"Any modifications however would make it impossible to return the amp to ClassD. " Without exception? Are you sure? Maybe you can check your policy and get back to us. What's the difference if it's used versus used over the course of a seven day gestation? In the end it is still used. msb64 lists a few issues with it you likely haven't heard on your home version. Maybe in production some parts got changed? These things do happen. The recommendations you gave make good sense and they're a good place to start trying to extract more from it but it's not so cut and dry. You're in no way guaranteed "subtle but worthwhile" anything if you modify it. In fact you're likely to just make it a lot worse if you have no clue what you're doing, remember this is class d. Acclimatisation to the new sound could take a lot longer than a an hour, it can take weeks for the new toy syndrome alone to subside, and it can take many months before you really get to know it and stop lying to yourself if that's what you're doing. Keep in mind this is not a class A amp, there's a lot more going on sonically and most ears are ignorant of it, at least at first and especially when blinded by a new toy and further deafened by the proud papa syndrome of having plugged it together themselves. If you're a more experienced builder and listener you'll go through those steps in lightning speed compared to anyone who's less experienced but it can still take some weeks. It stands to reason swapping a few components around or employing improved auxiliary supplies is the exact first thing anyone in the know might try to fix these little sonic quibbles that were mentioned. Of course in seven days you haven't even the time to have truly experimented with your own means before blaming it on the module. Here we have it reported it gets a lot noisier at higher output levels. Is that a layout issue of the module? Probably, as selling turnkey IC solutions in datasheet dress doesn't make you an RF layout wizard. Equally likely, it could simply be a layout or wiring issue at the end user side of things or a combination of the two is even probable, not everyone is experienced with high speed low noise RF techniques or pitfalls. They may take six months just to figure that bit of plugging it in, out. Before you guys go on talking about diamond in the ruff I'd say give your tactics a try before cementing your conclusions publicly, you may find a sufficient bottleneck to render any further "improvements" all but pointless.
I haven't found an impedance chart yet, but did find this interesting review on the infinity-classics.de website.
hxxp://www.infinity-classics.de/infinity/models/Kappa-series-1987/Kappa-8-8A/Kappa_8_review_01.jpg
hxxp://www.infinity-classics.de/infinity/models/Kappa-series-1987/Kappa-8-8A/Kappa_8_review_02.jpg
I knew they dipped pretty low, but didn't know they went down that far.
That is definitely a challenge for most of the affordable equipment out there.
Checking out the crossover schematic and seeing all the NPE caps in there gives me an idea, another project to work on too, time to visit the Solen site again 🙂
The speakers can be biamped as well, I will try the classd on each half to see how it responds. Wish I had another one on hand to run all classd for the test.
Bl00dy H33l, I'd shout for an xover redesign with a load like that.
1.3 ohm minimum at 40hz rising no higher then 4 ohms up to 250hz. No wonder they are a struggle to drive.
I'm surprised the manufacture released them in their current form.
1.3 ohm minimum at 40hz rising no higher then 4 ohms up to 250hz. No wonder they are a struggle to drive.
I'm surprised the manufacture released them in their current form.
Too funny.
No need to state the obvious on methodology though, many of us have been there done that already.
Personally I would rather hear about first hand experience that others have had with these units, rather than arm-chair diy speculation.
That being said, suggestions made by anyone in a polite and non-condescending manner would be greatly appreciated as well 🙂
They do require some consideration when choosing power for them, but when fed properly they do sound pretty decent. They do work well with the larger vintage Threshold amps though, and even better with 2 of them. I have had success with stock and modded Hafler amps too.
The front and rear firing emits are pretty revealing too, so any high frequency anomalies are quickly revealed.
Last edited:
It would seem that the crossover from the bass unit to the first midrange @ 90hz is where a lot of the trouble could lie with this loudspeaker. If biamping was on the cards, I'd be seriously tempted to try using an active crossover to replace the passive one at that point.
Thanks for the recommendation 5th element, anything particular in mind?
I remember reading about some guys going the active crossover route when modding RS1Bs awhile back, might have possibly been Peavy units.
Would modding the existing crossovers be an option as well, or would that just create additional issues?
