"Mark, did you received my PM?"
No...
Pool, This is the first I've heard of this....
"It's getting really frustrating."
Yes it does... Bot if two or three orders out of several hundred that I've done over the last couple of years doesn't make it where its headed thats still a pretty good percetage, 98.7% that did. Mail is Mail and packages do get damaged/lost on occasion but the sucess rate is pretty high for what it costs.
There is always the more expensive Fed-Ex route that can cost upwards of 30 dollars if you feel more comfortable that way... But with postal items it is not possible to track any item once an item leaves the hands of the U.S. Postal Service when it arrives in another countrys Postal Service Hands.
Folks... If I don't get your PM through this site then contact the admin of this site for my real e-mail address. I have no problem with them giving it out. If that doesn't work then Snail Mail will.
Thanks!
Mark
No...
Pool, This is the first I've heard of this....
"It's getting really frustrating."
Yes it does... Bot if two or three orders out of several hundred that I've done over the last couple of years doesn't make it where its headed thats still a pretty good percetage, 98.7% that did. Mail is Mail and packages do get damaged/lost on occasion but the sucess rate is pretty high for what it costs.
There is always the more expensive Fed-Ex route that can cost upwards of 30 dollars if you feel more comfortable that way... But with postal items it is not possible to track any item once an item leaves the hands of the U.S. Postal Service when it arrives in another countrys Postal Service Hands.
Folks... If I don't get your PM through this site then contact the admin of this site for my real e-mail address. I have no problem with them giving it out. If that doesn't work then Snail Mail will.
Thanks!
Mark
Mark, judging by the way you have ignored my repeated emails and posts to you about this matter (by ignoring them), the facts seem to speak for themselves. I appreciate that items may go missing in the post once, but when this happens repeatedly, the facts speak for themselves, you did not send them, there is no slander here!!!
However, should you now consider the situation high profile enough to sort out, i will happily withdraw any statement made, until then i am quite sure i know what is going on.
However, should you now consider the situation high profile enough to sort out, i will happily withdraw any statement made, until then i am quite sure i know what is going on.
First Post
Hello,
I have read this forum thread from the start, congrts on a great project. Mark I have added my name to the wiki GB standby
I would also like to have the diodes if you still have enough
Did anybody compare the sound of the krells or clones against the chip amps eg. the "Amp2" from 41hz audio using the Tripath TK2350 Chipset??
I have done a lot of experimental work with audio and would like to progress this further if some of you are interested? Andrew? Pierre?Jacco?Flodstroem?
Regards,
Andrew...
Hello,
I have read this forum thread from the start, congrts on a great project. Mark I have added my name to the wiki GB standby I would also like to have the diodes if you still have enough
Did anybody compare the sound of the krells or clones against the chip amps eg. the "Amp2" from 41hz audio using the Tripath TK2350 Chipset??
I have done a lot of experimental work with audio and would like to progress this further if some of you are interested? Andrew? Pierre?Jacco?Flodstroem?
Regards,
Andrew...
Have compared it to my Mini Aleph, and Aleph 2's and the KSA-100 blows em all away... but again it also depends on what efficiency speakers you are trying to drive.... The Aleph's seem more suited to higher efficiency speakers and the Krells will drive just about anything. From a sound standpoint I prefer the Krell. Let me know how many boards and diodes you need.... I am sorting out diodes tonight. There are about 10 gold plated boards left at this stage.
Mark
Mark
Thanks, Can you give me 2 boards and 4x 1N5309 diodes
Price is $50 for the boards.
how much for the diodes? will they ship with the boards,
or will jacco send them?
tnx.
Also I would like to state that I have been recieving parcels from many many different people from all over the world in the last 10 yrs and in most cases they have all arrived safely, some delayed, and I lost 2 from about 200
I have found global priority very good and prompt delivery from 3 to 7 days to Ireland, so please do not blame Mark, if you are worried about delivery use an alternative, or go to your local shop and pay the Price!
Price is $50 for the boards.
how much for the diodes? will they ship with the boards,
or will jacco send them?
tnx.
Also I would like to state that I have been recieving parcels from many many different people from all over the world in the last 10 yrs and in most cases they have all arrived safely, some delayed, and I lost 2 from about 200
I have found global priority very good and prompt delivery from 3 to 7 days to Ireland, so please do not blame Mark, if you are worried about delivery use an alternative, or go to your local shop and pay the Price!
No problem. I agree with your findings on mail in general. I have had shipments come to me damaged and I have had at least one lost shipment comming to me that never showed up in my 52 years on earth. I have also had trouble to Denmark(not with this group) with Priority Mail twice in a row to the same address... wrong address probably. At any rate you can e-mail me directly for the boards... cinerama84106 at yahoo.com. You know what to do with the address when you type it out....
