sfogg Posted November 14, 2007 Share Posted November 14, 2007 To experiment more with my digital output mod on the DCX2496 crossovers I ended up picking up one of the Panny digital receivers. If anyone was looking for a very inexpensive amp to play around with bi/tri-amping this fits the bill. Using the 6 channel input one has access to 6 channels of digital amplification that is rated at 100w/ch but that is at a high distortion level. On the scope 70w out of one channel into 8ohms looked clean but I didn't yet check that on the spectrum analyzer. Because this is a full receiver one can run system volume using the Panny itself after your crossovers. This allows for maximum signal levels through a digital crossover for highest resolution/SNR. I'm going to try it on one speaker fed via S/PDIF from the DCX2496 to compare it against the analog outputs of the crossover into the Teac tripath amp. Just to test its noise level I hooked it up direct to a 44xt/K510 and its noise level depended upon the volume level. With the volume control turned all the way up it had too much hiss but was very quite with the control turned down. I can't image I doubt I will need the volume all the way up considering the power level of the Panny. With the noise reduction mods in the Teac they put out about 9w max into the 44xt/K510 and that is more power then I need. I also want to open up the Panny and see about adding a 3x S/PDIF input to the unit. Shawn Quote Link to comment Share on other sites More sharing options...
gaspr Posted November 14, 2007 Share Posted November 14, 2007 Shawn, looking forward to hearing more about your experiences with the Panny. They have had quite the following over at the AVS forums. Do these not also have signal alignment capabilities?? With a cheap digital crossover ahead of it, might be just the ticket for time aligned khorns... Quote Link to comment Share on other sites More sharing options...
sfogg Posted November 15, 2007 Author Share Posted November 15, 2007 These do have time alignment functions for surrounds and the center channel outputs. The manual is a little confusing about this though so I'm not sure if that will be active on the 6 channel input or not. I will have to hook up a few channels to the scope and alter time alignment on one of the channels to see if the timing between channels differs with the 6 channel input. It also has a bi-amp mode where it uses two channels for the L/R outputs. When in this mode it has a limited time alignment function as well but only for just under 1' of difference which isn't enough for the K'Horn. Should be interesting playing with this. I had it apart on the bench last night checking out signal routing in it. I think it will be straightforward to add 3x S/PDIF inputs to the amp, just need to design a board for that. Measurement wise these have more bandwidth then the Teac Tripaths. The A/Ds run at 96kHz so the amp is pretty flat to about 45kHz and is rolling off above there. And of course the ultrasonic switching noise is obvious on the output of the amp as it is with the Teac's too. Shawn Quote Link to comment Share on other sites More sharing options...
formica Posted November 25, 2007 Share Posted November 25, 2007 They have had quite the following over at the AVS forums. I noticed and I've been curious about these as well. Shawn, i'm also looking forward to hearing more about your measurements and experiments... Wondering how good they "sound".... Quote Link to comment Share on other sites More sharing options...
sfogg Posted November 26, 2007 Author Share Posted November 26, 2007 "i'm also looking forward to hearing more about your measurements and experiments... Wondering how good they "sound"...." Will let you know how it goes. I had to finish up another project which is now done and I will dig into this board for the Panny. As far as measurements what would you like to see? Like most/all digital amps it measures a little funny. From a FR standpoint it looks flat out to around 45kHz and then rolls off from their, at least into an 8 ohm load. The A/Ds run at 96kHz so the rolloff makes sense since the sampling rate in the A/Ds would limit response to 48kHz. I need to check it out more on the FFT and the scope. At lower levels the ultrasonic switching noise is obvious on the scope. Next time it is hooked up I'll take some pictures of that to show how that looks and what happens when you use averaging to show how it changes the SNR of the measurements. Shawn Quote Link to comment Share on other sites More sharing options...
formica Posted November 26, 2007 Share Posted November 26, 2007 Like most/all digital amps it measures a little funny. What causes this? How audible is it? At lower levels the ultrasonic switching noise is obvious on the scope. Next time it is hooked up I'll take some pictures of that to show how that looks and what happens when you use averaging to show how it changes the SNR of the measurements. Sounds like something interesting to compare... esp if we hook them up to high efficiency speakers.... ROb Quote Link to comment Share on other sites More sharing options...
sfogg Posted November 26, 2007 Author Share Posted November 26, 2007 "What causes this? How audible is it?" Most of it is the ultrasonic switching noise which is well above the audible range. Some thing noise like that may slightly heat a tweeter though to change its power compression somewhat. I haven't measured output impedance yet so I don't know how much the amp will interact with a speakers impedance. "Sounds like something interesting to compare... " Yup, a plain low level sine wave is ugly on a digital amp from the noise. With averaging (to bring the signal out of the noise) it looks much more normal though. I'll take some shots of that on the scope and post them. Shawn Quote Link to comment Share on other sites More sharing options...
sfogg Posted November 27, 2007 Author Share Posted November 27, 2007 Few quick measurements. This is a FFT of a 1000hz sine wave input at roughly 1w output. As you can see in the picture the unit is pretty clean on the FFT. There is a little notch at 2k which would be second order harmonic distortion then what looks like a little that would be 15 and 17th order. Load is 8 ohm 100w non-inductive resistor. Quote Link to comment Share on other sites More sharing options...
