bracurrie Posted March 14, 2013 Share Posted March 14, 2013 I am using the miniDSP with an Audio Research SP6B. I discovered that when I changed the input sensitivity of the miniDSP to 1.7mV that I heard more air in the high end. That's good. But I have determined the miniDSP is a major source of noise. I suspect either sloppy DSP work or a problem with the analog circuit at the output. Has anyone experienced the same thing or do any of the EE trained forum members have any thoughts or suggestions? The 2x4 miniDSP in a box is such a cost effective way to get flexible active crossover done. I just wonder if its just not going to cut it for getting that last bit of potential sound quality with my sound system. Maybe I should have passive crossover networks made with the settings that seem to work the best with my speaker components. Jubilee clone bass bins with Crites woofers and Eliptrac 400 horns with Faisal HF200 16ohm drivers. Or maybe a different active crossover. Quote Link to comment Share on other sites More sharing options...
Edgar Posted March 14, 2013 Share Posted March 14, 2013 It would be tough to be so "sloppy" with DSP implementation that it caused excessive noise. Distortion, perhaps; clipping, limit cycles (idle tones), etc., but generally not broadband noise. I suspect that the analog stages are at fault. I looked over the miniDSP site and found no evidence of digital I/O capability -- that would be the easiest way to determine if it is, indeed, a problem in the analog stages. A pity that the folks at miniDSP don't make a SDK (software developer kit). It looks like they're using some flavor of the Motorola/Freescale DSP56K -- I've done enough DSP56K work in the past that I could just about code in my sleep. Quote Link to comment Share on other sites More sharing options...
mark1101 Posted March 14, 2013 Share Posted March 14, 2013 I would contact Dave at FastLane. He swears by the mini-DSP. He could probably help you out. Quote Link to comment Share on other sites More sharing options...
Don Richard Posted March 14, 2013 Share Posted March 14, 2013 It sounds like you have the gain structure wrong - the input sensitivity of the Mini DSP is amplifying noise from the preamp. Try making the Mini DSP less sensitive so that it doesn't amplify the noise as much. Quote Link to comment Share on other sites More sharing options...
bracurrie Posted March 14, 2013 Author Share Posted March 14, 2013 the input sensitivity of the Mini DSP is amplifying noise from the preamp.My choice was .7V or 1.7V. I am using the 1.7V inputs. It was needed to make the SP6 happy as it needed the right load. BTW the preamp has been isolated as a not the source of the noise.[] Quote Link to comment Share on other sites More sharing options...
LarryC Posted March 14, 2013 Share Posted March 14, 2013 My choice was .7mV or 1.7mV. I am using the 1.7mV inputs. It was needed to make the SP6 happy as it needed the right load.Happy in what way? Do I remember you suggesting that the 1.7 v gave the sound more "air"? Perhaps so, which may mean the "sweet spot" in the miniDSPs level control doesn't mate up will with the SP6. Drawing on parallels with preamp circuitry, it sounds like the miniDSP circuitry after its own internal level control is too noisy for the high sensitivity of your amps and K-horn efficiency. Hard to know just what to zero in on here, but perhaps the suggestions to contact the miniDSP designer is as good a place to start as any. Quote Link to comment Share on other sites More sharing options...
NOSValves Posted March 14, 2013 Share Posted March 14, 2013 I am using the miniDSP with an Audio Research SP6B. I discovered that when I changed the input sensitivity of the miniDSP to 1.7mV that I heard more air in the high end. That's good. But I have determined the miniDSP is a major source of noise. I suspect either sloppy DSP work or a problem with the analog circuit at the output. Has anyone experienced the same thing or do any of the EE trained forum members have any thoughts or suggestions? The 2x4 miniDSP in a box is such a cost effective way to get flexible active crossover done. I just wonder if its just not going to cut it for getting that last bit of potential sound quality with my sound system. Maybe I should have passive crossover networks made with the settings that seem to work the best with my speaker components. Jubilee clone bass bins with Crites woofers and Eliptrac 400 horns with Faisal HF200 16ohm drivers. Or maybe a different active crossover. Brad I have a suggestion but I honestly think you would rather not hear it at this stage of the game [] Quote Link to comment Share on other sites More sharing options...
