RS duet

Posted on July 15, 2012 by Mark K

Paul Kittinger emailed me a number of months back asking if I collaborate with him on a design for the Midwest Audiofest, sponsored by Parts Express. Paul and I have known each other going back over ten years, if my memory serves me correctly, when he came to my home for a speaker DIY. Paul has since moved to Ohio, and I’m here in California.

Paul had a concept, and asked if I would design the crossover. He wanted to design an ML TL based on the RS225 and the RS28F. I had done a project similar, my RS22528A, a number of years ago and was quite familiar with the strengths and weakness of these units. I’d always been impressed with these drivers, and my testing of the new versions (RS225-8 unshielded and the RS28F) showed outstanding performance, regardless of price. As much as I liked my original design, there were a couple of improvements that I thought could be done, but I just didn’t have a reason to build a completely new design.

And along came Paul with his RS22528FMLTL. Of course, he came up with a better name, the duet.

Highly stylized version in black if you build it that way

An 8″ driver offers notably improved low end performance over the typical 7″ midwoofer based two way, and an ML-TL alignment allows outstanding bass performance and with control of midrange reflections. However, 8″ drivers need to be crossed quite a bit lower than a similar 7″ driver, ideally in the 1.3-1.8k range, which limits the number of tweeter choices significantly. My testing of the RS28F shows it to be up to the task, handling a crossover in the ~1.3k-1.4k range without difficulty.

Paul emailed me the tower dimensions, and I built a mock up test box. I used a sealed PE box and mated it to a baffle of the exact dimensions of Paul’s tower. For crossover design purposes, this would be sufficient. I measured the drivers on the baffle outdoors, then designed a suitable crossover.

A PDF with Paul’s writeup can be found at the link below. The design turned out quite well, so much so that it was awarded first place in its category at the Midwest Audiofest.

Paul K’s RS duet pdf file

More on the crossover design for the curious.

There are quite a few ways of designing a crossover network. I use Praxis, but no reason you couldn’t use SE, omnimic, ARTA or any number of the modern PC based software measurement systems.

Step one was to use a measurement point 2-3 meters out, and use lspCAD to target an eighth order LR curve in the region of the crossover. I try to get as high as possible and measure outdoors.  You can see my test box in the picture below. The baffle is identical to Paul’s Duet, but the back side isn’t exactly the same. It should be very close to the correct FR and adequate for crossover design. I can usually get usable window of ~7ms after subtracting time of flight.

The topology is a notched fourth order curve. The idea behind this topology is to start with a fourth order LR set of curves, then add an additional reactive component to generate a notch in the stop band. The notch has the effect of having a very steep slope. Then you just adjust the crossover Fc so that the steep slopes overlap and now you have a bit of a hybrid crossover that is overall fourth order, but in the region around Fc, approximately eighth order. Probably the easiest way to understand this crossover is through a series of graphs.

First, consider just a fourth order LR.

 

Then, add the reactive component and you end up with notches in the stopband. As noted by the blue comment, you have to use the optimizer to adjust the region of Fc to target an eighth order topology.

The beauty of this approach is that it completely suppresses the rather large and broad breakup of the raw RS225. Below you’ll see a set of raw curves for the two drivers. The region starting at 3k and peaking at around 6-7k needs to be suppressed aggressively.

Alternatively, a more typical fourth order LR with a notch could be used. The relative merits of this crossover have been discussed previously elsewhere, so I won’t go into it here. See my write up for the RS22528A, do a google search for old forum posts on the PE and HTguide boards.

Step two is to build the actual physical crossover and remeasure. Occasionally, the resultant FR curve is spot on. I target flat initially. If it’s not flat, then my job at this point is to figure out why not. I’ll typically go back to lspCAD and look at the topology, and adjust individual component values to engineer corrections. Then I’ll make component changes and remeasure over and over, one by one until I get the flattest possible curve.