Mark
Mark
Mark-
In the hundreds, if not thousands, of packages you have shipped out I'm sure you have had better than a 95% success rate on delivery. We all know your reputation is solid.
Everyone who buys parts knows that shipping is a gamble (usually safe, but still a gamble) and the package could be lost. If it is a concern they can opt for insured and tracked shipping at their cost. I would hope anger is directed at the cause and not a convenient target.
Congratulations on your marriage and I wish you the best.
In the hundreds, if not thousands, of packages you have shipped out I'm sure you have had better than a 95% success rate on delivery. We all know your reputation is solid.
Everyone who buys parts knows that shipping is a gamble (usually safe, but still a gamble) and the package could be lost. If it is a concern they can opt for insured and tracked shipping at their cost. I would hope anger is directed at the cause and not a convenient target.
Congratulations on your marriage and I wish you the best.
I'd have to agree with most of the people on this thread: Mark's contribution on the forum has been gigantic over the years and considering that it's non-profit and on the sideline I personally think he deserves a medal So if something takes a while or gets lost in the mail, compared to what didn't his track record is damn good. If it gets lost in the mail, deal with it: living in Dark Africa I've had to deal with lost packages and excessive customs delays a number of times.
Comparing the Krell to other topologies such as class-D ones, there is no clear answer. You can found extensive measurements and discussion of the KSA100 Mk2 somewhere on this thread around October 2006.
Class-D research being my day job, I can safely say that the class-D amps' distortion and noisefloor (the good ones such as Hypex anyway; the low-power Tripath does not fall in this category!) are considerably better than the Krell's. However, the Krell will beat most class-D designs when it comes to difficult loads such as the Apogees it was originally designed for. It is possible to design class-D for these loads, but will result in much higher complexity and cost, with little to no performance increase for normal loads. This is different to class-A amps where the relation between this capability and performance is more concrete. Designing for higher power is a more attractive and sensible choice, with its own set of challenges though.
Transient response is also exemplary on the Krell. It has been mentioned a number of times, and verified by my measurements, that the Krell's legendary bass can be attributed to its distortion under load rising fairly little compared to other amps. Class-D amps in particular suffer from this. I'm not going to mention names, but due to their more violent behavior under load class-D manufacturers tend to be very selective about their performance specs. One example would be a rated distortion of 0.01% at rated power into 8ohms, weighted. Using the same weighting, I measured 0.1% at 30W into 12ohms! And yes, it was with a subsequent hardware lowpass filter for the measurements. The published measurements of the Hypex amps are pretty much spot-on though.
Last year I've done an AB comparison (speakers switched with a relay box and both amps at the same volume) between the KSA100 and a TacT S2150 "digital" amplifier. The source was the same, a Denon DVD3910 universal player, with the digital output feeding the TacT directly. The analog out was fed to the KSA via a high-end DIY Audio Note M7 tube preamp. Speakers were ProAc 2.5 clones. Admittedly the KSA has improved somewhat afterwards with a few small adaptions, but my personal opinion as well as a professional violinist was that the TacT had more detail and clarity, but the Krell sounded more natural and full-bodied. We couldn't hear any difference to the bass, but may be attributed to the ProAcs being a fairly easy load to drive. The Krell would probably have won with lower impedance loads.
It is a very non-scientific sweeping statement, but my conclusion is that class-D amps' performance varies much more than class-A ones with different loads. Auditioning with your own speakers is always important when shopping for amps, but when considering a class-D amp it's vital.
So if something takes a while or gets lost in the mail, compared to what didn't his track record is damn good. If it gets lost in the mail, deal with it: living in Dark Africa I've had to deal with lost packages and excessive customs delays a number of times.Comparing the Krell to other topologies such as class-D ones, there is no clear answer. You can found extensive measurements and discussion of the KSA100 Mk2 somewhere on this thread around October 2006.
Class-D research being my day job, I can safely say that the class-D amps' distortion and noisefloor (the good ones such as Hypex anyway; the low-power Tripath does not fall in this category!) are considerably better than the Krell's. However, the Krell will beat most class-D designs when it comes to difficult loads such as the Apogees it was originally designed for. It is possible to design class-D for these loads, but will result in much higher complexity and cost, with little to no performance increase for normal loads. This is different to class-A amps where the relation between this capability and performance is more concrete. Designing for higher power is a more attractive and sensible choice, with its own set of challenges though.
