Are T amps just over hyped Distorting noise boxes? Or am I doing it wrong? - Page 4 - diyAudio
Go Back   Home > Forums > Amplifiers > Class D

Class D Switching Power Amplifiers and Power D/A conversion

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 16th November 2012, 11:25 PM   #31
diyAudio Member
 
Join Date: Apr 2010
Location: In an old house on a hill needing repair looking over an old cemertry.
Please let me know how it sounds. Thanks for the link. I am going to buy one of those media players
  Reply With Quote
Old 19th November 2012, 04:43 PM   #32
lausar is offline lausar  United States
diyAudio Member
 
Join Date: Sep 2010
Quote:
Originally Posted by radiosmuck View Post
I'm driving a pair if 15 inch 15 ohm vintage Tannoys with an SMSL 2020 amp and they sound just wonderful.
I swapped out my original Lepai for visual reasons, plus the SMSL is suppposed to have better components.
You are so right about speaker efficiency.
I'm currently driving my vintage 16ohm Altec's with a Topping 2020 and it sounds wonderful also.
  Reply With Quote
Old 23rd November 2012, 04:16 AM   #33
wmb is offline wmb  United States
diyAudio Member
 
Join Date: Nov 2004
Location: Hawaii
Regarding the Y148 boards on DealExtreme: From the picture, the output filter inductors appear to be small ferrites, so they are probably subject to inductance change due to temperature rise if played loud for awhile. The output capacitors are surface mount types, probably ceramic, so the will be subject to the distortion mechanisms mentioned in Understanding output filters for Class-D amplifiers and Trevor Marshall - Class D Audio Amplifier Design - TDA7498 Output filters . But hey, for $9.50, what do you expect? You won't get premium inductors and caps for that price.

Regarding driving 15 ohm speakers: The problem is that the output filter is designed for 8 ohm speakers. The purpose of that filter is to suppress the class-D switching frequency which is in the 50 kHz range. The optimum L and C values for class-D output filters depend on the load impedance. The inductors shown in the photo are 22 uH, which is the correct value for 8 ohm speakers. If you connect a speaker with a different impedance, the output filter will still operate, but its damping factor factor will be wrong. For 15 ohm speakers, the "Q" of the filter would be too high, which is the same as saying that it is "underdamped". What that means is that the filter will "ring", and more of the 50 kHz signal will be present on the output. That could cause the chip to see transient voltages or currents in excess of those it can easily handle. That could result in distortion or damage to the chip.

Class-d (and class-t) amps thus work best when the speaker impedance matches the design point for the output filter circuit. As a rule of thumb, if the filter was designed for 8 ohms but you want to use a 4 ohm speaker, you halve the output inductor values (2 inductors per channel) and double the capacitor values (2 or 3 capacitors per channel). To go from 8 ohms to 16 ohms, double the inductors and halve the capacitors. That's the rule of thumb - to really do it right you need to consider the speaker inductance, and you may also need to add an additional "snubber network" to deal with currents flowing in the circuit when the output transistors are turned off. The two links cited above give the gory details.
  Reply With Quote
Old 23rd November 2012, 02:01 PM   #34
diyAudio Member
 
radiosmuck's Avatar
 
Join Date: Jul 2012
Location: Montreal
Can the above be verified with ones ears? That is, could could an "expert" say to me, "You have the wrong inductors in your amp for those speakers, I can tell by the sound"?
The reason I ask is that I have several sets of still highly regarded speakers and they all sound great hooked up to my T2020 amp. They range from 4ohm to 15ohm and very small to huge.
The best are my huge old 15ohm jobs, they just purr along drawing just under 1 watt.
  Reply With Quote
Old 23rd November 2012, 02:43 PM   #35
Fynn is offline Fynn  Germany
diyAudio Member
 
Join Date: Nov 2012
Wouldn't it be way easier to put an 8ohm (10W+ rated) resistor in parallel to an 8ohm speaker, so it will count as a 4 ohm load for the entire output filter. Of course you "waste" half of the power on that resistor.
But: the ta2020 has almost double the power on 4ohm than on 8ohms. So in the end the power to the 8ohm speakers will stay the same, just the ta2020 will be running hotter (4ohm has lower efficiency and you will put out double the wattage).
  Reply With Quote
Old 23rd November 2012, 02:54 PM   #36
Pano is offline Pano  United States
diyAudio Moderator
 
Pano's Avatar
 
Join Date: Oct 2004
Location: Milliways
Blog Entries: 4
It's mostly the top end of the response that is affected by the output filter. See my measurements here:
Super-T Response
Sonic Impact Response
Sonic Tripath


Putting a parallel resistor across the terminals is a crude approach. Better to tailer the impedance of your tweeter, or the values of the output filter. It's the response above 8-10Khz that really changes with speaker impedance.
__________________
Take the Speaker Voltage Test!
  Reply With Quote
Old 23rd November 2012, 04:16 PM   #37
Fynn is offline Fynn  Germany
diyAudio Member
 
Join Date: Nov 2012
The upper frequency band still seems fine though. 1-2db variance in the last 1/3rd octave of your upper hearing range seems insignificant. Same goes to the sub 50hz band, seeing as that you aren't listening on a free field and your room has so many modes in that region which can easily reach > +-10db.

