when it´s try to hold down Q17-18? noted it with scope. 3. The distortion is 0.08% with 10Vpp with RL =100R and clipping with 15Vpp. 4. Good point there-> I think a resistor in serie with a capacitor about 56pF in parallel with R9 (and R10) should do the job, so you prevent RF in the circuits? Best regards

23rd March 2005, 12:04 PM	#7
peranders Electrons are yellow and more is better! diyAudio Member Join Date: Apr 2002 Location: Göteborg, Sweden Blog Entries: 4	The LM339 isn't unity gain stable unless you make a very slow output stage but remember also that the servo is injecting a small amount of audio signals back into the amp, therefore the servo have to have good audio properties to 100-1000 Hz depending where you place your target. You must also take into account the supply voltage of the servo and desired output voltage. Why don't you let the servo go to the inverting input like the rest of us would do? __________________ /Per-Anders (my first name) or P-A as my friends call me Tube Buffered Gainclone in work \|Thread

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
Help needed ! Inter M R300 schematics needed !	alexclaire	Solid State	30	9th September 2009 06:14 AM
Help Needed	d3cibel	Car Audio	0	30th April 2008 06:52 AM
LT1964-adj Help Needed or Part Needed	lgreen	Parts	2	29th June 2007 02:50 PM
a couple amps. critics / suggestions needed	millwood	Solid State	2	24th May 2003 01:46 AM

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 March 2005, 07:25 PM	#2
darkfenriz diyAudio Member Join Date: Jun 2004 Location: Warsaw	first impresions: 1. I don't know how lm339 behaves, but check if its output reaches close to rail voltage (10...0...-10 I guess) 2. I haven't calculated the poles, but am a little suspicious about C1 and stability issues- do you really need it? 3. It is global feedback free, so check if distortion is small enough 4. It's demon fast- think how to avoid RF interference, consider input C to the ground or paralel to R8, just several picoFarads 5. If above are pointless, then... GOOD WORK! Listen ihow it sounds and enjoy. cheers

23rd March 2005, 01:54 AM	#4
anatech diyAudio Moderator Join Date: Jun 2004 Location: Georgetown, On	Hi kimschips, The nifty thing about a diamond buffer is ... if you thermally couple Q7,12,17 and 18 the bias will track and remain extremely stable. You can delete Q15 & Q16. Your low distortion is expected from this buffer. -Chris

23rd March 2005, 06:57 AM	#5
kimschips diyAudio Member Join Date: Dec 2004 Location: Denmark	Hey Anatech My idea with Q15 and 16 was to make a CCS with Q17-18, then I could use 2SC2240 2SA970 (for Q15-16) and did not have to couple Q15-18 together with Q17-18. Q15-16 do also make a shortcircuit protectection. Butt it´s not needed now with diamond buffer?? Would it be better to lower R20,21 to 100R and R22-23 to 5ohm? I forgot to say that i measured the distortion fromm 100Hz to 20Khz (0.08%) with this: PC Soundcard Soundblaster Audigy 2 Software : Oscillometer 5.07 Low distortion generator. How about the resistors R3-R8? does anyone have a better solution to values? Best regards

23rd March 2005, 11:56 AM	#6
Eva diyAudio Member Join Date: Oct 2003 Location: Near the sea	The LM339 IC isn't an op-amp, it's a comparator thus it's designed with response times in mind and ignoring stability issues. There are no guarantees about comparator stability but be sure that it won't be unity-gain stable, it will oscillate like crazy when negative feedback is applied

23rd March 2005, 04:37 PM	#8
kimschips diyAudio Member Join Date: Dec 2004 Location: Denmark	Hey. (Peranders) I won´t send the servo back to the inventing input, because the circuit are also intended to be used balanced and in my cd player, which gives a balanced voltage out from the dac. About slewrate it´s not so fast as Peranders, butt I measure it tonight again. Anyone having ideas to make it faster and to lower the distortion? How good is a capacitor in 100uF to decouple the supply for freqencies higher than 100Khz? I´m asking because the supply is comming from a Shunt reg which impedance and Ft will be about 0.01R from DC to 400Khz. And I try to make the circuit so good (Powersupply) so the sound will be free as possible from "capacitorsound" Thanks for all Kim

23rd March 2005, 04:58 PM	#9
sam9 diyAudio Member Join Date: Jun 2002 Location: Left Coast	My limited experience with a comparator, LM339, says that the output will be binary, i.e. a square wave with the same period as the input signa. In this case "low" is ground/0V and "high" will be just a smodgen below the rail voltage (or whatever feeds Vcc). If the feedback cap is big enough (1,000uF ?) the square wave will become a sine again but with a DC offset. Is this what's intended here?

23rd March 2005, 05:17 PM	#10
sam9 diyAudio Member Join Date: Jun 2002 Location: Left Coast	I have a broader question/concern about DC servos: Some musical signals may naturally have brief (a few cycles of) DC offset in them even if the average offset over a longer period of time averages out to zero offset. In such a case, doesn't a DC servo introduce distortion (in the broad sense of: Output .NE. Input*A, not just harmonics)? Following that logic, is a simple adjustment that achieves quiecent zero offset not only all that is needed but all that is desired? These are not Socratic questions -- I'm really puzzeled on this metter.