Amp-Load interaction

zeus_threat · 2nd June 2006, 02:53 PM

Hi, i've built a prototype of sym-a-sym5 but i have found that i needed to add the attach filter to the amp's input to get good bass reproduction. Could anybody plz mention how the sound can be improved without this network?

AndrewT · 2nd June 2006, 04:32 PM

Hi,
it is quite normal to have a DC blocking cap and RF filter on the input to a power amplifier.
I would recommend you keep to that.

However your values are a bit unconvential.

4u7F and 221r gives an RC time constant of 1.04mS and rolls off the bass from 1.5kHz at -6db/oct.
Could that be a misprint and r1=22k? Then the -3db frequency changes from 1.5kHz to 1.5Hz.
If your source has a DC blocking cap at it's output, then you can bypass the input cap to the power amp.

The 10k series into the combined 47nF + 4k7 is beyond my maths/electronics analysis knowledge. But usually you aim for an RC time constant of 0.5uS to 2uS to roll off the ultrasonics and RF.
1k0 and 1nF achieve this plus a range either side depending on your preference and signal bandwidth.

anatech · 2nd June 2006, 04:38 PM

Hi AndrewT,
The correct value for R1 is in fact 22K.

-Chris

AndrewT · 2nd June 2006, 04:55 PM

Hi,
got an expert's attention!
How do you analyse the HF filter turn over frequencies, presumably pole and zero?

zeus_threat · 2nd June 2006, 05:00 PM

Sorry guys here is the actual rc network i added. But its not a question of time constant here its a matter of frequency response. Only when this network added that i get a good bass response as it attenuates the high frequency response by 6dB from the sym-a-sym5. Could it be due to the amplifier load interaction or some bias problem somewhere in the amp. Without this network there is excess high frequency response. This is not due to the load or the pre-amp they have been checked seperately and don't have this problem. Its with the amp.Can anyone help? Thanks

anatech · 2nd June 2006, 05:51 PM

Hi zeus_threat,
I would recheck your parts values used in your amplifier. I can only imagine that an error has been made. The amp design itself does not have this problem based on what the schematic looks like, and other reports from builders.

The main suspects would be the resistors and capacitors used in the feedback network and the input network as Andrew pointed out.

Bias current level may cause crossover distortion, but never a frequency dependant effect by itself. The output impedance is low enough to cause fuses to blow if it were an amplifier / load interaction.

AndrewT,
Thanks, but MikeB and Pavel are the experts on this amp!

-Chris

MikeB · 2nd June 2006, 06:49 PM

As i never had complaints about weak bass from symasym before, it's likely that there is an issue inside the amp...
Which version did you build, the latest ? Do you have two channels behaving the same or one prototype ? As Chris already told, have a look at the 470uF in the feedbackpath. If it's defect the paralell 100nf would do the job and kill all lower freqs.

That's a measured freq-response from symasym: (dominated by measuring setup)

Mike

zeus_threat · 3rd June 2006, 12:18 PM

Well all that has been mentionned has been checked. Thanks to your previous post mike i found an easy to measure the freq. response using rmaa. I have only 1 prototype. I fed the left out of my sound card to the amp input and sent the amp output to the left line in of the s.card. Taking care of course not to fry it. Attached is the freq response measurement. Its still not giving me a clue as to why am not getting any good bass from this amp. The speaker has been tested with a yamaha AVX496 amp(with all tone controls neutral) and works fine. But the tweeter is playing too loud with my protoype. I even plugged a headphone into my soundcard(SBLIVE5.1) and then plugged it into the prototype's output but there was a clear rise in high freq response from the amp. Any help would be welcome.

MikeB · 3rd June 2006, 12:39 PM

Hi zeus, this freq response is definitely a disaster, it must sound worst.
Symasyms freq response is definitely flat, have you checked for cold soldering ? Another possibility would be that it is oscillating...
The most amazing is that the response falls that steep above 10khz.
Have you checked if biasing is normal ? Start with the voltages across the 2 680 ohms (r5/6). These should read both ~1v. Also important are r15/17 (150ohms), these should read both ~375mv.
Have you double checked component values ? (especially caps) Maybe you got the pinout of one of the small signal transistors wrong ?

