Lejonklou Tundra
Moderator: Staff
Sounds great, Fredrik! And it is good to see the photo of it. It is a bit reminiscent of the earlier Exposure gear but that's no problem. Will there be a big LEJONKLOU logo centered between the two LEDs?
Interesting story on the Infinity speakers. Many of their models had a reputation as being difficult to drive as they had a nasty impedance curve from having a lot of drivers. Part of the reputation also came from stupid marketing. In the US Infinity was trying to become a bigger player in the Hi-Fi market so they went with one of the bigger chains at the time - Tech Hi-Fi. Tech was one of those places with a lot of equipment on the shelves and poor demonstration facilities and sales were based mostly on what had the best spifs. The go to electronics line there was Kenwood and their big "100+ watt" receivers. Well it turned out that the Kenwood receivers could not deal with low load impedances and would literally blow up when driving Infinity speakers - especially the bigger ones. Of course, due to the big spifs, that didn't keep Tech Hi-Fi from selling lots of the combination. Hence the reputation of Infinity speakers being difficult to drive was cemented. (A little side note: Infinity made a basic cheap box speaker with a big woofer at the time, it might even have been an exclusive for Tech Hi-Fi. Its distinguishing characteristic, other than its bad sound, was that it had an outrageously high profit margin! The real interesting thing was the model number: the POS 1. And, yes, it did mean "Piece Of S**t" although Infinity would never admit it to the general public!)
Several years after the Kenwood/Infinity debacle I had a customer come into the store where I worked looking at gear. I started up a conversation and asked him what he owned. The answer was a Kenwood 9600 (if I remember correctly) and some big Infinity speakers! I said: "Wow, I'm surprised the Kenwood didn't blow up!" His answer was that it had = Three Times! After the third time it was actually sent back to Japan and Kenwood "modified" it so it wouldn't blow up anymore (although I doubt it made that piece of electronics sound any good). I have to say I was rather surprised at his resistance to realizing he had a bad match and getting something else. Although the inability of Tech Hi-Fi to supply anything that could drive Infinity speakers without it costing several times the cost of the Kenwood receivers surely played a part. Somehow the Hierarchy wasn't exactly a part of their philosophy. Just sell 'em a cheap amp and a big pair of speakers and they'll be OK!
I had actually heard somewhere that the NAIT was about 12 watts. But whatever it was it was amazing how well it drove a lot of speakers. I once took the NAIT to a new Linn dealer we were training on LP12 setup (when I worked for the US Distributor) and told the two partners at the store that they should give it a listen. We did a direct comparison with about $3500 worth of Conrad-Johnson gear (this was in early 1984 when an LP12/Ittok was around $1500) and they were amazed that the NAIT totally blew the CJ stuff away (it sold for $395)!
On this same ability to drive subject when I bought my first pair of Isobariks I was driving them with a Marantz 8B - a 35Watt a channel tube amp - which actually handled them OK. (It was the amp I owned at the time and I didn't know any better.) After getting the Isobariks my interest in the Naim amps was piqued as I had read good things about them. Our store didn't carry Naim but I called the Linn/Naim distributor and asked if I could evaluate the new 42/110 (which they told me could drive Isobariks (they didn't know any better either). They were happy to send me a set as my evaluating the Isobariks resulted in the store carrying them and selling several pair. When I received the 42/110 I was amazed at how musical it was. Despite being rated at only 55 watts per channel into 4 ohms (hence the NAP110 designation) it easily outplayed every other amp in the store on the Isobariks including the big multi-hundred watt models from Threshold and Bryston. I thought the 110 was doing a fine job as did one of the other guys I worked with who bought the same combination. That was until later that year when we met Julian at a Hi-Fi show and my friend made the mistake of telling him how wonderful the 42/110 was with his Isobariks to which Julian replied "A 110 can't drive Isobariks." My friend (possibly not so cleverly at the time) said "But it sounds really good!" Julian: "A 110 can't drive Isobariks!" After a little more conversation Julian made it obvious that he felt the 250 was "the only decent amp in the world, period." and was what was needed for Isobariks. Once I got back home I rung up the distributor again and asked to evaluate a 250. They gladly obliged and I immediately heard for myself that a 110 couldn't drive Isobariks. I lived for many happy years listing to my Isobariks driven by a 250, and eventually a pair of 135s, until Linn came along with a preamp/amp combination that was easily more musical and a fair bit less expensive to boot.
The point I am really trying to make with that story is that the 110, with its limited power, was way better at driving Isobariks than any of the big North American muscle amps of the day and was only shown up when the 250 made it obvious what an Isobarik can really sound like. Interestingly, the Komri also has a reputation as being difficult to drive so, if as reported, the Tundra can handle it effectively in a largish room it is likely to be adequate for a lot of different speakers and rooms. As Fredrik says "Try it!" as this is really the only way to know if it works in your situation. I am certainly looking forward to being able to try it even though I will have to use a different set of speakers to do it. (I do have access to a few other speakers!)
