I'm personally not the best to debate on "the sound of Krill", but I haven't seen so far a shred of significant and credible measurement and eventual correlation with the topology and listening results.
This being said, for those who love the Krill, I'm not the one to poop your party. Enjoy!
I think I have said almost nothing about the "sound of Krill" other than I like it. I have attempted to stay away from that subject precisely because it is personal taste and not subject to measurement. Many people like the sound of me amps, many others prefer other amps.
If you do ever see any credible evidence, it will not come from me. I believe you know why. I will gladly work with anyone that is interested in running the tests.
In case it isn't clear to everyone here, I have no commercial involvement with any company. My venture into the commercial end of audio was many years ago and was over by 13 years ago. I made what most would consider very affordable equipment. The only thing I try to sell today is equipment I can no longer use and parts I still have in stock.
Steve, is it time to move on & reveal the next iteration in the design? There will never be the consensus that you hoped for - all those who are on-board are eager to hear what other clever design techniques you have up your sleeve. I believe it being widely recognised as a clever design is the highest accolade - it's like a DIY audio oscar 🙂
I will start another thread for that. I want to do some more on my "Build your own case" thread in the Construction Tips forum first. I don't handle multitasking too well any more.
Take all the time you want - I didn't mean it as any kind of pressure, just to mark the fact that there are lots of people here who would like to see/hear the further details. You do it in whatever time-frame is enjoyable to you - that's what a hobby is all about 🙂
There is one last loose end to tie up in this thread. We were looking at what differences there would be between LTspice and the program I use (Intusoft).
You're right, It was too good to be true. I checked the bias in the LTspice file and it was much higher than in the Intusoft file I had referenced earlier. I changed the value of Rbias to 882 ohm to set the bias at the same level as I had used earlier when I posted the minimum current values.
First, let my say I used the schematic I posted in post 127 with no changes other than the value of Rbias. I used the models that came with my program for the results I posted in post 107. I used Andy's models that were supplied with his sim file for the results I am posting now.
Here are the figures for minimum current flowing in an emitter resistor as calculated by both programs. The first is the results I posted in post 107 using Intusoft.
This is what I got using LTspice.
7.24ma minimum at 20K
9.15ma minimum at 1K
9.12ma minimum at 20 Hz
DC – 9.12ma with 28.3VDC at output with 8 ohm load
You can draw your own conclusions about these results, but I feel they are close enough, considering they were run with different models, to be called the same.
You're right, It was too good to be true. I checked the bias in the LTspice file and it was much higher than in the Intusoft file I had referenced earlier. I changed the value of Rbias to 882 ohm to set the bias at the same level as I had used earlier when I posted the minimum current values.
First, let my say I used the schematic I posted in post 127 with no changes other than the value of Rbias. I used the models that came with my program for the results I posted in post 107. I used Andy's models that were supplied with his sim file for the results I am posting now.
Here are the figures for minimum current flowing in an emitter resistor as calculated by both programs. The first is the results I posted in post 107 using Intusoft.
This is what I got using LTspice.
7.24ma minimum at 20K
9.15ma minimum at 1K
9.12ma minimum at 20 Hz
DC – 9.12ma with 28.3VDC at output with 8 ohm load
You can draw your own conclusions about these results, but I feel they are close enough, considering they were run with different models, to be called the same.
Less than 10ma ?
That sounds more like classB
And if "switching" looks more like classA, it must be special
That sounds more like classB
And if "switching" looks more like classA, it must be special
Last edited:
Ah, I get it, slowly
Its like when PP classA changes "bias" from side to side(+ to -), but never goes completely OFF leaving classA
Ehh, how does it look in 4ohm ?
Less output within these specs, or stays the same ?
Its like when PP classA changes "bias" from side to side(+ to -), but never goes completely OFF leaving classA
Ehh, how does it look in 4ohm ?
Less output within these specs, or stays the same ?
Last edited:
Those numbers are the minimum current in an emitter resistor for sine wave output at the frequencies shown. Compare that to the 1.46ma for the D. Self amp in post 2 of this thread. The output stage is dissipating 131W at idle. Mine is dissipating 55W at idle.
Thanks fore explaining, sounds really good
And into 4ohm ?
Hmm, stays the same, 55watt ?
I only checked at 1K for a 4 ohm load using LTspice. The minimum current was 1.69ma. See post 3 in this in this thread for the same sim with the D. Self amp.
Shouldnt be any problem to raise bias on a cool amp 😀 or ?
So, what happens when going OFF switching ?
Hard "classB-switching", or softer "classAB-switching" ?
I hope you know what I mean
Anyway, I rate normal speakers to 6ohm at the most, so dimensioning fore 3-4ohm would make sense
Last edited:
Hmm, and I dont know how to read curves like in post3
1.9amp in 4ohm before switching still means it doesnt go OFF still
But at how much output
The same 50watt
Which would be fine
But usually I expect x2 output into 4ohm
As I understand it, if it makes more than the 50watt into 4ohm, it goes OFF
And I ask, how do you rate bias/switching when forced into OFF/switching
Is it hard switching, or soft switching ?
How does this amp sound when forced into switching ?
But ok, a cool working 50watt that works like classA, and remains so into 4ohm seems quite an achievement
But wheres the hatch ?
Usually nothing is for free ?
Or is it just that it doesnt double output into half load, within the specs, like other classAB ?
And still better than loosing half the output into half load
Last edited:
That was 1.69ma, not Amps. That is maintaining the output at 28V peak. There is always some loss in the output stage when more current is delivered to the load. That means this was about 98W at 4 ohm.
Just for fun, I checked at 2 ohm (196W output). The outputs still have not shut off, but at 4uA, they are close.
Just for fun, I checked at 2 ohm (196W output). The outputs still have not shut off, but at 4uA, they are close.
and learned a bit too- Status
- Not open for further replies.