Mike

zeus_threat · 3rd June 2006, 01:23 PM

Sorry i forgot to mention the white line is frequency response and the green line is 2nd and 3rd order harmonic distortion. This attahcment is the correct one

2nd June 2006, 04:32 PM	#2
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	Hi, it is quite normal to have a DC blocking cap and RF filter on the input to a power amplifier. I would recommend you keep to that. However your values are a bit unconvential. 4u7F and 221r gives an RC time constant of 1.04mS and rolls off the bass from 1.5kHz at -6db/oct. Could that be a misprint and r1=22k? Then the -3db frequency changes from 1.5kHz to 1.5Hz. If your source has a DC blocking cap at it's output, then you can bypass the input cap to the power amp. The 10k series into the combined 47nF + 4k7 is beyond my maths/electronics analysis knowledge. But usually you aim for an RC time constant of 0.5uS to 2uS to roll off the ultrasonics and RF. 1k0 and 1nF achieve this plus a range either side depending on your preference and signal bandwidth. __________________ regards Andrew T.

2nd June 2006, 04:55 PM	#4
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	Hi, got an expert's attention! How do you analyse the HF filter turn over frequencies, presumably pole and zero? __________________ regards Andrew T.

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
Dynamic load line analysis using music and speaker load	Michael Koster	Tubes / Valves	0	7th March 2008 08:47 PM
loudspeakers and room interaction	jlo	Multi-Way	1	25th July 2007 08:17 PM
Aleph CS and interaction with PS	mchapman	Pass Labs	2	5th June 2006 08:46 PM
Crossover slope interaction ?s	toenail	Multi-Way	0	1st May 2003 01:06 AM
Magnet interaction?? (Car Audio)	WhtCastle	Everything Else	1	26th January 2002 02:24 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

2nd June 2006, 04:38 PM	#3
anatech diyAudio Moderator Join Date: Jun 2004 Location: Georgetown, On	Hi AndrewT, The correct value for R1 is in fact 22K. -Chris

2nd June 2006, 05:51 PM	#6
anatech diyAudio Moderator Join Date: Jun 2004 Location: Georgetown, On	Hi zeus_threat, I would recheck your parts values used in your amplifier. I can only imagine that an error has been made. The amp design itself does not have this problem based on what the schematic looks like, and other reports from builders. The main suspects would be the resistors and capacitors used in the feedback network and the input network as Andrew pointed out. Bias current level may cause crossover distortion, but never a frequency dependant effect by itself. The output impedance is low enough to cause fuses to blow if it were an amplifier / load interaction. AndrewT, Thanks, but MikeB and Pavel are the experts on this amp! -Chris

2nd June 2006, 06:49 PM	#7
MikeB diyAudio Member Join Date: Apr 2004 Location: Gütersloh	As i never had complaints about weak bass from symasym before, it's likely that there is an issue inside the amp... Which version did you build, the latest ? Do you have two channels behaving the same or one prototype ? As Chris already told, have a look at the 470uF in the feedbackpath. If it's defect the paralell 100nf would do the job and kill all lower freqs. That's a measured freq-response from symasym: (dominated by measuring setup) Mike

3rd June 2006, 12:39 PM	#9
MikeB diyAudio Member Join Date: Apr 2004 Location: Gütersloh	Hi zeus, this freq response is definitely a disaster, it must sound worst. Symasyms freq response is definitely flat, have you checked for cold soldering ? Another possibility would be that it is oscillating... The most amazing is that the response falls that steep above 10khz. Have you checked if biasing is normal ? Start with the voltages across the 2 680 ohms (r5/6). These should read both ~1v. Also important are r15/17 (150ohms), these should read both ~375mv. Have you double checked component values ? (especially caps) Maybe you got the pinout of one of the small signal transistors wrong ? Mike

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)