I speculate it takes you longer than seven days to figure out the importance of twisting wires. In the interim, we've a listening report of increased noise with increasing power, that apparently, we're to judge the quality of it by. Guffaw. You can tell more truth simply by looking at it. Such as it being conceivable and even probable that the increasing noise with increasing power you mentioned is due to design and layout issues gone ignored, and they won't be greatly diminished with a few twists or component upgrades. For the cost of some twisting though, it seems worthwhile exploring the efficacy of proper techniques that take longer than seven days to acquire, even for those who've been there and returned with the certification t-shirt. Fauteuil, to me!
I'm not entirely sure what your agenda is with these posts. No manufacturer, whoever they are, will accept a unit back once you've taken your soldering iron to it, or started to cut PCB tracks. You will have to do both of these if you wish to try out separate power supplies, or even replace some of the stock components.
In its standard form direct from ClassD all you have to do is apply +50V and Ground, that's it for power.
Then you've got the outputs and inputs.
There's not really a lot you can do with this. You're obviously going to keep them separate.
One thing I will mention is the power supply. TI state quite clearly that a 'long' wire (from the PSU) to the chip will result in an increase in low frequency distortion. I have measured this and it's significant.
The noise that one person mentioned could be the result of things not related to the TI module. If you've got silence with the inputs connected to ground then it's not the modules fault.
Obviously there's hi frequency switching noise thrown into the ether, if that's affecting another component in your system adversely, maybe you need to shield the 5630 better. Or perhaps the unit that's responding adversely to the switching noise is inherently unstable and the switching noise is pushing it to oscillate.
There are so many what ifs that it isn't worth speculating.
Besides where's the part where someone mentions increased noise with output power?
I managed to track down the source of the high frequency anomalies that I had encountered. After switching out various components, interconnects etc. I found out the issue was the recording that I had been using for testing, a track that was familiar to me but a different release.
I contacted Tom at ClassDaudio and he responded very quickly, made some helpful suggestions and immediately offered to look after the unit. Fortunately I was able to determine the issue before proceeding any further.
As far as the bass is concerned, a mildly modified S300 MKII did a much better job at dealing with the sub 2 ohm loads (as low as 1.3 ohms according to reports) imparted by the kappa 8s. I don't think this should affect the majority of people though, that will be looking at these classD amps.
As others have mentioned earlier, this is definitely a different animal though.
It sounds accurate, revealing, dynamic, is dead quiet and has a kind of presence that is hard for me to describe at this time. It is hard to say where it will fit into the scheme of things, as this hobby dictates that things rarely stay the same indefinitely. One thing I can say definitely say right now is that I do like it a lot and there is not a chance of this one going back, and I will be getting at least one more in the near future. I also have no problem recommending this to my friends, and will drag as many of them as I can to have them give it a good listen to over here.
I will give it a few weeks burn in time on the Kappa 8s, and then give it a good test run on IRS Beta mid/high towers.
I contacted Tom at ClassDaudio and he responded very quickly, made some helpful suggestions and immediately offered to look after the unit. Fortunately I was able to determine the issue before proceeding any further.
As far as the bass is concerned, a mildly modified S300 MKII did a much better job at dealing with the sub 2 ohm loads (as low as 1.3 ohms according to reports) imparted by the kappa 8s. I don't think this should affect the majority of people though, that will be looking at these classD amps.
As others have mentioned earlier, this is definitely a different animal though.
It sounds accurate, revealing, dynamic, is dead quiet and has a kind of presence that is hard for me to describe at this time. It is hard to say where it will fit into the scheme of things, as this hobby dictates that things rarely stay the same indefinitely. One thing I can say definitely say right now is that I do like it a lot and there is not a chance of this one going back, and I will be getting at least one more in the near future. I also have no problem recommending this to my friends, and will drag as many of them as I can to have them give it a good listen to over here.
I will give it a few weeks burn in time on the Kappa 8s, and then give it a good test run on IRS Beta mid/high towers.
I'd be tempted to say that in the hands of a good designer, with the right tools available to them, that they could most likely come up with a far easier to drive system using the same drivers - all passive. You might end up sacrificing a dB to get it, but it'd be worth it in my opinion. It'd also be quite difficult to do as you'll effectively be designing a brand new loudspeaker.
What you want to do is measure the individual responses of the bass and midrange drivers with the current passive xover in place. As the xover frequency is so low you can measure this near field to remove the effects of the room.
Having done that you can see what acoustic targets the designer was trying to reach. Hopefully you will see reasonably smooth and symmetrical roll off from both drivers. The roll off hopefully following a traditional filter transfer function, 2nd or 4th order linkwitz riley for example.