sfogg Posted November 27, 2007 Author Share Posted November 27, 2007 Same thing but looking at it on the scope. Looks ugly because of the noise contamination. But from looking at the FFT the noise isn't in the audible region and I also took it out to 40kHz and it wasn't showing up there either. (Teac Tripaths have a big spike in noise around 36kHz) Quote Link to comment Share on other sites More sharing options...
sfogg Posted November 27, 2007 Author Share Posted November 27, 2007 Same measurement but using averaging to bring the signal out of the noise. Quote Link to comment Share on other sites More sharing options...
sfogg Posted November 27, 2007 Author Share Posted November 27, 2007 Here is 1kHz again but around 17v max (not RMS) volts out. See 2nd,3rd and 15th and 17th order distortions. Quote Link to comment Share on other sites More sharing options...
sfogg Posted November 27, 2007 Author Share Posted November 27, 2007 Same thing on the scope... Quote Link to comment Share on other sites More sharing options...
sfogg Posted November 27, 2007 Author Share Posted November 27, 2007 As shown in that last scope picture those last two pictures were at 12.1 v RMS output... 18w. Quote Link to comment Share on other sites More sharing options...
sfogg Posted November 27, 2007 Author Share Posted November 27, 2007 35v peak out. Top trace is the whole waveform. Bottom trace is zoomed in on the top trace in the area the top trace is blue peak of the second wave from the left. Quote Link to comment Share on other sites More sharing options...
sfogg Posted November 27, 2007 Author Share Posted November 27, 2007 36.55 v out peak. Very even clipping top and bottom. Again bottom trace is the zoom in of the blue area of the top trace. No averaging in these last two pictures. The output level has increased enough that the signal is above the noise so you don't see the contamination of the sine wave on the scope. Voltage levels were too high to measure this on the FFT without putting on a 10x probe. Shawn Quote Link to comment Share on other sites More sharing options...
formica Posted November 27, 2007 Share Posted November 27, 2007 Even though it's pretty high frequency.... wouldn't that behaviour make it better suited to lower efficiency speakers… where it would be working at higher output levels most of the time? Quote Link to comment Share on other sites More sharing options...
sfogg Posted November 28, 2007 Author Share Posted November 28, 2007 "Even though it's pretty high frequency.... wouldn't that behaviour make it better suited to lower efficiency speakers? where it would be working at higher output levels most of the time?" The noise is still there in that case though. And even then you can still see the effects of the noise at 18w of output, it is only when it is really cranking that you don't see it. Within the audible range, and to at least an octave above, the signal is clean. If the audible signal itself was bad you would see that on the FFT as well as on the averaged readings on the scope. I'll have to swap out cards in my FFT for the 100kHz card I have and see what its output looks like up to there. The dual channel card I have in there now only goes up to 40kHz. FWIW the Panny is cleaner then the Teac Tripath which has a big spike around 36kHz. As far as noise goes a vinyl rig will have more then this and it is within the audible range as well as above it. SACD has increasing noise with frequency. Etc...etc... Ideally it wouldn't be there. But not too much is ideal so you want it to be far enough away from the audio to not cause problems or to cause other equipment to misbehave. Misbehave such as exciting a resonance in a metal tweeter (some ring in the 25ish kHz range), cause amps to oscillate (some can't handle ultrasonic noise well) and so on. Shawn Quote Link to comment Share on other sites More sharing options...
sfogg Posted January 13, 2008 Author Share Posted January 13, 2008 "I also want to open up the Panny and see about adding a 3x S/PDIF input to the unit." Finally had a chance to wire this board into the Panny. Only briefly tested but it works. The Panny now has 3x SPDIF input so it can plug into the modded 3x SPDIF out on the Behringer digital crossover to stay digital all the way through the crossovers and through the Panasonics digital amp. One downside is that there is only 5 channels of amplification available with the 3x SPDIF input right now. The sixth channel is basically going to the line out for the sub out on the Panasonic. I want to see if I can reroute that to get the sixth channel to an amplifier since I need six channels of amplification for the bi-amped L/C/Rs. Board on the bottom is the DSP section of the Panasonic amp, the board above it is the 3x SPDIF input board. I wired the 3 SPDIF inputs to one of the component video inputs on the Panasonic to avoid having to drill any holes in the Panasonic. Shawn Quote Link to comment Share on other sites More sharing options...
sfogg Posted January 13, 2008 Author Share Posted January 13, 2008 Viewed from above.... Quote Link to comment Share on other sites More sharing options...
DrWho Posted January 14, 2008 Share Posted January 14, 2008 Someone has been busy! I'm bummed I didn't see this post earlier...I love all the oscilliscope traces you've been posting. [Y] When you say you were averaging out the noise, are you just throwing on a lowpass or what's the deal there? I guess what I'm asking is what kind of amp topology is the Panny? Is it a tripath implementation or self-oscillating or...??? I love the clipping behavior on these amps...soooo well behaved. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.