Interesting story on the Infinity speakers. Many of their models had a reputation as being difficult to drive as they had a nasty impedance curve from having a lot of drivers. Part of the reputation also came from stupid marketing. In the US Infinity was trying to become a bigger player in the Hi-Fi market so they went with one of the bigger chains at the time - Tech Hi-Fi. Tech was one of those places with a lot of equipment on the shelves and poor demonstration facilities and sales were based mostly on what had the best spifs. The go to electronics line there was Kenwood and their big "100+ watt" receivers. Well it turned out that the Kenwood receivers could not deal with low load impedances and would literally blow up when driving Infinity speakers - especially the bigger ones. Of course, due to the big spifs, that didn't keep Tech Hi-Fi from selling lots of the combination. Hence the reputation of Infinity speakers being difficult to drive was cemented. (A little side note: Infinity made a basic cheap box speaker with a big woofer at the time, it might even have been an exclusive for Tech Hi-Fi. Its distinguishing characteristic, other than its bad sound, was that it had an outrageously high profit margin! The real interesting thing was the model number: the POS 1. And, yes, it did mean "Piece Of S**t" although Infinity would never admit it to the general public!)
Several years after the Kenwood/Infinity debacle I had a customer come into the store where I worked looking at gear. I started up a conversation and asked him what he owned. The answer was a Kenwood 9600 (if I remember correctly) and some big Infinity speakers! I said: "Wow, I'm surprised the Kenwood didn't blow up!" His answer was that it had = Three Times! After the third time it was actually sent back to Japan and Kenwood "modified" it so it wouldn't blow up anymore (although I doubt it made that piece of electronics sound any good). I have to say I was rather surprised at his resistance to realizing he had a bad match and getting something else. Although the inability of Tech Hi-Fi to supply anything that could drive Infinity speakers without it costing several times the cost of the Kenwood receivers surely played a part. Somehow the Hierarchy wasn't exactly a part of their philosophy. Just sell 'em a cheap amp and a big pair of speakers and they'll be OK!
I had actually heard somewhere that the NAIT was about 12 watts. But whatever it was it was amazing how well it drove a lot of speakers. I once took the NAIT to a new Linn dealer we were training on LP12 setup (when I worked for the US Distributor) and told the two partners at the store that they should give it a listen. We did a direct comparison with about $3500 worth of Conrad-Johnson gear (this was in early 1984 when an LP12/Ittok was around $1500) and they were amazed that the NAIT totally blew the CJ stuff away (it sold for $395)!
On this same ability to drive subject when I bought my first pair of Isobariks I was driving them with a Marantz 8B - a 35Watt a channel tube amp - which actually handled them OK. (It was the amp I owned at the time and I didn't know any better.) After getting the Isobariks my interest in the Naim amps was piqued as I had read good things about them. Our store didn't carry Naim but I called the Linn/Naim distributor and asked if I could evaluate the new 42/110 (which they told me could drive Isobariks (they didn't know any better either). They were happy to send me a set as my evaluating the Isobariks resulted in the store carrying them and selling several pair. When I received the 42/110 I was amazed at how musical it was. Despite being rated at only 55 watts per channel into 4 ohms (hence the NAP110 designation) it easily outplayed every other amp in the store on the Isobariks including the big multi-hundred watt models from Threshold and Bryston. I thought the 110 was doing a fine job as did one of the other guys I worked with who bought the same combination. That was until later that year when we met Julian at a Hi-Fi show and my friend made the mistake of telling him how wonderful the 42/110 was with his Isobariks to which Julian replied "A 110 can't drive Isobariks." My friend (possibly not so cleverly at the time) said "But it sounds really good!" Julian: "A 110 can't drive Isobariks!" After a little more conversation Julian made it obvious that he felt the 250 was "the only decent amp in the world, period." and was what was needed for Isobariks. Once I got back home I rung up the distributor again and asked to evaluate a 250. They gladly obliged and I immediately heard for myself that a 110 couldn't drive Isobariks. I lived for many happy years listing to my Isobariks driven by a 250, and eventually a pair of 135s, until Linn came along with a preamp/amp combination that was easily more musical and a fair bit less expensive to boot.
The point I am really trying to make with that story is that the 110, with its limited power, was way better at driving Isobariks than any of the big North American muscle amps of the day and was only shown up when the 250 made it obvious what an Isobarik can really sound like. Interestingly, the Komri also has a reputation as being difficult to drive so, if as reported, the Tundra can handle it effectively in a largish room it is likely to be adequate for a lot of different speakers and rooms. As Fredrik says "Try it!" as this is really the only way to know if it works in your situation. I am certainly looking forward to being able to try it even though I will have to use a different set of speakers to do it. (I do have access to a few other speakers!)
I thought it might be a good idea to bring some perspective into this whole "power" question. When I first got into Hi-Fi in the early 70s I read everything I could get my hands on about Hi-Fi. I learned what every specification meant and how to evaluate different products by specifications. Once I got more involved, and especially after I joined the industry in 1978, I found out that specs didn't mean anything as indicated by a couple of the stories above. But not everybody has had that experience yet so I thought I'd try to explain.
First off I'll use a tired automotive analogy and mention that horsepower has nothing to do with how fast a car goes. Why? because it is a single number whereas in reality it varies considerably with the engine's rotational speed for starters. But also because the quality of the transmission, the final gear ratio, the size and quality of the tires, the available torque, the weight of the car, etc., etc. all have an effect on how fast the car can actually go.