Don Richard Posted March 14, 2013 Share Posted March 14, 2013 the input sensitivity of the Mini DSP is amplifying noise from the preamp. My choice was .7mV or 1.7mV. I am using the 1.7mV inputs. It was needed to make the SP6 happy as it needed the right load. BTW the preamp has been isolated as a not the source of the noise. You said mV and it seems you meant V, that's a 1000:1 ratio. In that case what you need to do is turn the input attenuators on the amplifiers down, or if there aren't any on the amplifiers, use in-line attenuators. There is a thread here that was started 2 days ago you need to read: http://community.klipsch.com/forums/t/172296.aspx Quote Link to comment Share on other sites More sharing options...
bracurrie Posted March 15, 2013 Author Share Posted March 15, 2013 [+o(] the input sensitivity of the Mini DSP is amplifying noise from the preamp. My choice was .7mV or 1.7mV. I am using the 1.7mV inputs. It was needed to make the SP6 happy as it needed the right load. BTW the preamp has been isolated as a not the source of the noise. You said mV and it seems you meant V, that's a 1000:1 ratio. In that case what you need to do is turn the input attenuators on the amplifiers down, or if there aren't any on the amplifiers, use in-line attenuators. There is a thread here that was started 2 days ago you need to read: http://community.klipsch.com/forums/t/172296.aspx Thanks Don. The link was a good discussion. Yes my EE ignorance showing through. It is V. And just to make my ignorance complete, why if the volume knob on the SP6 is at 1 o'clock to get 85dB output from the whole chain would I want to further attenuate the output from the pre amp or the miniDSP? Is it safe to run the pre amp with the volume know at 70% or more of its output range? As you can tell I am the worst kind of amateur, I try to use a little knowledge. But I think I understand the concept, I just don't have measuring tools or the technical knowledge to do the calculations. I discussed this with the tech that repairs my equipment who patiently trys to explain the audio engineering and EE concepts to me. He confirms that gain structure is important to limiting noise and getting the optimum from my equipment. We reviewed the numbers and the only thing he found was that the miniDSP needed to be set at 1.7V because at .7V SP6's main outs were not being loaded optimally. And that adjustment was the equivalent to using attenuators. Maybe I am getting this wrong. Besides the miniDSP when running with nothing connected to it or whether it has the SP6 connected muted or unmuted outputs the same level of noise. Could it be that the gain level from the miniDSP is too high? That doesn't seem logical either as without the gain levels set to near maximum on the miniDSP I do not get enough sound. You have to remember, I am driving a 16ohm compression driver connected to a horn with no attenuation from a passive crossover network. It's ultra sensitive. Probably 120dB or higher. If there is any noise in the chain its going to be audible. I solved the problem for now by attenuating the compression driver by 12dB with a transformer based attenuator with a swamping resistor so the amplifier sees little or no change in load. But I will look again at the gain structure just to be sure since so many of you keep bringing it up as a strong possible source of the noise in my system. Quote Link to comment Share on other sites More sharing options...
bracurrie Posted March 15, 2013 Author Share Posted March 15, 2013 Brad I have a suggestion but I honestly think you would rather not hear it at this stage of the game Your nothing but a tease! Quote Link to comment Share on other sites More sharing options...
Don Richard Posted March 15, 2013 Share Posted March 15, 2013 Besides the miniDSP when running with nothing connected to it or whether it has the SP6 connected muted or unmuted outputs the same level of noise. Could it be that the gain level from the miniDSP is too high? That doesn't seem logical either as without the gain levels set to near maximum on the miniDSP I do not get enough sound. At this point the power amps need to be turned down either by their input level controls or with external attenuators placed between the mini DSP and the power amps until the hiss is reduced or eliminated. Turn the preamp volume knob to whtever it takes to reach the desired listening level. Of course remove the transformer based attenuator between the power amp and the compression driver. Quote Link to comment Share on other sites More sharing options...
bracurrie Posted March 15, 2013 Author Share Posted March 15, 2013 Besides the miniDSP when running with nothing connected to it or whether it has the SP6 connected muted or unmuted outputs the same level of noise. Could it be that the gain level from the miniDSP is too high? That doesn't seem logical either as without the gain levels set to near maximum on the miniDSP I do not get enough sound. At this point the power amps need to be turned down either by their input level controls or with external attenuators placed between the mini DSP and the power amps until the hiss is reduced or eliminated. Turn the preamp volume knob to whtever it takes to reach the desired listening level. Of course remove the transformer based attenuator between the power amp and the compression driver. I am confused by your answer. Thank you, but if I attenuate the output from the miniDSP, say 12dB, I would have to turn the volume up to 90% on the pre amp to get the desired listening level and it would leave the noise of the power amp, albeit the noise of the power amp is less than the noise of the miniDSP. What is wrong with having an attenuator between the amp and compression driver? If I had a passive crossover I likely would have some degree of attenuation. Quote Link to comment Share on other sites More sharing options...