Usually flat sounds pretty good. In this case though, when I emailed the preliminary crossover to Paul, he felt it was too forward. This leads to that age old, endless debate, “what is the right target for a speaker’s FR curve?” In general, I target flat, but occasionally flat doesn’t sound right. I had Paul mail me a tweeter and the crossover, just to verify that the tweeter was not the problem. I measured the FR of the crossover, as well as the distortion performance of Paul’s RS28F. Both of these measured fine. Although I can’t prove it conclusively, I suspect that the low crossover causes the tweeter power response to bloom quite a bit lower than we are used to. That is, the tweeter’s very wide power response causes quite a bit of off axis energy in the range of 1.3-3k compared with a more traditional crossover at 2-3k. It’s just not what we’re used to listening to.

So, it was time to engineer a dip. The easy part of this is that placing an amp side resistor on the tweeter leg attenuates the tweeter curve. Unlike a tweeter side resistor, placing the resistor on the amp side causes relatively more attenuation as you go down in frequency. This turns out to be perfect to engineer a modest dip which is most prominent immediately after Fc. In essence, a single resistor causes what is effectively a “BBC dip.” We can argue about the BBC dip, and what it really is, another time. I use the term because most folks have an idea of what it is.

If the dip bothers you, or you think it might bother you, just build the crossover with the resistor, and then use a switch to short out the resistor. When shorted, the response will be nominally flat, but leaving the switch open will leave the dip in. So you can have it both ways.

 

 

 

Posted in Completed Projects | Leave a comment

More UE3

Posted on February 12, 2012 by Mark K

While the athlon 630 seemed to have the computing horsepower to run UE3, opening a window-doing much of anything that stressed the graphics system-seemed to cause audible pops and crackles. If the computer screen saver kicks in, then you tap the mouse to  pick another song-well, that might cause some unpleasant extraneous noise.

The motherboard integrated video is a true dog. While fine for office graphics, it uses system memory and resources. So, I ordered a Sapphire 6570 from Newegg. The Sapphire isn’t really a “gaming card” but at least it has 1 gb of dedicated memory and it’s nice enough for the kids and to play HD video and audio. The good news is that it seems to make a noticeable difference in the performance of UE3. The processing time has dropped to about 6-7ms and opening windows and doing other tasks doesn’t seem to cause audible glitches. You do have to turn off the screen draw in the SE preferences page. With this setup, UE3 seems to work without any glitches.

Over the last week I’ve been playing with the prior pc setup, same as listed previously, but with the new card. I don’t have a raw set of drivers in a box currently available. (That’s the next step.) So, what I did was just use an already build RS22528A. I then took an ideal “full range” driver, generated in SE and set it as the woofer file. A praxis display of this “full range” driver is below.

You might say this is a trivial example. But really, what’s the difference between this and any other curve. The point is to further refine the setup, make sure the levels and gains are ok, and to do some more listening for distortion.

So far, it works very well. In the current setup with win7 and spdif passthru there are only two digital volume controls. There is an application volume control on the foobar. It’s supposedly a decent 24 bit dithered volume control. Initially I set it to 100%. The input levels don’t seem to distort with this setting. There is also the volume control on UE3. Curiously, I can only set the UE3 volume control to 60% or the output will show distortion. This is a curious result. UE3 shouldn’t apply any gain to equalize the curve above. It’s set to equalize the region between 100 hz and 8k. but there is no gain involved in equalizing the above curve. So why can’t I run the volume at 100%?

Anyway, it begs the question. Should I have the volume slider in UE3 set to max, and use the application slider to adjust the volume, or should I set the application volume to max and adjust the volume with UE3. And, should I get the passive volume control, like the MVC

I haven’t had a chance to try UE3 with the llano/fm1 based system, but the above result bodes well. The llano/fm1 system has a gpu that’s somewhere in the ballpark of a 6570 and  a processor that’s in the athlon range. Yes, the gpu and cpu have to share some resources, but the memory and bus is faster.

Why is that good? Well, if you’re building a new pc as an HTPC, the FM1 based ones are a good option. Low power consumption, well equipped motherboards, with onboard video that’s up to the task for HTPC. What I’m not sure of is how many channels can be driven. At this point I have to build a new system and measure it so I can run UE3 and see what

Posted in Uncategorized | Leave a comment

UE3 has arrived!!! Week 1 testing impressions

Posted on January 16, 2012 by Mark K

Well, I finally got a copy of UE3. It’s worth detailing the current PC I’m testing it on.