The same is true of wattage from an amp. Continuous power is tested into a resistive 8 Ohm load. This is the automotive equivalent of testing speed downhill with a tail wind! An actual speaker is anything but a resistive 8 Ohm load as can be easily seen in any of the speaker test graphs of impedance in Stereophile and other publications. A loudspeaker's impedance generally varies from about one half the rated impedance (which is one part that causes amps difficulties) to sometimes five to ten times the impedance at resonance points (woofer and port resonance being two common ones). But that is just the beginning. A passive speaker has a crossover that uses coils to roll off the top end of bass and mid drivers creating inductance that can be quite substantial. It also has caps to roll off the low end of tweeters, mids, etc. adding significant capacitance to the mix. And there are various resistors to balance out the levels of the different drivers. All these parts have to be heavy duty to handle the power involved but they, like all parts, change their characteristics somewhat depending on their temperature which is constantly varying (one of the reasons that active crossovers can work better). This creates a complex and constantly varying load for the amp and how well the amp deals with these variances has a great deal to do with how powerful it is in real life.
But also there is back-EMF. When a driver like a woofer responds to an amp signal and moves inertia causes it to want to keep moving. A good amp will help to damp this movement but there will still be some small movement that is beyond what the amp's signal required. This turns the driver into a generator that sends a voltage back to the amplifier. So the amp also has to deal with these constant small voltages well to minimize their effect on the music. So the job of a power amp is anything but simple and how well it deals with all these factors has much more to do with how powerful it sounds, and actually IS when hooked to a pair of speakers, that anything a standard measured power rating could tell you.
But that is only half the story (hold the groans - you know I can be long-winded). ;-) The other question is: How much power do you really need to play loud? This turns out to be quite interesting as the CONTINUOUS power you need for music listening is quite low. To figure this out we need to look at speaker efficiencies and understand how the ear works. The ear hears logarithmically. What this means is that twice the power (which is 3dB) does not sound twice as loud to the ear. For you to perceive it as twice as loud actually requires ten times the power (10dB). But the ear is also a sensitive device and too much volume can permanently damage it - as many rock stars can attest!
On to speaker efficiencies. Looking over the current and past Linn speaker catalog shows that the efficiency has varied from 85dB to 92dB (although I seem to recall the original Kan being 84dB but I can't find specs on it anywhere). Most of them are somewhere in the middle of that range:
Komri 87dB
Klimax/Artikulat series 87dB
Akurate series 87dB
Majik Series 88dB (which I find amusing)
Isobarik 86dB
Espek/Ninka 90dB
So pretty reasonable overall and not a lot of difference from one to the other. But what does this spec mean? It means that a single speaker fed one watt of power (normally measured at 1KHz) will put out the rated level (86dB for an Isobarik) at a distance of one meter from the front of the speaker. If I have it right this also means that two watts each into a pair of Isobariks will put out 86dB at two meters from the front of the speakers - maybe a bit close but not an unreasonable listening distance.
So how loud is 86dB? I remember many years ago I heard that 95dB continuous would make you deaf but nobody said how long you had to be exposed to it. I also heard that 130dB was the threshold of pain. In doing some research on it I found, of course, that it wasn't that simple as hearing varies from person to person so results can vary. But I also found those original figures do have some meaning. The threshold of pain is considered to be in the range of 120dB to 140dB depending on the person. However, and more importantly, 85dB continuous over an 8 hour period will cause PERMANENT hearing damage! Hearing protection is recommended for those working long periods with equipment that puts out a sound level over about 80dB.
In other words 2 watts each into a pair of Isobariks continuously, listened to at a distance of 2 meters will cause permanent hearing loss after 8 hours of exposure! What about that 95dB figure mentioned earlier? It turns out that if you double the sound level (3dB) you halve the time so at a continuous level of 94dB you would get permanent hearing damage in ONE hour! What is what power wise? Double the power for each 3dB: 89dB = 4 watts each, 92dB = 8 watts each, 95dB = 16 watts each. So 16 watts continuous into each of a pair of Isobariks at a two meter distance will permanently damage your hearing in under one hour! (If you are twice as far away you need four times the power: 4 meters, 64 watts for the same hearing damage.)
This is why a 12 watt NAIT was enough to drive inefficient speakers like Kans to good levels. But the real reason is that is is peak power, and even more importantly peak power that maintains the ability to drive a speaker really well when it is at its most difficult to drive, that is what is needed to handle music properly. Some more numbers: 10dB requires 10 times the power so with our Isobarik example 105dB would need 160 watts and 115dB (coincidentally the rated continuous output of my ATCs) would need 1600 watts! Incidentally that sound level of 115dB has a recommended maximum exposure time of 30 seconds and 120dB causes immediate hearing damage.
So Fredrik isn't just pulling a publicity stunt when he says that the industry standard continuous power ratings are meaningless. The only meaningful thing is how well an amp deals with a real loudspeaker and how well it handles the peak demands of the music into your actual speakers at the maximum listening levels you would listen to. Like pretty much anything else in Hi-Fi the only way to really tell is to listen in your own system, although I realize trying to do that with four Tundras might not be the easiest thing. But I would guess that, if a Tundra can drive Komris passively, they should be able to drive Keltiks Aktiv.
