Lejonklou Tundra
Moderator: Staff
-
rowlandhills
- Very active member
- Posts: 582
- Joined: 2008-01-27 19:25
- Location: York, UK
The adjustment is for each speaker output.Wolfie wrote:I'm not sure you can do that - surely if you did, both power amps would receive the same increase/decrease in power?rowlandhills wrote:So on the assumption that I can set a "+8dB" offset on my Exotik, is there any reason not to mix & match a Tundra with a 3200?
Mick
ADSM, Rega P3-24, 4200/D, aktiv Katans
MDS-I/D, 4100/D, aktiv Katans
Kiko
Sekrit DSM, 5110s
MDS-I/D, 4100/D, aktiv Katans
Kiko
Sekrit DSM, 5110s
Yes, knew someone using a balanced and unbalanced amp together, Linn suggested adjusting the pots to align the gains.lejonklou wrote:How do you do that? With the aktiv card level setting?anthony wrote:you can reduce a balanced "200" gain to 19.6 and unbalanced to 25.6
I'm asking because in my experience, there is a clear degrade in sound quality if one adjusts those levels to anything but the standard setting.
I never felt it was quite correct though.
Not sure how big a difference it would make for in wall rears and centre, probably not much.
Completely agree. I tried it when I had a 6100 and aktiv Ninkas. Linn must have supplied this option for convenience rather than quality.lejonklou wrote:How do you do that? With the aktiv card level setting?anthony wrote:you can reduce a balanced "200" gain to 19.6 and unbalanced to 25.6
I'm asking because in my experience, there is a clear degrade in sound quality if one adjusts those levels to anything but the standard setting.
I don't know if it's just me, but I also found that adjusting input sensitivity on pre amps (general, not just Linn) has a negative effect. Are specific components in a pre amp chosen with input sensitivity in mind?
Cheers
Paul
I think the two effects that you mention have different origins. I can guess why they're negative, but nothing I'm able to explain in a few words. In essence, you have to optimise the circuits around some fixed points. Change the fixed points and they become less optimised.marmite wrote:I don't know if it's just me, but I also found that adjusting input sensitivity on pre amps (general, not just Linn) has a negative effect. Are specific components in a pre amp chosen with input sensitivity in mind?
Not sure if you are already referring to the aktiv cards or not, but wouldn't this effect them as well, so the defaults used to tune the circuits originally would always be preferable, even if the aktiv cards had smaller adjustments than 1db?lejonklou wrote:I think the two effects that you mention have different origins. I can guess why they're negative, but nothing I'm able to explain in a few words. In essence, you have to optimise the circuits around some fixed points. Change the fixed points and they become less optimised.
Yes, we talked about the level adjustment on power amps that is intended for use with aktiv cards. And then marmite added the input sensitivity adjustment on preamps.Charlie1 wrote:Not sure if you are already referring to the aktiv cards or not, but wouldn't this effect them as well, so the defaults used to tune the circuits originally would always be preferable, even if the aktiv cards had smaller adjustments than 1db?
I'm not sure if one can generalise like that, Charlie. I'm thinking it most likely depends on how you do it.
As evident from Linn's original aktiv filters, where I (and others) eventually ended up fine tuning each drive unit in steps of 0.125 dB, there is certainly sound quality to be gained from this kind of adjustment. If the adjustment circuit can be made insensitive enough, it ought to be beneficial.
Thanks Fredrik. Sounds fair enough.
OR, just go passive :O) Passive seems to be gaining momentum at the moment, particularly with Klimax amps, and hopefully Tundra too, so good timing for Lejonklou.
I'm not anti-aktiv, but it took a lot of work optimizing my aktiv Ninkas before they matched passive setup in terms of overall cohesiveness. At first I thought it was just an inherent difference between them, but forum members urged me to persist.
I know you mentioned it before, but there is more to get wrong with aktiv. Even 'not quite right' is all it takes and I only had a single amp to worry about - the thought of aktiv 242s with Akurates really doesn't appeal.
OR, just go passive :O) Passive seems to be gaining momentum at the moment, particularly with Klimax amps, and hopefully Tundra too, so good timing for Lejonklou.
I'm not anti-aktiv, but it took a lot of work optimizing my aktiv Ninkas before they matched passive setup in terms of overall cohesiveness. At first I thought it was just an inherent difference between them, but forum members urged me to persist.
I know you mentioned it before, but there is more to get wrong with aktiv. Even 'not quite right' is all it takes and I only had a single amp to worry about - the thought of aktiv 242s with Akurates really doesn't appeal.