You'd then want to measure the drivers without their crossovers attached. This should be easy to do with the bass driver, but with the midrange you'll have to see exactly what parts are contributing to the hi pass and remove only those. The trouble here could be that some parts of the hi pass help shape the low pass a little, but that's a risk you'd have to take for trying this out.
Now you want to construct an active filter that tries to mimic what the old passive xover did. But to be perfectly honest I'd be tempted to ignore what the old filter might have done and get both the bass and midrange to arrive at a 4th order linkwitz riley acoustic. You'd also want to have a gain control on say the bass part in case you need to readjust the level. Having removed the old inductor and its associated series resistance you will probably gain some extra sensitivity.
If you can take a nearfield measurement of both the bass and midrange drive units without their respective lowpass/high passes in place, you can send me the files and I can whip up an active xover in LspCAD if you want.
ARTA + a decent mic and preamp, say the behringer ECM8000 + pre is all you will need to take good measurements.
If you're interested you might wish to PM me as discussing everything here in the ClassD thread isn't really appropriate.
The agenda is to peel away the rosy tint and see what we're really dealing with. To be kind I will agree, this thing is a steal, but that's based on price and not performance. So let's talk performance then, and that is not restricted to subjective impressions as the boundaries often seem to impose for some ominous reason.
If it's of sufficient caliber that with some know how performance of a standard indicated by the specifications of the ICs used could be had, then they should be flying off the shelves. The problem that I'd like to caution on is that intrinsic design flaws and weaknesses pertaining to layout and so forth will always be. They can't much be diminished by the typical end user no matter the tweaks. If it is in that category, and there's every chance that it is deeply in that category, then at least it's still cheap, but not such a diamond in the ruff as we've already seen prematurely concluded.
In one sentence you state all you have to do is 50V and 0, reiterating that plugging it is all that a user can do. Later "but the wires have to be really short, or else". Your typical module pluging-inner will know that how? Come on man, it's not a simple matter of plugging in if you actually want a listenable amp out of it.
Everyone knows the caliber of the simple 50 and 0 has massive influence in any amp and I promise it can be far worse with class d. Plugging it in explores nothing of that. There's also every chance you might plug it in and not like it, then ship it back for a refund in seven days, selling it short in doing so as with some proper techniques and the time and effort they require, it might really be something special, who knows? None of us do yet. Again this is why I say seven days is a laugh, which was the original agenda wasn't it.
Noise rising with increasing power issues were mentioned several posts back by msb64, that I found believable. Now, msb64, you're saying it is resolved in that you used a different release of a track you're otherwise familiar with. I don't believe that because, how is that release modulated by the power level of the amp. You need to twist your wires either way, start with the speaker wires, then DC supply wires where you've another nice big loop spewing all over your inputs. That would be the absolute minimum I'd do. You might find those kinds of problems diminished if the amp is any good at all.
I think you're probably hearing quite right though based on your descriptions and also the amount of time you've had it. In some time you'll better be able to describe that presence, and by then you'll be noticing a lot more. This presence you mention is a familiar signature. Is it a big problem? I'd say it depends on the extent of it, solely based on the fact that nothing is truly perfect, and so you have to accept some flaws, but said flaws also must be acceptable.
Preferably it wouldn't be at all different from a class A as in that respect they excel, otherwise they're imperfect as well. It can be greatly ameliorated with proper components and the very best layout and implementation, but I'm unconvinced in this case the board itself is capable of it, and honestly I suspect not. I would like to see some relevant measurements for it that might give some indication, particularly for the common mode noise on the output. Are there any?
Shifting back to 5th, "If you've got silence with the inputs connected to ground then it's not the modules fault". Blatantly incorrect, as you later indicate some awareness of. Ideas like that lead to people trying to sell modules with a 3k input impedance that we already know is quite wrong.
If the modules end up having EMI issues that infects other components that normally work fine, given all the basics that a user has control over were properly executed, then it is the fault of the modules and should be resolved at their source.
It is worth speculating when you're having problems, or rather hypothesizing, because that's how they get considered and later resolved. If other units are intrinsically unstable that's their problem, but I'm not looking to pass the buck. One set of measurements will tell the truth to this story and I'd trade them for a seven day grace period in a heartbeat because I know the rest can be more reasonably dealt with by typical users in some fashion or other.