First off I'll use a tired automotive analogy and mention that horsepower has nothing to do with how fast a car goes. Why? because it is a single number whereas in reality it varies considerably with the engine's rotational speed for starters. But also because the quality of the transmission, the final gear ratio, the size and quality of the tires, the available torque, the weight of the car, etc., etc. all have an effect on how fast the car can actually go.
The same is true of wattage from an amp. Continuous power is tested into a resistive 8 Ohm load. This is the automotive equivalent of testing speed downhill with a tail wind! An actual speaker is anything but a resistive 8 Ohm load as can be easily seen in any of the speaker test graphs of impedance in Stereophile and other publications. A loudspeaker's impedance generally varies from about one half the rated impedance (which is one part that causes amps difficulties) to sometimes five to ten times the impedance at resonance points (woofer and port resonance being two common ones). But that is just the beginning. A passive speaker has a crossover that uses coils to roll off the top end of bass and mid drivers creating inductance that can be quite substantial. It also has caps to roll off the low end of tweeters, mids, etc. adding significant capacitance to the mix. And there are various resistors to balance out the levels of the different drivers. All these parts have to be heavy duty to handle the power involved but they, like all parts, change their characteristics somewhat depending on their temperature which is constantly varying (one of the reasons that active crossovers can work better). This creates a complex and constantly varying load for the amp and how well the amp deals with these variances has a great deal to do with how powerful it is in real life.
But also there is back-EMF. When a driver like a woofer responds to an amp signal and moves inertia causes it to want to keep moving. A good amp will help to damp this movement but there will still be some small movement that is beyond what the amp's signal required. This turns the driver into a generator that sends a voltage back to the amplifier. So the amp also has to deal with these constant small voltages well to minimize their effect on the music. So the job of a power amp is anything but simple and how well it deals with all these factors has much more to do with how powerful it sounds, and actually IS when hooked to a pair of speakers, that anything a standard measured power rating could tell you.
But that is only half the story (hold the groans - you know I can be long-winded). ;-) The other question is: How much power do you really need to play loud? This turns out to be quite interesting as the CONTINUOUS power you need for music listening is quite low. To figure this out we need to look at speaker efficiencies and understand how the ear works. The ear hears logarithmically. What this means is that twice the power (which is 3dB) does not sound twice as loud to the ear. For you to perceive it as twice as loud actually requires ten times the power (10dB). But the ear is also a sensitive device and too much volume can permanently damage it - as many rock stars can attest!
On to speaker efficiencies. Looking over the current and past Linn speaker catalog shows that the efficiency has varied from 85dB to 92dB (although I seem to recall the original Kan being 84dB but I can't find specs on it anywhere). Most of them are somewhere in the middle of that range:
Komri 87dB
Klimax/Artikulat series 87dB
Akurate series 87dB
Majik Series 88dB (which I find amusing)
Isobarik 86dB
Espek/Ninka 90dB
So pretty reasonable overall and not a lot of difference from one to the other. But what does this spec mean? It means that a single speaker fed one watt of power (normally measured at 1KHz) will put out the rated level (86dB for an Isobarik) at a distance of one meter from the front of the speaker. If I have it right this also means that two watts each into a pair of Isobariks will put out 86dB at two meters from the front of the speakers - maybe a bit close but not an unreasonable listening distance.
So how loud is 86dB? I remember many years ago I heard that 95dB continuous would make you deaf but nobody said how long you had to be exposed to it. I also heard that 130dB was the threshold of pain. In doing some research on it I found, of course, that it wasn't that simple as hearing varies from person to person so results can vary. But I also found those original figures do have some meaning. The threshold of pain is considered to be in the range of 120dB to 140dB depending on the person. However, and more importantly, 85dB continuous over an 8 hour period will cause PERMANENT hearing damage! Hearing protection is recommended for those working long periods with equipment that puts out a sound level over about 80dB.
In other words 2 watts each into a pair of Isobariks continuously, listened to at a distance of 2 meters will cause permanent hearing loss after 8 hours of exposure! What about that 95dB figure mentioned earlier? It turns out that if you double the sound level (3dB) you halve the time so at a continuous level of 94dB you would get permanent hearing damage in ONE hour! What is what power wise? Double the power for each 3dB: 89dB = 4 watts each, 92dB = 8 watts each, 95dB = 16 watts each. So 16 watts continuous into each of a pair of Isobariks at a two meter distance will permanently damage your hearing in under one hour! (If you are twice as far away you need four times the power: 4 meters, 64 watts for the same hearing damage.)
This is why a 12 watt NAIT was enough to drive inefficient speakers like Kans to good levels. But the real reason is that is is peak power, and even more importantly peak power that maintains the ability to drive a speaker really well when it is at its most difficult to drive, that is what is needed to handle music properly. Some more numbers: 10dB requires 10 times the power so with our Isobarik example 105dB would need 160 watts and 115dB (coincidentally the rated continuous output of my ATCs) would need 1600 watts! Incidentally that sound level of 115dB has a recommended maximum exposure time of 30 seconds and 120dB causes immediate hearing damage.
So Fredrik isn't just pulling a publicity stunt when he says that the industry standard continuous power ratings are meaningless. The only meaningful thing is how well an amp deals with a real loudspeaker and how well it handles the peak demands of the music into your actual speakers at the maximum listening levels you would listen to. Like pretty much anything else in Hi-Fi the only way to really tell is to listen in your own system, although I realize trying to do that with four Tundras might not be the easiest thing. But I would guess that, if a Tundra can drive Komris passively, they should be able to drive Keltiks Aktiv.
