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 17th December 2003, 09:04 AM   #11
sreten is offline sreten  United Kingdom
diyAudio Member
 
Join Date: Nov 2003
Location: Brighton UK
Well according to my amplifier handbook (D.Self) the
earth lug is correct but the chassis ground is wrong.

According to the above the chassis should be connected
to the ground at the inputs and at no other points.

I wouldn't call that small section a ground loop at all,
and cutting the track will not help.

sreten.
  Reply With Quote
Old 17th December 2003, 09:50 AM   #12
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Strange, I've always been told that the inputs should be isolated from the chassis...... Maybe I'm not visuallising this correctly

I think I'm refering to the earth lug and the chassis ground as being the same thing, chassis basically being the case of the amp, have I got the terminology wrong? woops just noted that my diagram shows both the earth lug and the gnd lug, and I did ask about it.... so the mains earth is ok, but the star ground on the pcb shouldn't be connected back to the chassis lug????? That sounds strange to me....... especially since just about everything I've read about star grounding (admittedly not much) says to keep the large currents as far from your input ground as possible.

I think I'll just try some stuff (not too radical, at least not yet...) and measure and listen to what difference positive or negative (if any) it makes.

Oddly after doing the last measurements, I went downstairs and hooked the speakers back up, and the left channel was noticably quieter (especially in the tweeter), the noise from the tweeter changed from a sort of muli freq hum to a low hiss) but the right channel has not improved at all.....

The weird thing is that on examining the noise graphs from rmaa It doesn't look like one channel is consistantly worse than the other (the noise doesn't coincide the plots continually intersect one another one higher at one point then it swaps to the other) the differences in magnitude are one or two db at most (around the -120db mark) Note that in order to here this my ear needs to be within about 20cm of the drivers, so it's not really that big a deal, it basically annoys me (especially now that its much worse on one channel than the other!)

Yeah I kind of figured that the bit I thought might be a ground loop was a bit far fetched wishfull thinking on my part, ultimately either way the current is finding the same way to ground.

Tony.
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote
Old 17th December 2003, 10:07 AM   #13
sreten is offline sreten  United Kingdom
diyAudio Member
 
Join Date: Nov 2003
Location: Brighton UK
Quote:
I think I'm refering to the earth lug and the chassis ground as being the same thing, chassis basically being the case of the amp, have I got the terminology wrong? woops just noted that my diagram shows both the earth lug and the gnd lug, and I did ask about it.... so the mains earth is ok, but the star ground on the pcb shouldn't be connected back to the chassis lug????? That sounds strange to me....... especially since just about everything I've read about star grounding (admittedly not much) says to keep the large currents as far from your input ground as possible.
You would need to reconnect the PCB GND wire for the star ground.
Disconnect the ground lug, but not the earth lug.
Connect the chassis to PCB GND at the input requiring lowest noise.
There will be no large currents in this input ground.

sreten.
  Reply With Quote
Old 17th December 2003, 02:55 PM   #14
sreten is offline sreten  United Kingdom
diyAudio Member
 
Join Date: Nov 2003
Location: Brighton UK
I've studied the PCB and can't see how the PA sections
would have such high crosstalk on there own.
Something about the testing methodology perhaps ?

sreten.
  Reply With Quote
Old 17th December 2003, 09:12 PM   #15
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Yes I was wondering about that. I'm going to test it with no load resistors and see what the result is.

The other thing I wonder is whether it is the power supply. The fact that when I input a 1Khz sine wave, the ripple on the power supply has the 1Khz sine wave superimposed on it, makes me suspicious.

Also when I did a test at low volume (bumped up the gain on the soundcard to get an ok level) the crosstalk was around -65db which is much better.

certainly if I listen to music with very strong channel separation I don't hear any leakage from one channel to the other.

When I was testing with the soundcard directly connected to the amp, something was "singing" there was a high pitched noise bit like a very high tuning fork coming from somewhere in the amp, couldn't pinpoint where, and didn't want to leave it that way for too long in case something cooked (the load resistors were already getting too hot...... maybe that's part of the problem)....

Tony.

edit: Just tryed another test. I connected only the left channel of my dvd player, connected the right channel speaker to the B speaker output, selected speakers B and listened. At normal listening volume the music could be heard very faintly in the right speaker at full volume it was still probably 1/4 of the volume I would normally listen at, also was mainly in the upper freq range, no bass or mid bass). I'll try again with the resistor in place of the left speaker so that there is actually a load on the left channel when doing the test. But first I have to go shopping bah humbug
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote
Old 17th December 2003, 09:16 PM   #16
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Quote:
Originally posted by sreten


You would need to reconnect the PCB GND wire for the star ground.
Disconnect the ground lug, but not the earth lug.
Connect the chassis to PCB GND at the input requiring lowest noise.
There will be no large currents in this input ground.

sreten.
I don't understand this..... If I disconnect the existing star earth connection and instead connect it to the input ground, then ALL other currents will then be redirected back through the ground at the inputs to the chassis earth, and all of a sudden you DO have large earth currents at the input??????