If it's of sufficient caliber that with some know how performance of a standard indicated by the specifications of the ICs used could be had, then they should be flying off the shelves. The problem that I'd like to caution on is that intrinsic design flaws and weaknesses pertaining to layout and so forth will always be. They can't much be diminished by the typical end user no matter the tweaks. If it is in that category, and there's every chance that it is deeply in that category, then at least it's still cheap, but not such a diamond in the ruff as we've already seen prematurely concluded.
In one sentence you state all you have to do is 50V and 0, reiterating that plugging it is all that a user can do. Later "but the wires have to be really short, or else". Your typical module pluging-inner will know that how? Come on man, it's not a simple matter of plugging in if you actually want a listenable amp out of it.
Everyone knows the caliber of the simple 50 and 0 has massive influence in any amp and I promise it can be far worse with class d. Plugging it in explores nothing of that. There's also every chance you might plug it in and not like it, then ship it back for a refund in seven days, selling it short in doing so as with some proper techniques and the time and effort they require, it might really be something special, who knows? None of us do yet. Again this is why I say seven days is a laugh, which was the original agenda wasn't it.
Noise rising with increasing power issues were mentioned several posts back by msb64, that I found believable. Now, msb64, you're saying it is resolved in that you used a different release of a track you're otherwise familiar with. I don't believe that because, how is that release modulated by the power level of the amp. You need to twist your wires either way, start with the speaker wires, then DC supply wires where you've another nice big loop spewing all over your inputs. That would be the absolute minimum I'd do. You might find those kinds of problems diminished if the amp is any good at all.
I think you're probably hearing quite right though based on your descriptions and also the amount of time you've had it. In some time you'll better be able to describe that presence, and by then you'll be noticing a lot more. This presence you mention is a familiar signature. Is it a big problem? I'd say it depends on the extent of it, solely based on the fact that nothing is truly perfect, and so you have to accept some flaws, but said flaws also must be acceptable.
Preferably it wouldn't be at all different from a class A as in that respect they excel, otherwise they're imperfect as well. It can be greatly ameliorated with proper components and the very best layout and implementation, but I'm unconvinced in this case the board itself is capable of it, and honestly I suspect not. I would like to see some relevant measurements for it that might give some indication, particularly for the common mode noise on the output. Are there any?
Shifting back to 5th, "If you've got silence with the inputs connected to ground then it's not the modules fault". Blatantly incorrect, as you later indicate some awareness of. Ideas like that lead to people trying to sell modules with a 3k input impedance that we already know is quite wrong.
If the modules end up having EMI issues that infects other components that normally work fine, given all the basics that a user has control over were properly executed, then it is the fault of the modules and should be resolved at their source.
It is worth speculating when you're having problems, or rather hypothesizing, because that's how they get considered and later resolved. If other units are intrinsically unstable that's their problem, but I'm not looking to pass the buck. One set of measurements will tell the truth to this story and I'd trade them for a seven day grace period in a heartbeat because I know the rest can be more reasonably dealt with by typical users in some fashion or other.
Modules like this never fly off the shelves.
This is after all a one chip solution. - most audiophiles will squeak at that.
It's Class D, this already has a stigma attached.
It's got a mini switch mode chip on it that drives the linear regs for the opamps. Another thing to dislike.
There is the potential for interference between equipment with the switching frequency.
I could carry on, but there are enough small details to prevent one from perhaps buying.
But then it also depends on why the user is buying them. As a potential upgrade? Specifically to replace a piece of kit they've already got? Or are just interested in what the thing can do.
I didn't build mine expecting the world. I wanted to build one for the sake of building it! Because the idea struck me as intriguing, I've always been drawn to what Class D can do well, efficiency and ease of high power. This chip just made it convenient enough for me to bother putting money into it.
I'd be surprised if many people were buying these to replace the well liked amps they already have. The angle the reviews have come over at, seem to imply the buyers were interested in seeing what these could do, nothing more nothing less.
Then that problem is inherent to the chip itself and not the layout. ClassD have followed the TI datasheets that show quite clearly how to maximise the performance of the chip.
It says it quite clearly in the TAS5630 datasheet. I would expect anyone who's interested in buying a module to read it over a couple of times.