Thanks for the ideas, Marco! I will get back to this in four to six months, in another thread. :o)mrco99 wrote:That's a great perspective. Will it be a Kinki, Kikkin with multiple inputs and headphone amp combined? That would make an awesome combination. May i also suggest a name already?
Hattrik.
I know that Linn Keltiks have been at the centre of some controversy lately, mainly on the Swedish Linn forum. I haven't tried Tundra on them, so I guess it comes down to "Try it", and I guess the outcome will be a matter of personal opinion too. That is; how loud you want them to play. Keltiks do have some disco/dancehall/headbang qualities, so I can understand they have a following for those about to rock. But ultimately it's the room you put them in that determines how much power you will need. Komris in a rather large room sounded great with one Tundra, but I can't comment on Keltiks until I've tried them.SaltyDog wrote:Anyone have a thought on Keltiks?
Thomas: Thanks for an excellent post! Indeed it's much more complicated than a single number (ohms). Even the so called "power cube", that uses capacitive and inductive loads, predominantly used in the scandinavian magazine HiFi&Musik, is terribly inadequate when it comes to predicting how a power amplifier will actually handle a loudspeaker.
Wrestling with issues of enclosure finish at the moment, which threatens to delay the introduction. In the meantime, I will soon return with the story of how I got on the right track in the development of Tundra.
Still, getting in right is the most important thing as I'm sure you're going to acheive. Are there many options, or just trying to find someone that does a good enough job. Would SME be overkill?lejonklou wrote:Wrestling with issues of enclosure finish at the moment, which threatens to delay the introduction.
Hi Fredrik,lejonklou wrote:I know that Linn Keltiks have been at the centre of some controversy lately, mainly on the Swedish Linn forum. I haven't tried Tundra on them, so I guess it comes down to "Try it", and I guess the outcome will be a matter of personal opinion too. That is; how loud you want them to play. Keltiks do have some disco/dancehall/headbang qualities, so I can understand they have a following for those about to rock. But ultimately it's the room you put them in that determines how much power you will need. Komris in a rather large room sounded great with one Tundra, but I can't comment on Keltiks until I've tried them.SaltyDog wrote:Anyone have a thought on Keltiks?
When you plan a trip to The Netherlands and don't mind bringing 3 or 4 Tundras, you are more than welcome to try them in my system with Keltiks. In fact I'd be honoured...;-)
Cheers,
Marco
There are many options, but so far none are good enough. The best finish so far looks great and is not unreasonable in cost, but they can't guarantee a perfectly dust free environment. So on the sample they sent me, there was a dust mark (tiny hair) right above the logo. It looks fantastic at a distance, but when you get close or run your fingers over it, it's no longer fantastic.Charlie1 wrote:Are there many options, or just trying to find someone that does a good enough job. Would SME be overkill?
I know that SME are famous for their finishing jobs. Transporting the enclosures back and forth will of course become a bit expensive, but I'll ask them what they charge for a job like this.
The honour would be all mine! I'm not sure I will be travelling with 4 Tundras, but you never know. :)mrco99 wrote:When you plan a trip to The Netherlands and don't mind bringing 3 or 4 Tundras, you are more than welcome to try them in my system with Keltiks. In fact I'd be honoured...;-)
Tundra power cord
Fredrik,
does the Tundra have a fixed or removable power cord ?
does the Tundra have a fixed or removable power cord ?
Hi John and welcome to the forum!
I promised an introduction of Tundra in February. It will most likely take place at Tonläget in Gothenburg. It's almost becoming a tradition that I launch new products there. And after the lauch, I intend to travel around to other major cities with it.
As I play with open cards, I must confess the schedule is tight and I am starting to fear that I won't make it in time. There is currently a third prototype being built, with minor changes to the enclosure, which will be sent to a new finishing company (the fifth so far). There's also been some changes to the circuit board layout and an addition of two more layers, to make it radiate less noise. The Tundra audio circuit remains the same, however. I've tried to improve it a bit further, but it appears to already be at the peak of its potential. I spent the entire 2011 perfecting it, so I'm not surprised that it's difficult to tune it any further. My intention is to launch an amp that won't be possible to upgrade for a very long time.
I'm so much looking forward to let Tundra play you some music!
Stay tuned and I will announce when and where the launch will take place.
I promised an introduction of Tundra in February. It will most likely take place at Tonläget in Gothenburg. It's almost becoming a tradition that I launch new products there. And after the lauch, I intend to travel around to other major cities with it.
As I play with open cards, I must confess the schedule is tight and I am starting to fear that I won't make it in time. There is currently a third prototype being built, with minor changes to the enclosure, which will be sent to a new finishing company (the fifth so far). There's also been some changes to the circuit board layout and an addition of two more layers, to make it radiate less noise. The Tundra audio circuit remains the same, however. I've tried to improve it a bit further, but it appears to already be at the peak of its potential. I spent the entire 2011 perfecting it, so I'm not surprised that it's difficult to tune it any further. My intention is to launch an amp that won't be possible to upgrade for a very long time.
I'm so much looking forward to let Tundra play you some music!