Transient response is also exemplary on the Krell. It has been mentioned a number of times, and verified by my measurements, that the Krell's legendary bass can be attributed to its distortion under load rising fairly little compared to other amps. Class-D amps in particular suffer from this. I'm not going to mention names, but due to their more violent behavior under load class-D manufacturers tend to be very selective about their performance specs. One example would be a rated distortion of 0.01% at rated power into 8ohms, weighted. Using the same weighting, I measured 0.1% at 30W into 12ohms! And yes, it was with a subsequent hardware lowpass filter for the measurements. The published measurements of the Hypex amps are pretty much spot-on though.
Last year I've done an AB comparison (speakers switched with a relay box and both amps at the same volume) between the KSA100 and a TacT S2150 "digital" amplifier. The source was the same, a Denon DVD3910 universal player, with the digital output feeding the TacT directly. The analog out was fed to the KSA via a high-end DIY Audio Note M7 tube preamp. Speakers were ProAc 2.5 clones. Admittedly the KSA has improved somewhat afterwards with a few small adaptions, but my personal opinion as well as a professional violinist was that the TacT had more detail and clarity, but the Krell sounded more natural and full-bodied. We couldn't hear any difference to the bass, but may be attributed to the ProAcs being a fairly easy load to drive. The Krell would probably have won with lower impedance loads.
It is a very non-scientific sweeping statement, but my conclusion is that class-D amps' performance varies much more than class-A ones with different loads. Auditioning with your own speakers is always important when shopping for amps, but when considering a class-D amp it's vital.
PWatts said:
You may have been hearing the smoothing of the tube preamp.
With DIY I think the best strategy is to marry the amp to one speaker at a time by putting an active crossover PCB into the amp chasis and using the two(three) channels to drive one speaker. You could design the crossover PCB to include a connector for a daughter card with all of the passive crossover components so it would be easy to divorce and re-marry.(Sorry Mark).
Longer term, going ALL DIGITAL continues to promise attractive advantages like room equalization, steep slopes, and near perfect time delay and phase control. I think you have to break a few laws to do that today with DVDs..... something about copyright protection.
Actually I chose the M7 for its low distortion and non-valve characteristics. The "smoothing" effect of valves are a result of even-harmonic distortion, but the M7's fall well below -100dB and is therefore highly unlikely to make any major contribution to the sound. It can therefore be compared to any proper solid-state preamp.
A bit off-topic but in the spirit of LineSource's comments: The class-D amps I'm working with are in fact also of the digital kind, with results far surpassing those of the digital competition, as well as most of the analog class-D and many class-AB ones. Of course there actually is no such thing as a "digital amp"; the building blocks have merely shifted. Indeed, if you take the output of the processor directly into a lowpass filter it's a DAC. The only difference is that unlike conventional DAC's the PWM output has been specifically tailored to feed a switching output stage. However, it still means the DAC and it's associated circuitry is eliminated, leading to considerably lower cost and more control over the resulting fidelity by removing a variable from the chain.
Because all the calculations are done inside an FPGA, adding additional filters are easy without hardware changes. In fact, one of my final-year students last year has developed a digital valve emulator software function, that can switch between 5 popular guitar valve amps as well as the intensity of the addition. It worked perfectly. This semester I have one working on digital filters such as the Behringer 2496, but still in the FPGA so NO additional DSP's, ADC's or DAC's. This leads to much better performance, flexibility and lower cost. Just a software change. Same goes for room correction although implementing that in VHDL is a Masters or PhD degree on its own. One can add as many of these extras as the FPGA's capacity allows, and with our current software we can fit two channels with 20% to spare for additions inside a $20 FPGA, and 18 into a $200 one. One of the next developments is the so-called "hybrid speaker", where the amp is bolted directly on the speaker magnet as heatsink. Besides the obvious advantages this theoretically allows the LC demodulation filter to be omitted because the short "speaker cable" (pretty much the tinsel leads) radiates little EMI before the woofer's voice coil inductance does the rest. This is especially attractive for pro audio applications where each speaker now only needs a digital connection and power instead of massice rackmount amps and hosepipe speaker cables.
We're working closely with FireWire since it is a natural partner for digital amplification, and the recent development of FireWire transmission over CAT5 and fibre optic instead of a short lead makes it a serious rival for Ethernet-based systems. We've also done the hardware development of a FireWire bridge that's essentially a FireWire equivalent of an Ethernet switch and therefore allowing unlimited devices to be daisy-chained. The main advantage is naturally for Pro audio, but multiroom and multichannel consumer setups also have much to gain. Pretty much the same as Meridian does, but more refined (they use S/PDIF and a comms cable to each speaker, and inside a DSP crossover, DAC and analog amps).
