Oscillograms/Testing, and why layout matters.

Mooly · 14th July 2008, 03:59 PM

I was going to add some of these shots to another thread after a D.I.Y. member contacted me regarding their latest build, and he commented that all the amps he had constructed ( including my design on this Forum ) sounded totally different ( by the sound of it very very different ). Having seen the layout and pictures I couldn't help thinking that layout was as much to do with it as anything.
As it may help others I've decided to put it in it's own thread.
First shot then is the amp--- and how not to do it --- but as this is similar to many "veroboard type " constructions I think it valid.
The design is based on my own MOSFET amp but this is different, I've been playing around with some HEXFET's and this is the result so far.

Mooly · 14th July 2008, 04:01 PM

Here is the output (Top trace) and input (Bottom Trace) at 20 Khz with no load.

Mooly · 14th July 2008, 04:02 PM

And the same, correctly grounded and supplying a load of 2.7 Ohms. This is a very severe test even though the amplitude is low at 4 volts peak to peak.

Mooly · 14th July 2008, 04:05 PM

This time with the output returned to the wrong point so that the input ground is "contaminated" by the output current.
This is the only thing to have been changed -- compare the traces to the one above. Note the amplitude of the output now !

Mooly · 14th July 2008, 04:07 PM

A 0.1 Mfd cap directly across the output, but correctly returned.

Mooly · 14th July 2008, 04:08 PM

And incorrectly returned.

Mooly · 14th July 2008, 04:11 PM

A more severe test. The little PSU I am using cannot supply much current for sustained testing, but this is a good trick. Here is 12 volts Peak/Peak across the 2.7 Ohm load but with a lower duty cycle. The load was correctly terminated.

Mooly · 14th July 2008, 04:13 PM

Last one, a shot showing a detail of using an incorrect ground. This was as above but with a 0.1 Mfd again as the load.

AndrewT · 14th July 2008, 04:16 PM

why does the input signal change with the loading?

Mooly · 14th July 2008, 04:19 PM

This was all a bit of fun on my part, I wanted to play with some HEXFET's, but it seemed a good opportunity to put this down so to speak. On a side note the amp design I use (single ended input) is very very stable and an amp with a differential input pair would give far more extreme results than this, with this type of layout . Try as I might it was hard to provoke this lash up into oscillation. If you look at the first shot you will see there is NO decoupling on the plugboard at all, and still it's stable.

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
Water cooling, practical matters.	kiwi_abroad	Solid State	31	22nd September 2008 10:03 PM
What really matters? Quality loudspeaker design.	Rounder	Multi-Way	15	6th November 2007 08:17 AM
Pass and IP matters	Magura	Pass Labs	12	31st March 2004 10:07 PM
Minimum volume matters	Raka	Multi-Way	27	24th December 2002 10:03 AM
What REALLY Matters?	Javaman	Subwoofers	3	12th March 2002 11:43 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

14th July 2008, 04:16 PM	#9
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	why does the input signal change with the loading?

14th July 2008, 04:19 PM	#10
Mooly diyAudio Moderator Join Date: Sep 2007	This was all a bit of fun on my part, I wanted to play with some HEXFET's, but it seemed a good opportunity to put this down so to speak. On a side note the amp design I use (single ended input) is very very stable and an amp with a differential input pair would give far more extreme results than this, with this type of layout . Try as I might it was hard to provoke this lash up into oscillation. If you look at the first shot you will see there is NO decoupling on the plugboard at all, and still it's stable.