Simple 100W power amp - Page 30 - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

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 1st July 2013, 07:25 AM   #291
dvv is offline dvv  Serbia
Sin Bin
 
Join Date: Sep 2004
Location: Belgrade, Serbia
Quote:
Originally Posted by djk View Post
...

The small power supply bypass caps (less than 1F) may well need low values of resistors in series with them to stop oscillations. They also need to be grounded at the point the load current is consumed (speaker ground). This is generally impractical, and the inductance in the leads and traces leads to oscillations, and the series resistance de-Q's the resonances leading to the oscillations.
An excellent way to reduce the possibility of oscillation is to marry a medium sized cap to the collector of ech output device, as near to the device as possible.

Nige asked about 1,000 uF, I'd rather see that as 2,200 uF - bigger, with more current potential.
  Reply With Quote
Old 1st July 2013, 09:55 AM   #292
AndrewT is online now AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Quote:
Originally Posted by dvv View Post
An excellent way to reduce the possibility of oscillation is to marry a medium sized cap to the collector of ech output device, as near to the device as possible........
This is the same as I have been recommending.

But where does the other cap lead go?

I have my theory that I use and have posted it, but I would like to hear your recommendation.
__________________
regards Andrew T.
  Reply With Quote
Old 1st July 2013, 11:47 AM   #293
dvv is offline dvv  Serbia
Sin Bin
 
Join Date: Sep 2004
Location: Belgrade, Serbia
Quote:
Originally Posted by AndrewT View Post
This is the same as I have been recommending.

But where does the other cap lead go?

I have my theory that I use and have posted it, but I would like to hear your recommendation.
To the ground, of course, where else, Andrew? The PSU ground, that is.
  Reply With Quote
Old 1st July 2013, 12:18 PM   #294
AndrewT is online now AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
But where?
You have told us precisely where to start, the device collector (for an EF).
Where is the HF decoupling Ground relative to the device and/or relative to the MF decoupling Ground?
__________________
regards Andrew T.
  Reply With Quote
Old 1st July 2013, 01:08 PM   #295
diyAudio Member
 
Join Date: Mar 2007
Location: Oxfordshire
Quote:
Originally Posted by AndrewT View Post
to Power Ground right next to the output devices.
I often use flying wires to TO 3 devices when I build prototypes . One has to be more careful if so . Being that the output device is a current amplifier usually it causes no problems and can facilitate an amp like here . I think saying be careful covers it . It is that I am moving slightly away from conventional thinking when adding local decoupling and if you like Formula 1 KERS . Thus I am apologizing in advance for complications it might bring . Small ones I am sure ? Sad to get it wrong as this is a very big deal and cheap . Panasonic 1000 uF FC series 63V if compatible seems where to start ?

I suspect a Star arrangement to be best ? I use a big brass bolt and eye tags when I do it ( 4 mm nickle platted , silver would be better I am sure ) . It looks like a flower . If the PCB has a big ground plane that might be better ? Solder to any point I guess ? I have never done this so can not say if it works well . I have looked at 3 layer PCB as the people I work with do it all the time . If doing home brew PCB I think it is possible to do 3 layer . Just make 2 PCB's and bring them together .

The Douglas Self's advice on Star is helpful . However my test show that it is not as easy as said . Sometimes doing the wrong thing works better ! The spectrum analyzer will say it sometimes . Even the way the analyzer is connected to the circuit can give false results . It is difficult . As DF 96 said simulators assume superconductors sometimes .

If doing it the wrong way works better it can mean a rethink will get the right way to work better still . Williamson in 1947 ( aged 19 ? ) said sometimes a Star is not possible . If so an intelligently used bus-bar is good . When I build valve amps I keep them simple so as to have Star . Also I try not to use a PCB .
  Reply With Quote
Old 1st July 2013, 01:31 PM   #296
diyAudio Member
 
Join Date: Mar 2007
Location: Oxfordshire
Quote:
Originally Posted by djk View Post
"I think 6 V/uS is right for 100 watts "

The Marsh and Jung criteria suggest between 0.5V~2V per peak volt of output. That would be more like a minimum of 20V/S for a 100W/8Ω amplifier.

Running out of drive current usually makes the waveform look like someone took a bite out of it in the 2nd and 4th quadrants.