Stay tuned and I will announce when and where the launch will take place.
Fredrik,lejonklou wrote:The Tundra audio circuit remains the same, however. I've tried to improve it a bit further, but it appears to already be at the peak of its potential. I spent the entire 2011 perfecting it, so I'm not surprised that it's difficult to tune it any further.
can you use the Tundra audio circuit for the upcoming matching preamp or do you have to start from scratch ?
matthias
A preamp is very different in construction from a power amp, so I can't use much from the audio circuit.matthias wrote:can you use the Tundra audio circuit for the upcoming matching preamp or do you have to start from scratch ?
However, I've learned a lot in the process, so it certainly won't be a start from scratch. I've got one book full of notes from the work on Tundra and one book from Kikkin.
Stability with age and environment? (And congrats, too!)
Hello Mr. Lejonklou!lejonklou wrote:....I settled for the current circuit a year ago. Back then it was a tiny amp that could be housed in a Kikkin enclosure and the name of the project was 'Fling'....
I could have released Fling in spring 2011, but I ran into a problem. Variations in quality turned out to be big. In other words: The second prototype I made sounded different from the first, even though it was made the same way. And when I studied the phenomenon more carefully, I realised that the circuit was indeed terribly sensitive. Not just to the fine nuances of the music it amplified, but also to all component variations.
...I decided to make a more powerful Fling and just see how far I could push the quality, without any practical considerations on how it could ever be manufactured. I bought piles of components and manually measured and selected them in ridiculously small variations. I used twin switch mode power supplies that isolated it much better from variations in the mains power. And I just tuned the little monster in any way that I could think of; currents, voltages, temperatures, layout, cabling, directions, torques, mechanical issues and damping....
...this was not an amp that could be manufactured and sold. It was an extreme experiment, which could later result in something simpler and possible to manufacture....
So the last month has been spent on sourcing and selecting components in test rigs built for the purpose. I decided to do this myself and ship perfectly matched components directly to the circuit board manufacturer, to avoid any potential mistakes. Right now an assembly manual is being written, the longest and most detailed I've ever made. In January I will make sure it's being followed to the letter. There are only two parameters that I haven't been able to include in any practical way. These I will take care of myself with a final step of individual tuning of every Tundra, when it arrives from the assembler.
Great news about the upcoming Tundra; this sounds like a component I'd very much love to own someday, and congrats in advance on the upcoming product launch! =)
Now, may I please ask a sort of engineering philosophy question? In practical terms, the question is simply:
Q: How do you expect the Tundra to age, given that component parameters can drift over time? Will the Tundra be as stable as competing products? Or if not, will it be simple to restore the performance of an aging Tundra through component refurbishment or firmwear?
I'm asking this somewhat from-out-of-nowhere question because of your fascinating summary of the Tundra development story (snippets of which are quoted above), which echoes development-philosophy issues I've seen elsewhere -- and since you've succeeded with an approach that has seen many others fail, it'd be great to hear more about how you did it!
To explain: as you might remember from past years' posts, I used to work in robotics -- and in that field we too encountered the same conundrum you appear to have faced in developing the Tundra. To wit: it's possible to get decent performance out of an electronic system that's inherently forgiving of environmental and component variances; and it's similarly possible to achieve great performance with a system that's painstakingly tuned both to its specific operating environment and to the particular components it's assembled with -- but it's really, really hard to combine both performance and forgiveness!! That is, in robotics (as apparently is the case with amplifiers) awesome performance often means a design that can't be manufactured easily and/or that won't work reliably in real-world settings (which can differ considerably from the lab bench or test bay) -- and yet you appear to have overcome the conundrum!
So the underlying question is: How (in more detail!) did you do this?? It's one thing to grade your components and to give your fabricators & assemblers robust instructions (though figuring all this out must have been a staggering amount of work!) -- but you've still an apparently complex system that's highly sensitive to variance, and that will be installed in rooms situated in dry-as-a-bone Arizona as well as in damp-and-cold Scotland, and then plugged into the mains power of many different countries. How is the Tundra stable under such variances? And how have you corrected for component drift and aging? From an engineering philosophy point of you, you've done something fairly amazing here -- and I for one would love to hear more!
Anyway, such is the (rather obscure) question. =) Best wishes for the product launch(es), and I look forward to reading the first user reports from the forum gurus (e.g., ThomasOK, Charlie1, Musiclover, and others). All the very best!
- W
Re: Stability with age and environment? (And congrats, too!
That's very kind of you Ceilidh, but my experience really is very limited compared to many others on the forum. Good to hear from you and if I do get a chance to hear the Tundra (I hope so), then I'll report back.Ceilidh wrote:I look forward to reading the first user reports from the forum gurus (e.g., ThomasOK, Charlie1, Musiclover, and others).
Well, I'm definitely no music guru, but since I've been using Tundra "Prototype 2" in my own system for a few weeks, I guess it's appropriate to write a report anyway (not least because I promised Fredrik to do it while nagging about buying the thing).
My Tundra is playing in a system consisting of ADS/D, Kikkin and a pair of 109's. It has replaced the 2200/D that I've been using (and loved) for about a year.
I sold the 2200 in November, so I have not made any A/B comparisons. But since I've been playing the 2200 in the identical system for a few months, I feel that I can draw a few quite reliable conclusions.