That's what threads like this are for, to find out perhaps, if modding is successful. But aside from that no one seems to be disappointed in the least with their purchases. In fact everyone praises the modules highly! Seven days isn't enough to coax the most that the modules may be capable of providing, no. But the inherent qualities will still remain. If you dislike the sound, you send it back. You're not going to perform a 180 on the way it sounds with a few modifications. Only snake oil pushers will try and claim things like that.
It's rather simple really. The noise he was hearing was present in the recording. The louder he turned it up, the louder the noise got. That is all.
I will be posting a whole load of measurements of my build when I get some decent inductors. I hadn't really thought about going into much detail about the noise. But I'll see what I can do.
The comment was merely stating that if the signal going into the amplifier is removed (shorted to ground) and the noise stops, then its likely another piece of equipment upstream is causing the noise. Or rather could be being affected by the module and producing noise as a result.
It's not the fault of the module if it's naked to the sky. If it's in a properly shielded case and still you've got problems, then yes, I agree with you.
Yes I would also like to know why most of the websites that sell various amplifier modules come with very little measured data. You'd think it would be easy enough for them to provide you with a few graphs and if they can't, why can't they!? I don't want to be buying modules off of companies that can't or don't measure their products. I would at least expect any company to test each product before shipping to ensure it's in full working order.
Brief impressions TAS5615 160WX2 vs Tripath Sure 100Wx4
I though I would post some initial thoughts about the "Class D Audio" 160W X 2 300W single supply module based on the TI TAS5615.
Since I until recently I also was running the Sure 4X100W module the obvious differences are still fresh in my mind. The 4x100w expired prematurely.
Left with the Meanwell SMPS I picked "Class D Audios" 160W X 2. Now you can get a lot more watts from the same company for about what I paid ($135 + Ship), however I didn't want to buy a new dual voltage linear supply.
I hooked the 160x2 to the very same old slightly modified but relatively inefficient KLH acoustic suspension speakers driven from a pioneer CD directly through 2 50K potentiometers e.g. no preamp. In other words apples were compared to apples.
Now because I haven't run the 160X2 long enough to form a strong opinion, I will report only what I am very certain of.
First, the TI device does not sound way better than the tripath 4x100. No knockout was scored in the initial subjective listening area. Keep in mind that even with the Meanwell cranked to 28V, I am nowhere near the voltage limit of the TI module. I wonder if these things are optimized for a voltage like the tripath modules.
The tripath 4x100 was distinctly brighter than the TI 2x160 although both were too bright for my particular taste and made me yearn for a tone control.
The TI clearly delivered a better low end out of the box. Kick drums and bass transients. I did a double take on a tuba blat which surprised me.
The TI had a much higher input sensitivity than the 4x100 with the pot way below a quarter revolution. Incidentally, read the manual (RTFM) and avoid my stupid error of connecting my single ended drive signal differentially. In other words use + and GND not + and -. My error resulted in an annoying line buzz that I though was originating in my drive signal. Rest assured that when connected correctly the module had no audible hiss or hum, but I did not stick my head in the speaker cabinet for that last db.
I had opportunities to blast both modules while playing Steely Dan. I was vacuuming for my wife (no comments!) which is also an excuse to push out some watts. The Sure 4x100 ran fairly hot in this situation while the TI was slightly warm. Think touching radiator vs touching a pet.
The modules definitely sounded different I think mostly due to imaging. I can say the 4x100 tripath had excellent high frequency imaging. The TI had a more pronounced midrange and fatter bass with no mods.
I am holding off on a judgment since the tripath grew on me over a series of days. Maybe the TI will do the same. I still wonder if these modules actually change sound or do our brains rewire themselves to adapt to the tonal balance. I am increasingly convinced that with DSPs and recording studio effects, sound systems are becoming the equivalent of tinted glasses with people preferring lots of different shades e.g tube sound, class D, class a ETC. 🙂
I though I would post some initial thoughts about the "Class D Audio" 160W X 2 300W single supply module based on the TI TAS5615.
Since I until recently I also was running the Sure 4X100W module the obvious differences are still fresh in my mind. The 4x100w expired prematurely.
Left with the Meanwell SMPS I picked "Class D Audios" 160W X 2. Now you can get a lot more watts from the same company for about what I paid ($135 + Ship), however I didn't want to buy a new dual voltage linear supply.