 

The current PC candidate–

It’s an Athlon II x4 630  processor on an MSI K9N6PGM2-V2  mainboard. It has 4GB of ddr2 with nvdia graphics (GF7025) and a recent vintage samsung 1gb HDD (back when they were $50!!). It’s a reasonably modern motherboard with a bios date of 11/10. It’s the most modern amd mb I could fine that was budget and allowed me to use ddr2. I wanted to reuse my 4GB of legacy ddr2 (this is the “new” kids PC). In retrospect, given how ridiculously cheap ddr3 is now, that was a mistake. Anyway, it represents a kind of low end. It’s an older processor and on board video but still reasonable processing power. This was made from parts of my old main PC that has become my kids’ PC. It also has a delta 410 and Xonar sound card inside. The OS is a new copy of Win 7 home premium 64 bit.

 

The actual setup gritty details–

It’s also worth taking a moment to describe the setup for UE3. It’s a little convoluted, no pun intended. UE3 was not designed to work directly with a software player, like foobar or JRiver. Rather it was meant as a PC based xover with the sound card inputs as analog inputs from a CD player into the soundcard. For whatever technical reason, UE will not accept a direct software audiobitstream. If you want to use a software player, you have to route it out of one sound card via analogue out and then back through the analogue in of another soundcard associated with UE. That’s right–two separate soundcards. The problem is that there is an extra DAC–>ADC conversion in those two separate soundcards. This is not very audiophile.

There is an alternative. Most motherboards have an SPDIF digital stream out. This out is actually usually in two places. Most of the time there is an SPDIF out on the back of the PC on the sound out header. However, the SPDIF signal comes in an optical wrapper, and the sound card UE3 uses has a coaxial wrapper. You can buy optical to RCA converters, but most motherboards have a TTL level SPDIF signal out that’s meant to go into the graphics card or other sound device. However, there is no reason you can’t use this pinout with an adapter plate like this one.

http://www.amazon.com/SPDIF-RCA-Plate-Cable-Bracket/dp/B0035J9Z7W/ref=sr_1_1?ie=UTF8&qid=1326728114&sr=8-1

If you’re wondering what the bracket is doing, well, very little. However, you can’t just connect the output of the TTL level pinout on your mainboard to an RCA female plug. The signalling voltage is different. See Rod Elliot’s appendix at the bottom of his project page. Does Rod have a day job, or is he independently wealthy?

(To be more explicit, I did buy and am using the header in the amazon listing above. However, you can build one yourself if you want from the info at Elliot Sound Products.)

http://sound.westhost.com/project85.htm

So what you have is the following.

digital SPDIF out via Motherboard–>spdif header–>outboard RCA patch cable–> coaxial SPDIF in on the back of the 410. A bit circuitous, but it does avoid an unnecessary set of DAC/ADC conversions.

Step one is to route the output of foobar into the onboard spdif output. Then a single rca patch cable on the back from the spdif output header into the delta 410. Even though the 410 is a legacy card, there are current drivers available. Start foobar, and the M-audio mixer showa a nice digital signal on the spdif input channel. The application is set up to control the volume, and the mixer features of the M-audio are disabled. To test this part of the setup, I routed the 410 output directly from the input. It works flawlessly. In this step, UE3 isn’t used. It’s just a test to see that the spdif routing via patch cable works correctly, and it does. Step two is to reroute the 410 so that the inputs and outputs are under software control, i.e. UE3 and see how it all works.

Now that we know foobar correctly routes a signal into the delta 410, the real question, “How does UE3 work?”

On a very preliminary weekend review, it works well. I’m using a number of flac rips of my CD collection. So this means all the files are 16bit/44.1k. The spdif in are correctly routed to inputs 1 and 3, (front right and front left). I used the sample 2×2 stereo file in UE3 and just hooked it up to a speaker. I individually listened to each of the output channels, L tweeter, L woofer, R tweeter, R woofer (i.e. delta 410 outputs 1-4 and not necessarily in that order). Yes, yes, the frequency response is screwy since I used BR’s sample FR with my speaker, but the point was to see if this works correctly, and indeed it does.

