Krill - The little amp that might...

Status
Not open for further replies.
Jump to Latest
#1,422
Thanks for posting those results, Steve. Hope the tests go well.

I feel a need to summarize the various data so far. Is this right?

"high" is 0.113% THD or 1130ppm, 1Khz and 47Vpk
"medium" is 0.016% THD or 160ppm, 1Khz and 48Vpk
"low" is 0.008% THD or 80ppm, 1kHz and 29Vpk

You have measured a real circuit at 50ppm at 20kHz, 400W and 4 ohms (therefore 56Vpk) - post #78


How about a bit of competition to see who (other than Steve D), through cunning an ingenuity, can squeeze the most out of this design? Can anyone beat these figures in simulation or in a real circuit?

I don't know if cheating counts as cunning or ingenuity (no it does not :shy: ), but by using ZTX458 and 2SC5200 models and their exact complements, 1N4148 diodes and ideal CCSs, I managed to get my LTSpice THD20kHz to 160ppm into 8-ohms at 30Vpk.

Without cheating, using ZTX558 and 2SA1943, I get THD1kHz 180ppm at 30Vpk 8 ohms, and THD20kHz 250ppm.

Can anyone beat that in sims or in copper and sand? :c_flag:

Brian
 
#1,423
Sorry folks.....

I have a genuine hard time evaluating a circuit based on numbers...To me what it soundslike is the key....and when it comes to an output stage the important this is how well does it behave...is it stable..does it thump whe turned og...is it expansive or supressive in it's sound signature...how natural does cymbals sound how is the deccay in the base...Sure these things have nothing to do with one simpel parameter such as distortion and for sure not with simulated distortion....

Please hav a look at superb amplifiers such as the VTL Sigfreds or S400.. they sound fantastic musical coherrent, makes the emotions of the artist flow out of the speakers...and have distortion measured more than 10 times higher than this circuit....

What i find interesting when talking distortion is that this circuit appers to have lover distortion at higher frequencies..At this point it resembles nature better than convetional output stages.....maby a good point.....
 
#1,425
Originally posted by anatech
Hi Joshua,

I really think you are reading far too much into things. I know that syn08 never questioned Steve's integrity. It was the design, not the man and that is fair within limits.
…
Nope. It wasn't that bad. Underdog complex?
Click to expand...


When one's design, along with one's claims about that design, are questioned, then one's integrity is questioned. That is, when the questioning is done without actual measurements, not based on facts.

It's regrettable that some people don't see it.


Originally posted by anatech
There is no need for you to take up sword and shield to defend anyone. This type of posting will generally only prolong "hostilities".
Click to expand...


I wonder, especially when I know how deeply Steve was hurt by some posts in this thread.

If my resenting to hurting people brings about hostility, so be it.
 
#1,427
traderbam said:
Thanks for posting those results, Steve. Hope the tests go well.

I feel a need to summarize the various data so far. Is this right?

"high" is 0.113% THD or 1130ppm, 1Khz and 47Vpk
"medium" is 0.016% THD or 160ppm, 1Khz and 48Vpk
"low" is 0.008% THD or 80ppm, 1kHz and 29Vpk

You have measured a real circuit at 50ppm at 20kHz, 400W and 4 ohms (therefore 56Vpk) - post #78


How about a bit of competition to see who (other than Steve D), through cunning an ingenuity, can squeeze the most out of this design? Can anyone beat these figures in simulation or in a real circuit?

I don't know if cheating counts as cunning or ingenuity (no it does not :shy: ), but by using ZTX458 and 2SC5200 models and their exact complements, 1N4148 diodes and ideal CCSs, I managed to get my LTSpice THD20kHz to 160ppm into 8-ohms at 30Vpk.

Without cheating, using ZTX558 and 2SA1943, I get THD1kHz 180ppm at 30Vpk 8 ohms, and THD20kHz 250ppm.

Can anyone beat that in sims or in copper and sand? :c_flag:

Brian
Click to expand...