I hooked the 160x2 to the very same old slightly modified but relatively inefficient KLH acoustic suspension speakers driven from a pioneer CD directly through 2 50K potentiometers e.g. no preamp. In other words apples were compared to apples.
Now because I haven't run the 160X2 long enough to form a strong opinion, I will report only what I am very certain of.
First, the TI device does not sound way better than the tripath 4x100. No knockout was scored in the initial subjective listening area. Keep in mind that even with the Meanwell cranked to 28V, I am nowhere near the voltage limit of the TI module. I wonder if these things are optimized for a voltage like the tripath modules.
The tripath 4x100 was distinctly brighter than the TI 2x160 although both were too bright for my particular taste and made me yearn for a tone control.
The TI clearly delivered a better low end out of the box. Kick drums and bass transients. I did a double take on a tuba blat which surprised me.
The TI had a much higher input sensitivity than the 4x100 with the pot way below a quarter revolution. Incidentally, read the manual (RTFM) and avoid my stupid error of connecting my single ended drive signal differentially. In other words use + and GND not + and -. My error resulted in an annoying line buzz that I though was originating in my drive signal. Rest assured that when connected correctly the module had no audible hiss or hum, but I did not stick my head in the speaker cabinet for that last db.
I had opportunities to blast both modules while playing Steely Dan. I was vacuuming for my wife (no comments!) which is also an excuse to push out some watts. The Sure 4x100 ran fairly hot in this situation while the TI was slightly warm. Think touching radiator vs touching a pet.
The modules definitely sounded different I think mostly due to imaging. I can say the 4x100 tripath had excellent high frequency imaging. The TI had a more pronounced midrange and fatter bass with no mods.
I am holding off on a judgment since the tripath grew on me over a series of days. Maybe the TI will do the same. I still wonder if these modules actually change sound or do our brains rewire themselves to adapt to the tonal balance. I am increasingly convinced that with DSPs and recording studio effects, sound systems are becoming the equivalent of tinted glasses with people preferring lots of different shades e.g tube sound, class D, class a ETC. 🙂
What modifications had you made to the Sure amp?
On a slightly different note does anybody have any experience with the oxicaps on board the classdaudio? I asked Tom at classdaudio about replacing input caps and he seemed to think that the oxicaps were the best thing available for these amps. I'm not sure if he thought this simply because they were the best things that would fit easily on the board, or if he really thought they were the best thing sonically available. Anybody have any experience with these caps or replacing input caps on these boards?
Joe
On a slightly different note does anybody have any experience with the oxicaps on board the classdaudio? I asked Tom at classdaudio about replacing input caps and he seemed to think that the oxicaps were the best thing available for these amps. I'm not sure if he thought this simply because they were the best things that would fit easily on the board, or if he really thought they were the best thing sonically available. Anybody have any experience with these caps or replacing input caps on these boards?
Joe
Hi i just ordered the 120 watt board and should arrive by the end of the week. This is the first neg comment i have read about the treble, i am wondering if it has since calmed down any for you. Others comments have been highs much less edgy than previous amp. My speaks dont need any help in the brightness dept 🙁
Last edited:
With those 28 Volts you'll get about 40 watts of below 1% THD power into 8 ohms and twice that into 4 ohm. I'd guess you'll probably get about 160 watts into 2 ohms too, since the output can still provide the extra current necessary.
Clarification on "too bright".
I should expand on my previous "too bright" comment. Neither the tripath nor the TI had a problem with clarity on the high end. I believe that some of the effect is possibly due to impedance matching to the amps. I know that just a few picofarads of capacitance or 5 feet of shielded cable on the inputs brought both amps into decent balance.
Driving with a low impedance such as a headphone amp made both amps accentuate treble. Series resistance on the inputs reduced the treble.
Not a real problem. What I am really looking for is dynamics and imaging.
Incidentally, the workmanship and to the extent I can see it, the design and layout of the "Class D Audio" module looked great. My sense is that it is not breaking a sweat even with a heatsink less than a third the size of the Sure.
However my real motivation in releasing $135 hard earned dollars from my hand was reliability. So far so good.
The Sure amp I had sounded amazing even though I never got the opportunity to modify it. I defiantly will try the 2X100 module at some point since people see to do better with that version and at < $50 it is a poor mans dream.
Last thing. Forget about repairing the Sure 4X100. I have decades of soldering experience and access to pretty good equipment. The pins, pads, spacings and etches are highly susceptible to breaking, lifting and shorting.