Sorry, didn't mean to hijack the thread.lejonklou wrote: Yes, we talked about the level adjustment on power amps that is intended for use with aktiv cards. And then marmite added the input sensitivity adjustment on preamps.
Charlie, I understand where you are coming from regarding Aktiv use. I never felt the Ninkas were quite as cohesive in aktiv mode, although there were other benefits. The funny thing is I never even compared aktiv/passive before buying the aktiv cards, I just assumed (believed) it was the next step.
As much as I like my active ATC's, I'm intrigued by the thought of hearing them passive on the end of a Tundra. (Back on topic at last)
Paul
I've intended to tell you about two major breakthroughs in the development of Tundra and now is a good time to start. I'm basically stuck waiting for all the parts to arrive, and on monday I'm travelling to watch over the finishing of the enclosures (black powder coating). I can't afford the 50 first enclosures to be coated wrong (the exterior is all black, but all mating surfaces need to be metal against metal for the enclosure to become completely sealed), so I thought it best to be there myself.
Part one: Signal preservation
During 2010, I started working on a power amp. I had done a lot of reading and intended to try a couple of designs to see if any of them stood out in terms of delivering music. This is a tricky stage, as one has to work on a design for long enough to reach a reasonable understanding of it, but at the same time not wasting a lot of time and money on something that doesn't show any real promise.
For some reason, my mind was stuck with a vision of how a power amp needs to deliver voltage and current. In other words: Power! Nearly all the audio gurus that have put their ideas in print emphasise the power issue of power amps (the name implies it, does it not?). Indeed the power delivery is a bit tricky and a major source of distortion. But then again, loudspeakers are an even greater source of distortion. This is the typical hierarchy, based on the total harmonic distortion figures: Speakers are most troublesome, power amp come second, preamps are close to perfect and sources - at least digital ones - have been virtually perfect for decades. Many gurus swallow this bait, but those of us who judge by musical performance instead of distortion figures have realised that it's the other way around: The closer to the source, the larger the impact of quality on our musical satisfaction. We'd rather listen to Dylan playing the harmonica in a noisy bathroom, than to a decent interpreter of Dylan in a quiet concert hall.
Late in 2010, this knowledge finally broke through my concept of a how a power amp should work. Although I'd understood it for about 20 years, I somehow didn't see the forest for all the trees. The idea that struck me was that I should be working with signal preservation instead of power delivery. This apparently simple change of viewpoint made me look at everything from a completely different angle. Instead of battling with the output stage, I put all the effort into the input and driving stage, attempting to keep the quality of the signal preserved for as long as possible in its journey through the power amplifier. I tossed all the output stage plans in the bin and decided that it would have to adapt to what preceded it, regardless of how many watts it delivered. The idea of working with signal preservation was a major breakthrough, all from a simple change of viewpoint.
Next part will be about how the design for Tundra was chosen.
Part one: Signal preservation
During 2010, I started working on a power amp. I had done a lot of reading and intended to try a couple of designs to see if any of them stood out in terms of delivering music. This is a tricky stage, as one has to work on a design for long enough to reach a reasonable understanding of it, but at the same time not wasting a lot of time and money on something that doesn't show any real promise.
For some reason, my mind was stuck with a vision of how a power amp needs to deliver voltage and current. In other words: Power! Nearly all the audio gurus that have put their ideas in print emphasise the power issue of power amps (the name implies it, does it not?). Indeed the power delivery is a bit tricky and a major source of distortion. But then again, loudspeakers are an even greater source of distortion. This is the typical hierarchy, based on the total harmonic distortion figures: Speakers are most troublesome, power amp come second, preamps are close to perfect and sources - at least digital ones - have been virtually perfect for decades. Many gurus swallow this bait, but those of us who judge by musical performance instead of distortion figures have realised that it's the other way around: The closer to the source, the larger the impact of quality on our musical satisfaction. We'd rather listen to Dylan playing the harmonica in a noisy bathroom, than to a decent interpreter of Dylan in a quiet concert hall.
Late in 2010, this knowledge finally broke through my concept of a how a power amp should work. Although I'd understood it for about 20 years, I somehow didn't see the forest for all the trees. The idea that struck me was that I should be working with signal preservation instead of power delivery. This apparently simple change of viewpoint made me look at everything from a completely different angle. Instead of battling with the output stage, I put all the effort into the input and driving stage, attempting to keep the quality of the signal preserved for as long as possible in its journey through the power amplifier. I tossed all the output stage plans in the bin and decided that it would have to adapt to what preceded it, regardless of how many watts it delivered. The idea of working with signal preservation was a major breakthrough, all from a simple change of viewpoint.