The figures I gave for the 100W amp can be seen here.

http://www.diyaudio.com/forums/showthread.php?postid=1776001#post1776001

What I may or may not have measured at some point on another amp is not relevant to this thread. That is NOT the amp this thread is about. That has been pointed out repeatedly, but you and some others seem unable to grasp that simple fact.

I see you have chosen to ignore or disregard simulations that gave results similar to or better than mine. Many simulation results have been posted here by many members. Some have been better than I believe possible, some have been close to mine and some have been quite bad. Just pick one you like.

Simulation results were requested (insisted on?), so I posted mine. Now it seems simulation results are useless unless performed by a select few using models not available to the rest of us. I am certainly glad I built and measured this design BEFORE I tried to simulate it.

What distortion did you get at 1KHz? That is where I was measuring the amps this thread is about.

The low distortion results are attached. All 3 should have been at the same output. That was my oversight.
 

Attachments

#1,429
Steve Dunlap said:
That is NOT the amp this thread is about. That has been pointed out repeatedly, but you and some others seem unable to grasp that simple fact.
Click to expand...
Are you suggesting I judge your posts based on what other people suggest you meant by your posts?

It must be this Guinness that's affecting my perception.

Actually, I did post THD for the 100W amp being discussed in this thread. THD at rated power (100W) into 8 ohms, measured at 1KHz is 0.01% or less.
Click to expand...

Ah, the world suddenly seems much more sensible. Must be this Guinness.

I see you have chosen to ignore or disregard simulations that gave results similar to or better than mine. Many simulation results have been posted here by many members. Some have been better than I believe possible, some have been close to mine and some have been quite bad. Just pick one you like.
Click to expand...

Good. They can speak for themselves, I'm sure. Unless they've had too much Guinness.

What distortion did you get at 1KHz? That is where I was measuring the amps this thread is about.
Click to expand...

I believe I said. Are you drinking Guinness too?

The low distortion results are attached. All 3 should have been at the same output. That was my oversight.
Click to expand...

It's the poison nectar of the Leprechaun. It'll do it everytime. :drunk:
 
#1,433
Re: Some real measurements...

Andypairo said:
At 80W output power ,1 kHz,
input THD 0.006%- 0.007%
output THD 0.018%-0.03%
Click to expand...

Thanks - useful data point. I make that 180ppm to 300ppm, at 36Vpk (assuming 8 ohm load).
Are you using the same transistors as in SD's latest schematic?

Regarding other measurements...what can you measure? What equipment have you got?
 
#1,434
Hi Phil,
as I have said in the construction thread I heatsinked Q7 and Q10.

Before taking the measurements I however found that nulling the output stage offset (with R32/R33) resulted in a much higher bias current than before, so for the first time I could decrease R27 and still have a reasonable bias current. This seems to make sense, since adding say 20 Ohms results in nearly 200 additional millivolts to bias the output stage.

Ciao

Andrea
 
#1,435
Re: Re: Some real measurements...

traderbam said:


Thanks - useful data point. I make that 180ppm to 300ppm, at 36Vpk (assuming 8 ohm load).
Are you using the same transistors as in SD's latest schematic?

Regarding other measurements...what can you measure? What equipment have you got?
Click to expand...

Actually the load is more 10 Ohms (2 5R1 series resistors), but it shouldn't matter that much.
I'm using 2SC3067 for the diff pair, 2SC2240-2SA970 for the "small" transistors, 2SA1837 and 2SC4793 from Steve and MJL1302A-3281 for driver and output.

My "equipment" for distortion measurement is a sine generator, oscilloscope and PC soundcard.

Ciao

Andrea
 
#1,436
Hi Andrea --

I remember you saying you had a sub-optimal heatsinking arrangement and was not aware that you got things working. Good to hear all is OK.

I also need to heatsink Q7 and Q10 before going to higher power. I hope to post some preliminary results later today or tomorrow.

Phil
 
#1,437
Hi Andrea & Phil -

Great that you are making progress with your construction and thanks for the distortion figures.

How about some of your listening impressions when you have time to get around to them.

If you can take a look at the odd harmonics that could show some helpful information.

:cheers:
 
Status
Not open for further replies.
Top Bottom