Even if you are a pcb surgeon, devices usually fail in groups. While tripath devices and clones are available, the devices alone usually will cost more than a replacement module. To reduce mystery failures, use a wrist ground when working on these modules. Static can kill in weird ways.
I should expand on my previous "too bright" comment. Neither the tripath nor the TI had a problem with clarity on the high end. I believe that some of the effect is possibly due to impedance matching to the amps. I know that just a few picofarads of capacitance or 5 feet of shielded cable on the inputs brought both amps into decent balance.
Driving with a low impedance such as a headphone amp made both amps accentuate treble. Series resistance on the inputs reduced the treble.
Not a real problem. What I am really looking for is dynamics and imaging.
Incidentally, the workmanship and to the extent I can see it, the design and layout of the "Class D Audio" module looked great. My sense is that it is not breaking a sweat even with a heatsink less than a third the size of the Sure.
However my real motivation in releasing $135 hard earned dollars from my hand was reliability. So far so good.
The Sure amp I had sounded amazing even though I never got the opportunity to modify it. I defiantly will try the 2X100 module at some point since people see to do better with that version and at < $50 it is a poor mans dream.
Last thing. Forget about repairing the Sure 4X100. I have decades of soldering experience and access to pretty good equipment. The pins, pads, spacings and etches are highly susceptible to breaking, lifting and shorting.
Even if you are a pcb surgeon, devices usually fail in groups. While tripath devices and clones are available, the devices alone usually will cost more than a replacement module. To reduce mystery failures, use a wrist ground when working on these modules. Static can kill in weird ways.
When I was originally building mine, with the 18.5v bench PSU I've got, I could run em without a heatsink with no problems.
I know the chip turns itself off when it gets to 155
degrees C, and one side of me wants to see how far you can push the chip sinkless. But the other side says no don't do that it's reckless and you'd be annoyed if you blew the thing up.
From what I know the DKD44 package is similar in both the TAS56xx and the Sure chips. I have managed to solder/desolder the TAS5630 4-5 times off and onto boards. It's a lot easier to do if you don't mind destroying the chip, but if you want to use it again >.< It takes ages to do and you have to be very very careful doing it, but it is possible.
Of course this leaves the PCB very battered and fragile around the area where the chip solders to. I guess it makes it 'easier' to do if you know you can just whip up another board if it comes to it, but it's best not to fry the thing in the first place.😀
I hope you enjoy the sound of yours. I should be getting some decent inductors soon. I've been talking with Ice Components from the Inductor Evaluation thread. Once they arrive I'll post a series of measurements of how the TAS5630 performs.
imaging: TAS5615 160WX2 vs Tripath Sure 100Wx4
Ok, this is what I, a flawed organic meat puppet, think I am hearing through my
possibly tone deaf ears:
The "Class D Audio" images well up to the upper mid-range then it gets a little diffuse in the tweeter side of cymbals and high frequency transients.
However, voices, snare drums, horns etc image very well.
The Sure had great imaging through out the entire audio range. An visual analogy would be looking at a fireworks display and being able to see the exact point and width of each burst. Its almost as if the Sure had tighter damping and control over the tweeters.
Is this output filter related stuff? Could using a power supply up closer to 50V improve the imaging. I suspect so. Wasn't inclined to mention this cause I know moving a speaker to a slightly different position or setting a different e.q can turn things completely around.
Then again, the last real fireworks here WAS the Sure amp. Maybe this is like eulogizing a dead uncle.
Ok, this is what I, a flawed organic meat puppet, think I am hearing through my
possibly tone deaf ears:
The "Class D Audio" images well up to the upper mid-range then it gets a little diffuse in the tweeter side of cymbals and high frequency transients.
However, voices, snare drums, horns etc image very well.
The Sure had great imaging through out the entire audio range. An visual analogy would be looking at a fireworks display and being able to see the exact point and width of each burst. Its almost as if the Sure had tighter damping and control over the tweeters.
Is this output filter related stuff? Could using a power supply up closer to 50V improve the imaging. I suspect so. Wasn't inclined to mention this cause I know moving a speaker to a slightly different position or setting a different e.q can turn things completely around.
Then again, the last real fireworks here WAS the Sure amp. Maybe this is like eulogizing a dead uncle.
- Status
- Not open for further replies.