The computer shows ~15% cpu utilization and works flawlessly-assuming you don’t do certain tasks while using UE3. Curiously, trying to resize the UE3 window will causes audible clicks and pops, and if sometimes will actually corrupt the filter permantly. That is, if you resize the window enough, the audio will become choppy and even if you stop resizing, the audio continues choppy. The only way to recover is to stop and restart UE3. Somehow the realtime filter becomes corrupted. Even more curiously, I can resize and open almost any other application and the effect is much less severe.

It does seem to work fine if you don’t do other things on the PC. I don’t think it’s a cpu issue. I think the hangups have more to do with drivers, the system bus and video. The on board video is pretty weak and simple screen redraws place some stress on the cpu, or at least interrupts that hang the cpu. I ended up buying a 6770 card for my i5 and I think I’ll buy a 6570 card for this machine. Not because UE3 needs a graphics card, but to offload the cpu a little bit. The onboard video has to borrow memory and cpu resources across the bus, so this probably hampers things. Plus, the PC does not have an HDMI out, so it needs a video upgrade anyway. This PC will also get an SSD. Why do 10 year olds need SSD’s you might ask? Well, actually I want the 1gb Samsung hard drive as a network music HDD. I’d by one, but they’re not $50 anymore, they’re $150. So, it turns out to be cheaper to by a 64gb SSD (enough for my kids’ pc for now) and swipe the 1gb drive. It’s a crazy world we live in…

So far though it looks like the delta 410 works well at least in a simple stereo configuration with an athlon 630 based system. As I posted before, I have a couple other systems I might try this on. My wife’s PC is one. It uses an A6-3500 on an FM1 mainboard with 8gb and an SSD. This is basically the same/similar Athlon II core (well, 3 of the 4 cores anyway) with a much faster mainboard bus and video. So it will be interesting to see if I can surf the web and resize UE3 while listening.

If you’re interested, I tried selecting the Xonar card and also the usb amp (uses standard usb class 1 audio drivers). It’s a no go. SE gives an “unsupported audio” something something error. It appears for now, that only cards officially recommended by BR are useable, and even the delta 410 only works “unofficially” but it seems to work.

Here’s a screenshot. Click on the image to get a 1900×1000 full size jpeg.

More to come in weeks ahead.

 

 

 

Posted in Digital Music | 2 Comments

Subwoofer excursion

Posted on January 2, 2012 by Mark K

This post is a result of some musings on the PE board about subwoofer excursion. I was convinced that the RS315HF could be overdriven in a 1.8 cu ft box and the Dayton SA240 plate amp. Others weren’t so sure. Turns out Pete was probably more right than I was, as the amp “theoretically” can drive the sub to its max excursion, but not much more.

When thinking of amps and subs, you have to keep track of a couple of things. “Watts” gets thrown around all the time, but it’s also important to keep track of the maximum voltage and current that an amp can deliver, the listed xmax as well as xmech, and the box size. As the box size has effects on current and power driven, as well as excursion.

So let’s take the SA240. Taken from the spec sheet–

Rated Power Output: 150 watts RMS into 8 ohms @ 0.1% THD

240 watts RMS into 4 ohms @ < 1.0% THD

*Based on one-third power duty cycle

The amp is claiming to deliver 240 watts across 4 ohms. This comes out to a peak voltage of 44 volts and 11 amps. Note also the duty cycle is 33%. Meaning it can’t deliver this voltage and load continuously.  This means the amp will engage in some sort of thermal shutdown if driven to it’s maximum for any length of time. I’d be curious to know what the maximum continuous power rating is, but it’s not given.

Anyway, lets but the RS315hf into a sealed 1.8 cu ft box and see what happens. Below you’ll see the excursion plot. Yes, Pete’s right, we’re not driving the unit beyond xmax. These graphs are courtesy of Unibox.

So we’re done. Everything’s cool right. Pete’s right. Well, yes, but it’s still helpful to look at some other graphs. Below, you’ll see the voltage and current draw graphs for 240 watts of power.

Yes, I know the voltage plot is a boring, constant line. I put it there to make the point that amplifiers are constant voltage sources. When 240 watts of nominal power is placed across the drivers leads, the current varies with the impedance of  the driver. So the actual power dissapated varies akin to the current graph.

In a sub application, the driver would be rolled off above 80, so the current, i.e. power peak would be limited. But look what happens below 20 hz. The current is rising to the maximum limites of the plate amp, which can only deliver this 1/3 of the time anyway.