Edgar Posted March 15, 2013 Share Posted March 15, 2013 ... if I attenuate the output from the miniDSP, say 12dB, I would have to turn the volume up to 90% on the pre amp to get the desired listening level There's nothing wrong with that. I have my system set up so that when the volume control is set at 100% the amps will just barely start to clip with the source at maximum level. That way I cannot possibly overdrive my system. ... and it would leave the noise of the power amp, albeit the noise of the power amp is less than the noise of the miniDSP. That is as it should be. Basically, you want as little amplification as possible after the noisiest stage in your signal chain. In your case that is the miniDSP. Otherwise you are also amplifying the noise. What is wrong with having an attenuator between the amp and compression driver? It's fine if what you want to do is heat the room. You paid good money for those Watts, so why not turn them into sound instead of heat? Think about it: with 12 dB attenuation, if you send 16 Watts to the driver, only 1 Watt actually gets there. Quote Link to comment Share on other sites More sharing options...
bracurrie Posted March 15, 2013 Author Share Posted March 15, 2013 It's fine if what you want to do is heat the room. You paid good money for those Watts, so why not turn them into sound instead of heat? Think about it: with 12 dB attenuation, if you send 16 Watts to the driver, only 1 Watt actually gets there. Would you please explain the calculation. I would find it hard to believe I am losing that much. Besides the quite is worth the tradeoff. Quote Link to comment Share on other sites More sharing options...
Don Richard Posted March 16, 2013 Share Posted March 16, 2013 It's fine if what you want to do is heat the room. You paid good money for those Watts, so why not turn them into sound instead of heat? Think about it: with 12 dB attenuation, if you send 16 Watts to the driver, only 1 Watt actually gets there. Would you please explain the calculation. I would find it hard to believe I am losing that much. Besides the quite is worth the tradeoff. One of the benefits of biamping is the removal of the crossover or any other devices from between the power amplifier and the compression driver or woofer driver. Passive level controls between the power amplifier and driver control the output of the driver by reducing power going to the driver. The excess power is dissipated in the form of heat. Once again, remove anything from between the power amp and the speaker and hook the voice coil directly to the power amp. Make sure the proper crossover points are in the Mini DSP. Turn everything on and listen for hiss with no signal present. Turn the power amp down until the hiss is just inaudible. Play some music and turn the preamp's volume control to a low listening level. If the low frequencies and high frequencies are balanced that should do it. If not, adjust the outputs on the Mini DSP until they are. Don't be concerned about the preamp volume knob setting if it is higher than before. If it won't get loud enough with the volume control wide open, then set the sensitivity of the Mini DSP to .7V. Quote Link to comment Share on other sites More sharing options...
Edgar Posted March 16, 2013 Share Posted March 16, 2013 Would you please explain the calculation. I would find it hard to believe I am losing that much. Besides the quite is worth the tradeoff. DeciBels (dB) are a measure of power ratio. The equation is as follows: dB = 10*log10(power out / power in) You are using 12 dB of attenuation, or "-12 dB". Let us assume that "power in" equals 16 Watts: -12 = 10*log10(power out / 16) -1.2 = log10(power out / 16) 10^(-1.2) = power out / 16 16 * 10^(-1.2) = power out 1 = power out Quote Link to comment Share on other sites More sharing options...
bracurrie Posted March 16, 2013 Author Share Posted March 16, 2013 Would you please explain the calculation. I would find it hard to believe I am losing that much. Besides the quite is worth the tradeoff. DeciBels (dB) are a measure of power ratio. The equation is as follows: dB = 10*log10(power out / power in) You are using 12 dB of attenuation, or "-12 dB". Let us assume that "power in" equals 16 Watts: -12 = 10*log10(power out / 16) -1.2 = log10(power out / 16) 10^(-1.2) = power out / 16 16 * 10^(-1.2) = power out 1 = power out Many thanks for the detail. There are some among us who would simply say such a thing with little basis. If the power needed to drive 90dB of sound was 2watts before attenuation then the power needed would now be 32watts, correct?I still maintain that while I am wasting some power, it's worth it to have some quiet. Quote Link to comment Share on other sites More sharing options...
DrWho Posted March 16, 2013 Share Posted March 16, 2013 What is the amplifier and speaker you're connected to? Do you have their datasheets available? Quote Link to comment Share on other sites More sharing options...
Edgar Posted March 16, 2013 Share Posted March 16, 2013 If the power needed to drive 90dB of sound was 2watts before attenuation then the power needed would now be 32watts, correct? Correct. I still maintain that while I am wasting some power, it's worth it to have some quiet. Unless your amplifier noise floor is objectionably high, you would be better off putting the 12 dB attenuation ahead of the power amp than behind it. Quote Link to comment Share on other sites More sharing options...
bracurrie Posted March 16, 2013 Author Share Posted March 16, 2013 What is the amplifier and speaker you're connected to? Do you have their datasheets available? I am biamping an active crossover 2 way speaker system. The HF is a Eliptrac 400 horn with a Faital HF200 16 ohm driver. http://www.critesspeakers.com/HF200_datasheet.pdf The amp driving the HF is a Manley Labs 'Euro' Stereo 100 that uses the GY50 tubes. Specs:Max. power @ 1.5% THD: 95 watts Input sensitivity: 800mV Output load impedance: 8 Ohm Freq. Response +-1 dB: 10 Hz-40 KHz S/N ratio: 80dB 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.