I would rather be concerned about possible distortion or other artifacts, than some attenuation in the most upper and lower frequency ranges. Because those could be very audible.
  Reply With Quote
Old 23rd November 2012, 04:20 PM   #38
Pano is offline Pano  United States
diyAudio Moderator
 
Pano's Avatar
 
Join Date: Oct 2004
Location: Milliways
Blog Entries: 4
Agreed. And some fullrange drivers can use the boost at the top, anyway.
  Reply With Quote
Old 23rd November 2012, 06:13 PM   #39
wmb is offline wmb  United States
diyAudio Member
 
Join Date: Nov 2004
Location: Hawaii
Regarding the audibility of output filter mis-damping: I suspect that it is not directly audible in and of itself, but I could imagine situations where it caused amplifier instability that would be audible. It probably depends on several factors including the supply voltage (if close to the maximum, the chip would be more susceptible to overvoltage induced by filter ringing). Other factors could include the inductance and capacitance of the cables and the presence of RF interference or other noise coupled back into the circuit. A well-tuned output circuit will be less susceptible to such problems - but if those problems don't exist, maybe the mis-tuning itself is benign. If, as you say, the system sounds good to your ears in your setup, you win.

I once connected a 2-ohm load (four 8-ohm speakers in parallel) to one channel of a Sure 4x100 T-amp and it sounded just awful, even at low volumes. The amp was clearly freaking out. It sounded fine at 4 ohms. The problem mechanisms for too-high impedance are rather different - overvoltage for too-high impedance vs. overcurrent for too-low - but it is clear that, in some circumstances, impedance does matter.

Regarding a parallel resistor: If the output circuit is turned for 8 ohms - as appears to be the case for the two T-amps that I have (HLLY 20 and Sure 4x100) - dropping it to 4 ohms will make things worse (albeit overdamped instead of underdamped). But paralleling a 16 ohm resistor across a 15 ohm speaker to bring it down to 8-ish ohms would indeed do the right thing for an 8-ohm-tuned output filter. It would not change the amount of power instantaneously available to the speaker, because the voltage across the speaker would be the same in either case. The amplifier would still be operating within its rated I-V conditions. The only possibly-audible downside might be a reduction in the reserve power available for extended bass passages. I doubt that would be a problem if the power supply is decent.

The resistor power rating would not need to be all that high for normal listening. For music, you want to listen at average power levels lower than about 1/10 the rated amp power. At higher average levels, the musical transients will exceed the amp power and clip, and that is definitely audible. For a low-power T-amp, a 5W resistor would be more than adequate. Full-power bench testing is obviously a different story.
  Reply With Quote
Old 23rd November 2012, 06:35 PM   #40
Pano is offline Pano  United States
diyAudio Moderator
 
Pano's Avatar
 
Join Date: Oct 2004
Location: Milliways
Blog Entries: 4
Quote:
Originally Posted by wmb View Post
But paralleling a 16 ohm resistor across a 15 ohm speaker to bring it down to 8-ish ohms would indeed do the right thing for an 8-ohm-tuned output filter. It would not change the amount of power instantaneously available to the speaker, because the voltage across the speaker would be the same in either case.
If the output impedance of the amp is low enough, yes. In this case it mostly is.
__________________
Take the Speaker Voltage Test!
  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
What is wrong with op-amps? Gordy Solid State 270 31st August 2012 06:28 AM
Buss Channel with Noise after wrong soldered bypass Cap. suchenderxxx Analog Line Level 12 11th July 2012 09:58 PM
Wrong Voltage / current on BoZ... or Wrong Measuring ? gionag Pass Labs 8 14th May 2008 11:24 AM
Amps mounted on speaker boxes? mr__mcgoo Car Audio 6 10th June 2004 03:58 PM
What are the things that may go wrong with amps. downhere Solid State 1 9th December 2001 07:28 PM


New To Site? Need Help?

All times are GMT. The time now is 04:18 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