Copyright is only an issue with 24/96 & 24/192 audio, and as long as the hardware developer has a license for access to unencrypted interface chips it's not a problem. The same chips are available to anyone, but the encrypted variety. This means that 24/96 can be transmitted as long as it's not encrypted, which is still attractive for PC-based audio. For S/PDIF there's no such workaround if using a DVD player: law stipulates that encrypted 24/96 material must be downsampled to 16/48. This can be bypassed by ripping the audio with a PC, removing the copyright and either play it from the PC or write it to a DVD.
Anyhow, back to the Krell
A bit off-topic but in the spirit of LineSource's comments: The class-D amps I'm working with are in fact also of the digital kind, with results far surpassing those of the digital competition, as well as most of the analog class-D and many class-AB ones. Of course there actually is no such thing as a "digital amp"; the building blocks have merely shifted. Indeed, if you take the output of the processor directly into a lowpass filter it's a DAC. The only difference is that unlike conventional DAC's the PWM output has been specifically tailored to feed a switching output stage. However, it still means the DAC and it's associated circuitry is eliminated, leading to considerably lower cost and more control over the resulting fidelity by removing a variable from the chain.
Because all the calculations are done inside an FPGA, adding additional filters are easy without hardware changes. In fact, one of my final-year students last year has developed a digital valve emulator software function, that can switch between 5 popular guitar valve amps as well as the intensity of the addition. It worked perfectly. This semester I have one working on digital filters such as the Behringer 2496, but still in the FPGA so NO additional DSP's, ADC's or DAC's. This leads to much better performance, flexibility and lower cost. Just a software change. Same goes for room correction although implementing that in VHDL is a Masters or PhD degree on its own. One can add as many of these extras as the FPGA's capacity allows, and with our current software we can fit two channels with 20% to spare for additions inside a $20 FPGA, and 18 into a $200 one. One of the next developments is the so-called "hybrid speaker", where the amp is bolted directly on the speaker magnet as heatsink. Besides the obvious advantages this theoretically allows the LC demodulation filter to be omitted because the short "speaker cable" (pretty much the tinsel leads) radiates little EMI before the woofer's voice coil inductance does the rest. This is especially attractive for pro audio applications where each speaker now only needs a digital connection and power instead of massice rackmount amps and hosepipe speaker cables.
We're working closely with FireWire since it is a natural partner for digital amplification, and the recent development of FireWire transmission over CAT5 and fibre optic instead of a short lead makes it a serious rival for Ethernet-based systems. We've also done the hardware development of a FireWire bridge that's essentially a FireWire equivalent of an Ethernet switch and therefore allowing unlimited devices to be daisy-chained. The main advantage is naturally for Pro audio, but multiroom and multichannel consumer setups also have much to gain. Pretty much the same as Meridian does, but more refined (they use S/PDIF and a comms cable to each speaker, and inside a DSP crossover, DAC and analog amps).
Copyright is only an issue with 24/96 & 24/192 audio, and as long as the hardware developer has a license for access to unencrypted interface chips it's not a problem. The same chips are available to anyone, but the encrypted variety. This means that 24/96 can be transmitted as long as it's not encrypted, which is still attractive for PC-based audio. For S/PDIF there's no such workaround if using a DVD player: law stipulates that encrypted 24/96 material must be downsampled to 16/48. This can be bypassed by ripping the audio with a PC, removing the copyright and either play it from the PC or write it to a DVD.
Anyhow, back to the Krell
Hello
Interesting writings from PWatts, indeed. Im just in the start right before going to take advantage of a Firewire Solo mic-amp for my speaker testing project.
But back to the Krell building project.
How far have you come with it? Anyone done som testing yet?
I lack some components (the CP extra caps) and Im tweaking with the heat sink for the drivers, but finally I have got an idea for a solution. I will use the AAVID part #0S499 (3" in length). Guess it has a diss. factor of ca 1.2°C/W. I am going to use 3 pair of drivers to the 6 pair of powers.
I will post pictures when I have my first main board up and running so you could see my solution.
Happy building
Ragards
Interesting writings from PWatts, indeed. Im just in the start right before going to take advantage of a Firewire Solo mic-amp for my speaker testing project.
But back to the Krell building project.
How far have you come with it? Anyone done som testing yet?
I lack some components (the CP extra caps) and Im tweaking with the heat sink for the drivers, but finally I have got an idea for a solution. I will use the AAVID part #0S499 (3" in length). Guess it has a diss. factor of ca 1.2°C/W. I am going to use 3 pair of drivers to the 6 pair of powers.
I will post pictures when I have my first main board up and running so you could see my solution.
Happy building
Ragards