Next part will be about how the design for Tundra was chosen.
Last edited by lejonklou on 2012-02-29 18:29, edited 1 time in total.
Very, very interesting. Love to hear more.
Concerning signal preservation there seems to be a tendency for power amp manufacturers to have the gain of their power amps at around 20dB.
The benefit is that the preamp works at a higher voltage output and the signal does not need so much attenuation at low listening levels.
Concerning signal preservation there seems to be a tendency for power amp manufacturers to have the gain of their power amps at around 20dB.
The benefit is that the preamp works at a higher voltage output and the signal does not need so much attenuation at low listening levels.
I don't remember exactly what I wrote, Matthias, but I suspect my comment was theoretical and regarding an alternative way of distributing the gain in a HiFi system. It's true that most of the time, we use the preamp as an attenuator. Which means that it lowers the volume of the signal from the source, and then the power amp raises it again. This isn't exactly optimal.
Zero gain is, however, too little when using the common preamps of today. A Kikkin or a Linn preamp has a maximum output level in the region of 7-10 V rms. If the power amp had zero gain, you wouldn't be able to get more than 6-12.5 w in 8 ohms or 12-25 W in 4 ohms. To make a power amp useable with these preamps, it needs to have some gain.
When it comes to Tundra, a gain of 11 times was the best sounding and most stable alternative for the circuit. Luckily, it's also a practical figure that makes it useable with all sources and preamps.
Zero gain is, however, too little when using the common preamps of today. A Kikkin or a Linn preamp has a maximum output level in the region of 7-10 V rms. If the power amp had zero gain, you wouldn't be able to get more than 6-12.5 w in 8 ohms or 12-25 W in 4 ohms. To make a power amp useable with these preamps, it needs to have some gain.
When it comes to Tundra, a gain of 11 times was the best sounding and most stable alternative for the circuit. Luckily, it's also a practical figure that makes it useable with all sources and preamps.
Fredrik do you have an exact date for the release of Tundra now?lejonklou wrote:I don't remember exactly what I wrote, Matthias, but I suspect my comment was theoretical and regarding an alternative way of distributing the gain in a HiFi system. It's true that most of the time, we use the preamp as an attenuator. Which means that it lowers the volume of the signal from the source, and then the power amp raises it again. This isn't exactly optimal.
Zero gain is, however, too little when using the common preamps of today. A Kikkin or a Linn preamp has a maximum output level in the region of 7-10 V rms. If the power amp had zero gain, you wouldn't be able to get more than 6-12.5 w in 8 ohms or 12-25 W in 4 ohms. To make a power amp useable with these preamps, it needs to have some gain.
When it comes to Tundra, a gain of 11 times was the best sounding and most stable alternative for the circuit. Luckily, it's also a practical figure that makes it useable with all sources and preamps.
Mike
Part two: The circuit
One day just before the Christmas holidays in 2010, I had a conversation with my friend Andreas Wahlberg, MD of Tunerlux AB. Andreas is not only a brilliant engineer, he's a real wizard in electronics and has filed a bunch of patents. Many times he's cracked the nut open for me when I've been stuck with a problem. And today was one of those days.
We were discussing the concept of "current dumping". Current dumping is a power amplifier topology, originally patented by Quad in 1975 and still used today in different shapes by various audio companies. One of them is Linn Products and the circuit they call Chakra. I had been working on my own variation of current dumping, which I felt suited my recently adopted viewpoint of signal preservation. I presented my ideas to Andreas. He wasn't impressed. He pointed out various compromises and imperfections and ended by stating that the concept of current dumping was inherently flawed. From my many arguments with him, I was expecting this. As I was getting ready for my "Ok, but listen...", he suddenly threw a quickly drawn sketch at me. "This is better", he said. "Tried it years ago and I haven't seen it in any commercial amp, ever."
I went quiet and studied the sketch for some time. This one actually suited my signal preservation viewpoint better than my version of current dumping. No, it was more than that: It was practically the incarnation of the concept of signal preservation! And it appeared to be a perfect match for the input stage I had already worked on for a good while. I knew I had to try it.
In January 2011, the first crude prototype was ready. It was one channel only and showed promise from the very first note it played, hooked up to an experimental speaker in my lab. There was just something right about it. When I later brought it home, after weeks of fine tuning it, my wife named the project Fling (as mentioned on page 2 in this thread).