The small power supply bypass caps (less than 1F) may well need low values of resistors in series with them to stop oscillations. They also need to be grounded at the point the load current is consumed (speaker ground). This is generally impractical, and the inductance in the leads and traces leads to oscillations, and the series resistance de-Q's the resonances leading to the oscillations.

As I have said in other threads I think it is greate advice that isn't fact based . If you like Malaria . London had the " bad air " idea for illness . The sewers built by Bazalgette are still in use today . They just about slew enough today . He used Portland cement which was experimental . Good that he did as these sewers are still working with serious modern flows . Almost like Techno music ? It is said to need high slew rates and crest factor allowance .

The last point is excellent . Dvv and I debated this in another way . I think your reason is the true reason and the resistor is required . A 1 R series resistance to the 1 uF if doing it without test gear ? To make an improved electrolytic into a device with a snubber seems Ideal . I did it on a 500 V capacitance multiplier recently . It was a hunch I just could not ignore . No evidence for any need . I just thought the resistor would do more good than harm . I used the well tested Zobel type of 10R + 0.1 uF . I had a bag of caps to use at no cost so did it . X2 suppression caps . I see no reason to use Audiophile ones . 2 x 220 nF in series at 500 V . Usual rating 385V DC per cap .
  Reply With Quote
Old 1st July 2013, 07:41 PM   #297
dvv is offline dvv  Serbia
Sin Bin
 
Join Date: Sep 2004
Location: Belgrade, Serbia
Quote:
Originally Posted by AndrewT View Post
But where?
You have told us precisely where to start, the device collector (for an EF).
Where is the HF decoupling Ground relative to the device and/or relative to the MF decoupling Ground?
Andrew, I have always used the PSU ground for decoupling from the PSU. This method has never failed me.

I have on several occasions also tried a special grouning line to the central star earthing point, but I feel I have gained nothing but complication.

Nice in theory, but in practice, nothing to write home about.

However, all that said, I must remind you that by default I use separate PSU lines for the input stage and the VAS on one hand, and the current gain stage on the other. Also, much along what Niogel's been talking about lately, rather than use a relatively complex discrete voltage regulator, I tend to use a "virtual battery" type of regulation, based on either Motorola/ON Semi MJE 15031/15032 BJTs or IRF 510/9510 MOSFETs. The MOSFET version is more finnicky because of fairly large voltage drops across the MOSFETs, but that's not really a problem if one plans for it from the outset.
  Reply With Quote
Old 2nd July 2013, 10:35 AM   #298
AndrewT is online now AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Quote:
Originally Posted by dvv View Post
Andrew, I have always used the PSU ground for decoupling from the PSU. This method has never failed me.
How long is that route during a half cycle of an AC waveform?

i.e. how far away is the PSU ground from the output devices?
__________________
regards Andrew T.
  Reply With Quote
Old 2nd July 2013, 09:07 PM   #299
dvv is offline dvv  Serbia
Sin Bin
 
Join Date: Sep 2004
Location: Belgrade, Serbia
Quote:
Originally Posted by AndrewT View Post
How long is that route during a half cycle of an AC waveform?

i.e. how far away is the PSU ground from the output devices?
Fairly near. Very near if you count the upper side ground plane. Let's say about 1 cm on average.

Remember, it's not so much proximity as it is trace width and overall area of leads. Thicker copper plating also helps. I mean, the industry routinely works with 35 uM of copper, better models use 50 uM, and I use 70 uM. If I could lay my greedy hands on some Mil Spec 120 uM glass fibre, I would commit a crime and nick it.
  Reply With Quote
Old 3rd July 2013, 09:22 AM   #300
AndrewT is online now AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Quote:
Originally Posted by dvv View Post
Andrew, I have always used the PSU ground for decoupling from the PSU. This method has never failed me.......
That PSU Ground is on the PCB and is only 1cm from the output devices.
Is that correct?
__________________
regards Andrew T.
  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
Which class D amp for a 100W +/-28VDC power supply ? 00940 Class D 3 29th August 2009 06:17 PM
Aem 6005 100w Power Amp Modules GeeVee Swap Meet 1 7th January 2007 09:36 AM
looking for simple discrete 100W DIY amp halusha Solid State 27 8th December 2006 11:30 PM
100W Amp from car battery as power source subashcobi Solid State 2 18th October 2006 03:13 AM
simple designs for 50-100W amplifer TomJ Solid State 2 7th February 2003 06:17 PM


New To Site? Need Help?

All times are GMT. The time now is 09:58 PM.


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