So technically we are not exceeding xmax. However, the amp and driver are at their limits when pushed hard and there may be some distortion, more from the amp than the driver. Most drivers can exceed xmax by a fair margin, this “xmech” which is when the excursion exceeds the mechanical limits of the driver, not just an arbitrary distortion figure. Usually xmax is defined as the displacement that causes 10% THD, but there are a number of definitions. Exceeding xmax just causes some extra distortion. Exceeding xmech may permanently damage the driver.

It’s informative to look at some other scenarios. Let’s increase the box volume to 3 cubic feet. The air spring is more compliant, so it’s much easier to drive the unit into over excursion.

Also, the current required goes down a little at the low end. It doesn’t take as much current to push against the springier air, so to speak.It’s not much, an amp or so, but it does reduce the amp load.

Now let’s try the opposite, and put it it a tiny 30 liter box.

The air spring is too stiff to allow 240 watts to drive the unit to xmax. The problem is this-

The amp might very well engage in a thermal shutdown if there is any sub 20 hz content even though it’s not putting out much bass in this region–Now granted, there may not be much sub 20 hz content, but if there is, it will be exceedingly hard on the amp. Remember, the amp has a 1/3 duty cycle. so it can’t sustain these 10 amp loads. Just by looking at the excursion curves you’d think everything was fine, but it may not be.

 

 

 

 

Posted in Design Concepts and Tools | Leave a comment

Waiting on UE3

Posted on November 30, 2011 by Mark K

Looking forward to UE. Recently the question came up, “how much computer horsepower do you need?

Good question. I’ve built 2 new computers recently, and will probably build another shortly. I had to update my kids’ pc. The oldest is 10, so they don’t need much horsepower. (Sorry, no FPS games yet. High end graphics not an option!) I ended up reusing my older pc parts, an athlon 630 quad core, 4gb of ddr2, a samsung 1gb (back when they were $50 bucks. Whatever happened to those days…?) This represents a low end platform. Absolutely fine for office and Dragon, internet surfing. Fine for my kids. Interesting to see if UE3 runs on this well. I’m not expecting 10 channels of 192khz audio, but a much older dual core athlon runs 2×3 stereo through the delta 410 without a hiccup, so this should be more than adequate for a stereo setup 2×4. Question is will it support 2×4 at 96 or 192k?

The next system is my new system. It’s a core i5 overclocked to 4200, 8gb of ddr1600 memory, crucial m4 SSD. I need the speed for pshop and lightroom, but it should have no problem with an HT UE3 setup. (PS I migrated SE v 17 to the new win 7 machine just by copying the Bodzio folder and installing the dongle driver-works without a hitch.)

See the screenshot below. Double click the image to see the full screen shot. I know, Windows experience index is not at all the bomb, but it’s still a ballpark estimate of speed. sorry about the pedestrian graphics. This is with the on board hd3000 video. I don’t game, so I haven’t decided if I want a separate graphics card. Yes, some of pshop filters use the graphics engine. Whether it’s really worth it or not is unclear.

PS yes it’s hot, but intelburn on max settings is brutal. Try it on your machine on max setting with a real time cpu temp monitor and well, see what happens. Do it at your own risk however… The system will run fine, apparently at 4.4 and even boot at 4.8 but isn’t completely bulletproof at those speeds.

Virtually every i5-2500k will oc to 4ghz without difficulty. Even if you aren’t into overclocking, this is an easy one. It would be very unlikely a system like this would have trouble with 10-20 high res audio streams.

The final system-my wife now wants her home office pc updated-will probably be one of the llano chips, maybe an i3. Both very nice and reasonably priced. The i3 has better cpu horsepower, but the llano is a very nice integrated solution. i3 and llano systems are seeing a lot of use in HTPC’s, and based on their architecture, they should do high res stereo bitstreams without a hiccup. Could a quad core llano do 2×4 192k, or more? Don’t know.

 

Posted in Digital Music | Leave a comment

The Mitey Mic II preamp is alive!

Posted on July 10, 2011 by Mark K

I had an old MMII preamp. This was a mic preamp designed by Joe D’Appolito. It had 6dB of gain, but more importantly, had the correct bias voltage for the standard panasonic mic capsules. It died on me a while back. It didn’t matter when I switched to praxis, since praxis has the correct bias voltage.

