Second attempt at Building my own amp from scratch - Page 8

Vostro · 7th December 2012, 12:48 PM

Do you think C4 and C9 (post69) will help, I've seen it on a NAD C372,
I'll make come changes next week,
Diode clamps look nice, another thing to try

thanks

Vostro · 7th December 2012, 06:10 PM

Maybe this is a second option, if I don't succeed.
What you guys think ?

AndrewT · 8th December 2012, 11:20 AM

Quote:

Originally Posted by djk

................66 is the preferred connection for the cascode voltage reference to the emitter resistor rather than to the rail.................

i.e. the cascodes exclude R4 & R20.

What difference does this make?

djk · 06:53 PM

IIRC it's similar in idea to the Lender circuit, a kind of local feedback if you will.

Its been a while since I read the Hawksford paper on this subject, my copy is at work (I think).

The APT is from about 1977, the Hawksford paper is from about 1988.

https://docs.google.com/viewer?a=v&q...euspUYKfxBvEuA

Read section 3.2 and see fig.6B

AndrewT · 9th December 2012, 09:42 AM

Thanks.

Vostro · 10th December 2012, 06:27 PM

Finally.
Looks like my biggest problem, was initially being scared and running the amp from my bench 30V-0-30V supply, instead on 63V-0-63V it was designed for.

Most changes have been made, and I'm nearly ready to make the final boards.
This circuit posted is fully working,

Should I use a BC560 for Q20, instead of 2N5401.

I also think I might get rid of R28.

I've added D8 to help clipping behaviour.

With using matched pairs in ips and mirror, offset is very low.
But thinking of including a trim pot for offset.

Any comments would be appreciated on changes.

Regards

AndrewT · 07:44 AM

Q19 & Q20 are current limiting devices. Neither should ever see high voltages. Low voltage devices are suitable.

How does D8 work?
Don't insert a trim pot in the emtter circuit of the LTP.
Better to look at VAS current trimming and/or adding a mirror trimming with a resistor from the bases to -ve rail.

Why are R41 & R42 different value?

What compression did you set for your jpeg?
I find the values/details quite difficult to read.

Vostro · 11th December 2012, 10:50 AM

Quote:

Originally Posted by AndrewT

How does D8 work?

Without the diode the 'darlington VAS' struggles to come out of saturation.
I have not seen this problem with a single transistor VAS, but when over driven really hard, I guess it could have a negative effect.
More experienced members would explain this better.
From my point of view, when the VAS transistors is driven with a large current from the IPS LTP then
the collector voltage of the vas transistor (Q3) drops quite low compared to its base voltage,
making the b/c junction forward bias, with diode added the current has a new path to flow, allowing Q3 to come out of saturation easier.
I'm not sure if I'm explaining well, I'm sure others will add to this.

Quote:

Originally Posted by AndrewT

Why are R41 & R42 different value?

Value of R41 needed to be low when cascode was for zener diode, the zener needed more current to pass zener
knee. I increased R37 for about 2mA current, thinking id get get close to 0.7V per diode.
Value of R42 was originally the same as R41, But R4 feeds current to Q12 and Diodes D6/7,
that means that the value of R4 gets affected by the value of R42 (D6/7 current).
I decided to decrease current in D6/D7 to about 1mA which is still sufficient (greater than 20x Ib Q24)
As a side effect, max output before clipping was 185W with zener cascode, the circuit shown now has a
max output of 195W before clipping, but I still consider it 185W amplifier.

Quote:

Originally Posted by AndrewT

What compression did you set for your jpeg?
I find the values/details quite difficult to read.

Circuit was exported as 1:1 screen res but at 1bit black/white bitmap.
Attached is a better resolution.

Regards.

AndrewT · 11th December 2012, 11:41 AM

Same question

Quote:

Why are R41 & R42 different value?

Far too big.
What is needed is more clarity, not more pixels.

Vostro · 11th December 2012, 12:19 PM

Can't export as bitmap with low res and still be very visible.
On proteus default font is very 'thin' and small.
In other words, theres high jpg compression, saved as jpg from bitmap in photoshop at highest quality '12' jpg setting.
Even if I posted .bmp straight from proteus it would look the same.

Only other way is to export is pdf, I prefer jpg to export to web.

Regards

8th December 2012, 06:43 PM	#74
djk diyAudio Member Join Date: Feb 2001 Location: USA	IIRC it's similar in idea to the Lender circuit, a kind of local feedback if you will. Its been a while since I read the Hawksford paper on this subject, my copy is at work (I think). The APT is from about 1977, the Hawksford paper is from about 1988. https://docs.google.com/viewer?a=v&q...euspUYKfxBvEuA Read section 3.2 and see fig.6B __________________ Candidates for the Darwin Award should not read this author. Last edited by djk; 8th December 2012 at 06:53 PM.

11th December 2012, 07:35 AM	#77
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	Q19 & Q20 are current limiting devices. Neither should ever see high voltages. Low voltage devices are suitable. How does D8 work? Don't insert a trim pot in the emtter circuit of the LTP. Better to look at VAS current trimming and/or adding a mirror trimming with a resistor from the bases to -ve rail. Why are R41 & R42 different value? What compression did you set for your jpeg? I find the values/details quite difficult to read. Last edited by AndrewT; 11th December 2012 at 07:44 AM.

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
Building an amp from scratch - what features would i want	speakerfritz	Tubes / Valves	35	26th January 2011 03:43 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

7th December 2012, 12:48 PM	#71
Vostro diyAudio Member Join Date: May 2012 Location: Other side of the fence	Do you think C4 and C9 (post69) will help, I've seen it on a NAD C372, I'll make come changes next week, Diode clamps look nice, another thing to try thanks

9th December 2012, 09:42 AM	#75
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	Thanks.

11th December 2012, 12:19 PM	#80
Vostro diyAudio Member Join Date: May 2012 Location: Other side of the fence	Can't export as bitmap with low res and still be very visible. On proteus default font is very 'thin' and small. In other words, theres high jpg compression, saved as jpg from bitmap in photoshop at highest quality '12' jpg setting. Even if I posted .bmp straight from proteus it would look the same. Only other way is to export is pdf, I prefer jpg to export to web. Regards