The first impression when I connected the Tundra was that something must be wrong. It sounded extremely flat and dull, quite like the short period when I used a LK100 with my 109's. I left the system playing anyway, since I didn't have time investigate it, I had to go to bed soon. After ten minutes or so I came back to the living room, and realized how good that Dixie Chicks song was! I sat down in the sofa and stayed there for two hours (I was tired for the whole week after that).
Its very easy - although somewhat obvious by now - to conclude that a warmed up Tundra is better then 2200/D. The difference is not subtle either, it is one of those "transforming" upgrades that makes you want to listen to all your records again, even the ones you didn't really love, just to hear if there is something you missed.
One of the things I really liked about the 2200 was that I could play really loud (louder than healthy-loud) without any perceived decrease in musical quality. I have not had sufficient time to play loud with the Tundra yet, but I do feel that the maximum level with remaining quality has lowered a bit. This should probably be investigated with both amps present, especially since they have different grades of amplification.
So: If you live in a flat and rarely have dance parties - get one! But if you live in a house and your children are big enough not to wake up when you play your super-loud music every day - you might want to try it before you buy it. The question will then be if you are willing to sacrifice quality for quantity?
/Az
My Tundra is playing in a system consisting of ADS/D, Kikkin and a pair of 109's. It has replaced the 2200/D that I've been using (and loved) for about a year.
I sold the 2200 in November, so I have not made any A/B comparisons. But since I've been playing the 2200 in the identical system for a few months, I feel that I can draw a few quite reliable conclusions.
The first impression when I connected the Tundra was that something must be wrong. It sounded extremely flat and dull, quite like the short period when I used a LK100 with my 109's. I left the system playing anyway, since I didn't have time investigate it, I had to go to bed soon. After ten minutes or so I came back to the living room, and realized how good that Dixie Chicks song was! I sat down in the sofa and stayed there for two hours (I was tired for the whole week after that).
Its very easy - although somewhat obvious by now - to conclude that a warmed up Tundra is better then 2200/D. The difference is not subtle either, it is one of those "transforming" upgrades that makes you want to listen to all your records again, even the ones you didn't really love, just to hear if there is something you missed.
One of the things I really liked about the 2200 was that I could play really loud (louder than healthy-loud) without any perceived decrease in musical quality. I have not had sufficient time to play loud with the Tundra yet, but I do feel that the maximum level with remaining quality has lowered a bit. This should probably be investigated with both amps present, especially since they have different grades of amplification.
So: If you live in a flat and rarely have dance parties - get one! But if you live in a house and your children are big enough not to wake up when you play your super-loud music every day - you might want to try it before you buy it. The question will then be if you are willing to sacrifice quality for quantity?
/Az
-
rowlandhills
- Very active member
- Posts: 582
- Joined: 2008-01-27 19:25
- Location: York, UK
That's very interesting review, especially as I'd interpreted Fredrik's comments as meaning that it gets better the louder you go, so I was concerned that it would not sound as good at low to medium volumes!
Thinking of this comment in particular:
Thinking of this comment in particular:
lejonklou wrote:Hi Matthias!
Tundra has a lower maximum power than Linn Akurate 2200. That's maximum power without any sound quality considerations.
So if you're in a situation where an Akurate 2200 clips the signal or shuts down when used with a particular speaker+room combination, you can safely assume that Tundra won't be powerful enough for your needs.
If your concern is that sound quality degrades when playing loud with 2200 or any other amp, the situation is different and you definitely need to try a Tundra. One of the unique features of Tundra is that its sound quality just gets better and better the louder you play. Until you reach its limits, where it will rather abruptly tell you that you need to lower the volume a click or two.
Re: Stability with age and environment? (And congrats, too!
Hi Ceilidh! Nice to see you on the forum again!
I think you're overestimating the complexity of audio amplifiers when comparing them to robotics. First of all, the environment for the audio circuit is very stable. It's used indoors in a dry (non condensing) environment and will heat up due to the idling current to at least 40 deg Celcius. Above 70 deg, it will shut down.
That the quality of mains power can vary in voltage and quality was initially a problem, and the reason why I decided to use dual switch mode power supplies. Those have been thoroughly tested in different environments and will provide the amplifier circuit with exactly what it needs, using all the common mains standards in the world.
Until now, my designs have been tuned with resistors, after assembly. As this is quite time consuming, I decided to do it the other way around this time. The components are selected in small groups before assembly. I select all the semiconductors; transistors, diodes etc. The power transistors have been selected into six groups and I use only one of them for Tundra. But I'm saving the others for coming products, when a different group might prove better. Diodes that had a 2% variation (the tightest tolerance that was offered by the manufacturer) have been selected to a few tenths of 1%, etcetera. I have even selected the colour of the blue LEDs on the front. Not necessary for performance, but it can be annoying if one LED has a green tint and the other glows with a reddish blue.
I have a little amp that's been powered up for close to 5 years and I have repeatedly measured the semiconductors in it. They show practically zero change, and so do the resistors I've used. I don't know exactly what happens in a decade or two, but the most likely cause for a loss of sound quality is that some capacitors will start to dry out. The ones I use have the longest life expectancy figures I've been able to find (10 000 hours at 105 deg Celcius), but eventually they will degrade. I retested a few that had been running warm for 3 years and found to my surprise that they sounded better than brand new ones. Haven't tried any that are older than that, so I don't know exactly when they start degrading.