The problem is that I’m curious about some of the other software out there, like ARTA. I’m trying to set up a new, portable measurement system. The problem is that most sound cards have no bias voltage, or the higher pro level voltage. You can buy one, but I’ve already got an m audio transit, which is a fine card for measurements (48k sampling). So I needed a working preamp. I pulled the MMII and it just turned out to be some loose wires where the 9v battery was soldered to the board. (Uh, it was me that originally soldered it.)

Below you’ll see the transfer function. This is about as flat as you get. You’d have to obsess about parasitic reactance in the cables to get much better.

Why should you care? Hopefully in the next couple of months I’ll have a review of ARTA. I can also get an SE measurement system up and running.

Posted in Design Concepts and Tools | 1 Comment

TB 3″ copper final build

Posted on July 9, 2011 by Mark K

Well, hasn’t the price on this driver and TB in general gone up…? It’s a nice driver, but I’m not sure if they’re priced to move, so to speak…

Anyway, I had mine on sale from the original price. That was a good deal. Now, well, it’s up to you to decide. The final curves in the box shown in the prior post are shown below.

First, an on axis FR curve, normalized to 2.83v. It’s not super sensitive, but about right for a 3″ driver.

So, 78dB sensitive. I ignored the trap for the 5-20k region given the off axis response-if anything a full range driver has less high frequency energy in it’s power response. Listen 15-30 degrees off axis if you wish. See the power response below.

What kind of low frequency response do you get? Below you’ll see a “flex curve,” which is praxis’ way of adding low frequency info after the reflection free interval into the gated FR above the first reflection. It gives you a rough idea of in room bass response. You have to smooth it, so you should only use it to make general impressions of relative bass, mid and highs.

There is a 6-12 db/oct rolloff starting roughly at 150 hz. That is, it needs a sub for anything serious.

What about distortion?

If you crossed it over at 150 hz, then the most you could get out of these cleanly is in the low 90dB/1m peaks. Pretty good for a bedroom or small apartment speaker. Pushing it 4 dB higher gives the following graph

You can see the higher order distortion start to creep in, just getting to -60dB. That’s about the most you can get.

So, how do they sound? Well, I’ve got them in my daughter’s bedroom as an alternative to the typical pathetic boom box setup. For voice, small acoustic ensembles in a small room, they work very well. They don’t have the “air” that comes with a small tweeter, but the power response is pretty forgiving and they don’t have that forward sounding tendency that the typical 2 way 4LR/1″ dome has.

Let me know if you want the details of the box and the crossover, which is just a 3 component trap. The LCR trap combines with the natural curve and the baffle response of the box to give almost complete baffle step compensation. It’s not too much-mostly because of the significant roll off at 150 hz.

If you’re wondering why a simple trap works, the graph below explains it. The natural response of this driver is not flat. However, placing the driver on the baffle results in the following curve.

The net effect of the natural driver response and the baffle is a low Q peak centered between 1-2k. All that was necessary was a trap to flatten this.

Posted in Uncategorized | 3 Comments

More Topping

Posted on January 29, 2011 by Mark K

Here’s a graph of Vo/Vin for the Topping TP30 and a more traditional budget amp, an older Audiosource amp. Nothing too special, but a more traditional design. Under a straight resistive load, both amps look reasonably flat. However, under typical speaker loads, i.e. my RS22528a and a full range TB 3″ driver with a parallel notch, the frequency response variation seems quite pronounced. Maybe the Topping has a nice tube sound…

Below, there are two grouping of curves.

First, the relatively flat black, aqua and red at about 4 dB represent the FR of the audiosource under a 5.6 ohm resistive load, and the RS225 and TB respectively.

The second set consists of the blue, orange and brown and are all of the Topping. The blue line at 3 dB is the Topping with a 5.6 ohm load. The orange and brown represent the Topping TP30 FR under the loading of the RS22528A and the Tang Band, respectively.

Ignore the relative gain levels. What’s important is that an amp should have a frequency response that is a flat line.

Seems like the Topping is not acting like a low impedance voltage source…

A google search show references to class D amps having some output impedance variation with frequency, and that looks to be the case.