Tony .... confused
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote
Old 18th December 2003, 02:37 AM   #17
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Default latest tests

Ok,

I tried the listening test with and without dummy load on left channel with only the dvd left channel connected. at all but very high volumes, switching in the dummy load caused a very slight increase in the volume comming from the right channel speaker. If pushed to the point just before clipping the change was more noticable. If I pushed it way past clipping the sound from the right channel became quite a bit louder and very distorted. Note that at full volume without the dummy load on the left channel the sound from the right channel was not distorting. From this I'm tempted to conclude that the problem lies with the power supply.

I also did the rmaa test again without the dummy load, result was that the stereo crosstalk figure improved to -79.2db quite respectable really, I would again say probably an indication of power supply weakness.

If I have time tonight I'll re-wire the power supply and test again. Also interestingly the marked difference in noise levels between the left and right channels is now showing up in rmaa. could be something I've desoldered and resoldered while separating stages of the amp. All tests from now on, I'm going to do through both preamp and power amp sections, treating the amp as a whole.

If I get really adventurous I may pull out the spare 300VA torroidal I have, only prob is I don't have any filter caps to use with it, and buying a set as an experiment is a little bit expensive! I guess I could parralel up the two windings on each and use one torroidal for -ve rail and one for the +ve rail. At least that way I wouldn't need any more caps, and I'd have twice the current at my disposal

Tony.
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote
Old 18th December 2003, 11:00 AM   #18
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Well I rewired the power supply and if nothing else I learned that 8 gauge wire is a b*tch to work with, probably should have got the 12 gauge!

Having said that after only changing the wire between the caps and the rectifier, and changing the way the wire linked back to the earth point, and finally changing the main wire from the star ground on the pcb to the chassis ground to 8 gauge, the crosstalk with the load resistors improved from -37.5db to -43.5db, another 6db improvement...... what is it with 6db

Still not real good but at least it's an improvement. The downside is that the noise got worse (probably not surprising when fiddling with the ground wiring....... I don't think moving the 0V wires from the xformer was such a good idea..... I'll move it back and hopefully it will go back to previous noise levels. When I say got worse, it's only noticably worse when the amp is cranked right up with no signal (and rmaa shows signs of the high freq noise comming back..... Oh well maybe I'll get somewhere tomorrow

Tony.
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote
Old 18th December 2003, 12:10 PM   #19
sreten is offline sreten  United Kingdom
diyAudio Member
 
Join Date: Nov 2003
Location: Brighton UK
Quote:
Originally posted by wintermute


I don't understand this..... If I disconnect the existing star earth connection and instead connect it to the input ground, then ALL other currents will then be redirected back through the ground at the inputs to the chassis earth, and all of a sudden you DO have large earth currents at the input??????

Tony .... confused
The existing ground (chassis connection) is not the star point.

The star point is on the PCB and should connected via heavy
wire to the centre point (OV) of the power supply capacitors.
Heavy connections for the +/- rails will reduce rail ripple.
All the P.S. wiring should be heavy duty,
especially to and from the diode bridge.

The case should be connected to mains earth at the mains input.
The earth lug is correct.
The ground lug should be disconnected.

Asumming the case is single insulated mains earth and the star
point will now be not connected, but the amplifier works.

The case is connected to the star point with a short link to
the single ground return track next to the phono socket.
No extra operating currents will flow through the ground return.

(P.S. does your PC have a two wire or 3 wire mains connection ?)

sreten.
  Reply With Quote
Old 18th December 2003, 09:10 PM   #20
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Thanks sreten. I'll give that a try... Yes the pc has a 3 pin plug, I'm assuming that the case is grounded.

I guess thinking about it 0V is just a reference, and it doesn't have to be earth.... so the current flow will be back to 0V not to the earth point.

I wired up the 0V wiring on the caps as in the diagram thinking this should be better than the existing way. Should I put the ceter tap from the transformer to the centre of the 0V wiring or directly onto one of the caps? I'm leaning towards directly to a cap at the moment (its currently on the gnd lug)......

Tony.

edit: it's times like this I wish I had a digital camera
Attached Images
File Type: gif caps.gif (2.8 KB, 211 views)
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  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
Hi All new here quest. about kappa pro drumsz Subwoofers 0 7th November 2008 10:06 PM
The FR125-Spiral project has begun! eVITAERC Full Range 66 25th April 2008 05:56 PM
813 Project has begun ! Steven-H Tubes / Valves 2 27th June 2007 06:14 PM
The Quest millwood The Lounge 14 26th July 2003 04:59 PM


New To Site? Need Help?

All times are GMT. The time now is 04:16 AM.


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