I soon became aware of the high sensitivity of the circuit. On the positive side, this brought vibrant life into the music in a way I hadn't heard before, from a prototype still far from being perfected. On the challenge side, it's very high speed and needs a careful layout and perfect matching of components not to go haywire. Months of trying different adjustments followed and I gradually narrowed down the parameters and conditions in which it worked best. Apart from the Kikkin-sized ten Watt Fling, I made a version called Fling Large. It was powered by a big conventional transformer with massive regulation. It sounded very good. I could have settled for that design and released it in the summer of 2011.
One thing that annoyed me, however, was that mains quality affected the performance of the amp. Between different locations, it could go from great to good only due to the mains. While this could perhaps have been regarded as tolerable, it just didn't feel right. I tried to isolate it better, but nothing seemed to work. In the back of my mind, I knew that the optimal solution would be to use switch mode power supplies (SMPS). I had been working with them before and I was convinced they could make Fling perform even better, while at the same time remove the mains related variations.
But just like an amplifier circuit, a SMPS is an active device with hundreds of parameters, which means they are expensive to develop and can be tuned endlessly. After a period of agony over having to choose between Fling-with-variations and diving into a new period of development work, I settled for the latter. After a couple of months I arrived at a 100 W SMPS that was much better than my conventional designs. When I let two of them feed the stereo amplifier board, Tundra was born.
One day just before the Christmas holidays in 2010, I had a conversation with my friend Andreas Wahlberg, MD of Tunerlux AB. Andreas is not only a brilliant engineer, he's a real wizard in electronics and has filed a bunch of patents. Many times he's cracked the nut open for me when I've been stuck with a problem. And today was one of those days.
We were discussing the concept of "current dumping". Current dumping is a power amplifier topology, originally patented by Quad in 1975 and still used today in different shapes by various audio companies. One of them is Linn Products and the circuit they call Chakra. I had been working on my own variation of current dumping, which I felt suited my recently adopted viewpoint of signal preservation. I presented my ideas to Andreas. He wasn't impressed. He pointed out various compromises and imperfections and ended by stating that the concept of current dumping was inherently flawed. From my many arguments with him, I was expecting this. As I was getting ready for my "Ok, but listen...", he suddenly threw a quickly drawn sketch at me. "This is better", he said. "Tried it years ago and I haven't seen it in any commercial amp, ever."
I went quiet and studied the sketch for some time. This one actually suited my signal preservation viewpoint better than my version of current dumping. No, it was more than that: It was practically the incarnation of the concept of signal preservation! And it appeared to be a perfect match for the input stage I had already worked on for a good while. I knew I had to try it.
In January 2011, the first crude prototype was ready. It was one channel only and showed promise from the very first note it played, hooked up to an experimental speaker in my lab. There was just something right about it. When I later brought it home, after weeks of fine tuning it, my wife named the project Fling (as mentioned on page 2 in this thread).
I soon became aware of the high sensitivity of the circuit. On the positive side, this brought vibrant life into the music in a way I hadn't heard before, from a prototype still far from being perfected. On the challenge side, it's very high speed and needs a careful layout and perfect matching of components not to go haywire. Months of trying different adjustments followed and I gradually narrowed down the parameters and conditions in which it worked best. Apart from the Kikkin-sized ten Watt Fling, I made a version called Fling Large. It was powered by a big conventional transformer with massive regulation. It sounded very good. I could have settled for that design and released it in the summer of 2011.
One thing that annoyed me, however, was that mains quality affected the performance of the amp. Between different locations, it could go from great to good only due to the mains. While this could perhaps have been regarded as tolerable, it just didn't feel right. I tried to isolate it better, but nothing seemed to work. In the back of my mind, I knew that the optimal solution would be to use switch mode power supplies (SMPS). I had been working with them before and I was convinced they could make Fling perform even better, while at the same time remove the mains related variations.
But just like an amplifier circuit, a SMPS is an active device with hundreds of parameters, which means they are expensive to develop and can be tuned endlessly. After a period of agony over having to choose between Fling-with-variations and diving into a new period of development work, I settled for the latter. After a couple of months I arrived at a 100 W SMPS that was much better than my conventional designs. When I let two of them feed the stereo amplifier board, Tundra was born.
Tundra will be launched the 21st of March at Tonläget, Gothenburg.
Please visit http://tonlaget.com/afton-lejonklou.html to make preliminary booking. This is just to let Tonläget know approximately how many are coming.
Please visit http://tonlaget.com/afton-lejonklou.html to make preliminary booking. This is just to let Tonläget know approximately how many are coming.