Posted in Digital Music | 3 Comments

Topping TP30

Posted on January 19, 2011 by Mark K

So, I’ve been playing with the Topping TP30 USB DAC and integrated T amp. You can find them at a number of ebay sites for under $100 shipped.

What do you get? You get an integrated low power T amp based on the tripath 2024 chip, along with a usb DAC based on the ubiquitous TI2704 and a single 12meg crystal. The specs as best as I can tell are 16/48 and the usb uses standard windows audio class 1 drivers. The Tripaths are well regarded as an efficient, but generally low power solution and the TI27xx series is considered a pretty solid usb DAC solution in the midrange for 16/48. Below you’ll find a higher resolution image if you’re so inclined. Just click on it…

Question is, how good is it? Is it relegated to the office, streaming background classical, or can it really play?

Well, a couple tests first.

Let’s look at FR using praxis and running 96k through the m audiophile card.

Well, not bad except for the rising high end. The output is measured into a high impedance load, with the input adjusted to give Vp of 2.85v. Yes, somewhat arbitrary, but repeat measurements at different gains show the same behavior.

(UPDATE-Note that the FR into a high impedance load doesn’t change based on level, but it does change based on load. See the FR measurements into multiple loads for a number of T amps at Michael Mardis’ site. Under load the FR flattens. You don’t have to worry about oscillation. However, into a 4 ohm load there is actually extra roll off in some of the T amps. So, I’ll repeat the FR into a 4 ohm load and into a couple of speakers to see what the real world FR would be.)

What about distortion? Below you’ll see two graphs. One is a measurement with the volume set all the way down, as the control, and the other with the gain adjusted to 5.6 Vp.

Is this good? Well, notice that the spikes are multiples of 60 Hz. The spikes turn out to be multiples of 60 Hz line frequency. It’s a problem with the power supply and filtering. The switching PS is very noisy. The distortion is improved by locating the unit as far away as the power supply as possible, but I suspect the distortion would be improved with a better PS. I’ll retest it with a car battery-I suspect the numbers will look much better.

How about peak V and distortion? Well, let’s look at it a different way, with a wavelet. This is a nice way of doing it because it’s not likely to destroy the unit, since the level can be gradually increased such that only a single cycle will clip.

Praxis allows you to window a sine wave. The stimulus is a blackman windowed 7 cycle sine wave of 1k. You can see the center peak wave clips at 12 volts, the power supply rail when driving a high load. The red curve is with a 5.6 ohm dummy load. You can see the rail/clipping sag to 11 volts. The moral of the story is that you have about 11 volts and 1.5-2 amps of usable voltage and current.

Physically it’s a good looking, well built unit. The volume control is quite smooth and linear, and it is very quiet unit in operation. It does have a bit of turn on thud though. Did I mention it’s only $94 shipped… Windows recognizes it instantly as a usb audio device and it becomes the primary sound output device. When unplugged, windows drops back to the prior sound device, so there don’t appear to be any software glitches. As an alternative to on board sound, with the right speakers (i.e. relatively high sensitivity, low impedance-I think a 4 ohm mtm, 90 dB sensitive, would be perfect) it’s an attractive budget solution. It definitely needs a better power supply. Or, just run it off a battery.

Feel free to post comments on what other tests you might like, what you think of these so far.

Posted in Digital Music | 24 Comments

W3-1878 FR curves

Posted on January 8, 2011 by Mark K

Below you’ll see the FR curves for the Tang Band 3″ full range driver. Both curves are with the driver in the the box, measured at 0.5m but normalized to 1m/2.83v so the sensitivity is 80dB. There is 5-6dB of net baffle step correction. Two curves are shown. The black curve shows the response with a simple parallel LCR trap. The red curve adds a LR shelf. It appears the LR shelf to roll off the rise above 8k isn’t necessary, but makes the curve look better. I’m not bothering with the shelf in the final design. Full range drivers roll off anyway off axis much more so than a typical 2 way with 1″ dome, so peaking isn’t really noted. The 15 degree off axis curve for the unit without the shelf looks just like the red curve, so toe the speakers if you think you hear the rise.

Posted in Current Project, Drivers | 8 Comments