I think you're overestimating the complexity of audio amplifiers when comparing them to robotics. First of all, the environment for the audio circuit is very stable. It's used indoors in a dry (non condensing) environment and will heat up due to the idling current to at least 40 deg Celcius. Above 70 deg, it will shut down.
That the quality of mains power can vary in voltage and quality was initially a problem, and the reason why I decided to use dual switch mode power supplies. Those have been thoroughly tested in different environments and will provide the amplifier circuit with exactly what it needs, using all the common mains standards in the world.
The answer is that the components which are critical to performance in Tundra hardly age at all.Ceilidh wrote:Q: How do you expect the Tundra to age, given that component parameters can drift over time? Will the Tundra be as stable as competing products?
Until now, my designs have been tuned with resistors, after assembly. As this is quite time consuming, I decided to do it the other way around this time. The components are selected in small groups before assembly. I select all the semiconductors; transistors, diodes etc. The power transistors have been selected into six groups and I use only one of them for Tundra. But I'm saving the others for coming products, when a different group might prove better. Diodes that had a 2% variation (the tightest tolerance that was offered by the manufacturer) have been selected to a few tenths of 1%, etcetera. I have even selected the colour of the blue LEDs on the front. Not necessary for performance, but it can be annoying if one LED has a green tint and the other glows with a reddish blue.
I have a little amp that's been powered up for close to 5 years and I have repeatedly measured the semiconductors in it. They show practically zero change, and so do the resistors I've used. I don't know exactly what happens in a decade or two, but the most likely cause for a loss of sound quality is that some capacitors will start to dry out. The ones I use have the longest life expectancy figures I've been able to find (10 000 hours at 105 deg Celcius), but eventually they will degrade. I retested a few that had been running warm for 3 years and found to my surprise that they sounded better than brand new ones. Haven't tried any that are older than that, so I don't know exactly when they start degrading.
I think I already answered this, but I don't mind doing it again:rowlandhills wrote:I was concerned that it would not sound as good at low to medium volumes!
Of course it sounds great at low to medium volumes. I certainly wouldn't release it otherwise. My initial comment was due to how I feel that the sound seems to effortlessly increase in volume when I turn it up.
Henrik's comment about power is probably due to him driving Tundra to just where it starts clipping. It can only deliver so many volts and when you go past that, the signal clips. I can personally recognise that clipping straight away, but I think many will rather perceive it as the sound getting hard and a bit choked. A few volume steps below that point, it sounds fantastic. And a few steps above, it sounds terrible.
-
rowlandhills
- Very active member
- Posts: 582
- Joined: 2008-01-27 19:25
- Location: York, UK
Thanks Fredrik.
Another question for you. I thought that you mentioned earlier in this thread that the gain was not the same as Linn power amps, and hence if there was a need for multi channels (e.g. for going aktiv) it wouldn't be possible to mix Tundra with Linn amps. I have to admit that I now can't find this comment, but I'm sure that I read it somewhere!
Anyway, I'm wondering whether in a multi-channel passive speaker environment, it would be sensible to use a Tundra on the main stereo pair and a 3200 on the centre/surrounds? Buying three Tundras and having a spare channel seems like a bit of a waste...
I guess that I could adjust the volume offset of the centre and rear channels in my Exotik+DA to be a dB or two higher/lower than the main pair, but is that enough?
Thanks.
Another question for you. I thought that you mentioned earlier in this thread that the gain was not the same as Linn power amps, and hence if there was a need for multi channels (e.g. for going aktiv) it wouldn't be possible to mix Tundra with Linn amps. I have to admit that I now can't find this comment, but I'm sure that I read it somewhere!
Anyway, I'm wondering whether in a multi-channel passive speaker environment, it would be sensible to use a Tundra on the main stereo pair and a 3200 on the centre/surrounds? Buying three Tundras and having a spare channel seems like a bit of a waste...
I guess that I could adjust the volume offset of the centre and rear channels in my Exotik+DA to be a dB or two higher/lower than the main pair, but is that enough?
Thanks.
KRDSM, Tundra to 242s
Silvers, K400, Hutter rack
Silvers, K400, Hutter rack
Looking at the manual, I believe the Exotik +DA has adjustment for -/+ 15 dB, in steps of 0.5 dB. I can't immediately work out if the non +DA version is the same.lejonklou wrote:Tundra needs an input signal 8 dB louder than a Linn power amp to play equally loud. That's 8 steps on a Kikkin or a Linn preamp.
I don't remember how large the adjustment range is on an Exotik, maybe someone else does?
I'm pretty sure it is the same for the analog Exotik and that the level adjustment is done in the analog part of the preamp on an Exotik D+A as well.tmilligan wrote:Looking at the manual, I believe the Exotik +DA has adjustment for -/+ 15 dB, in steps of 0.5 dB. I can't immediately work out if the non +DA version is the same.lejonklou wrote:Tundra needs an input signal 8 dB louder than a Linn power amp to play equally loud. That's 8 steps on a Kikkin or a Linn preamp.
I don't remember how large the adjustment range is on an Exotik, maybe someone else does?