Phonoclone_noise.wav
24-96 recording of the phonoclone output noise. It is recorded at unity gain, with the DL103 connected to the input. The phonoclone gain is 58 dB. No, you aren't supposed to listen to it, but feed it into your favorite analysis software to pop out a power spectrum to compare against other phono stages, your build of the phonoclone, etc.
/R
24-96 recording of the phonoclone output noise. It is recorded at unity gain, with the DL103 connected to the input. The phonoclone gain is 58 dB. No, you aren't supposed to listen to it, but feed it into your favorite analysis software to pop out a power spectrum to compare against other phono stages, your build of the phonoclone, etc.
/R
The dB scale of the Fourier power spectrum plugin in Audacity is referenced to 0.05 Vrms = 0 dB
Conventionally it is referenced to 1 V rms = 0 dB.
This means that the dB data in Audacity is shifted up by 26 dB compared to most other software.
Attached is the phonoclone noise data, linked above, plotted using the conventional reference. The measurement threshold is about -120 dB.
The noise at 1 kHz is -103 dB. Input refered, gain = 58 dB, is -161 dB. The datasheet noise for the op27 is -167 dB. That these numbers match, with the measured figure slightly higher, tells us that everything is ticking over as it should.
Conventionally it is referenced to 1 V rms = 0 dB.
This means that the dB data in Audacity is shifted up by 26 dB compared to most other software.
Attached is the phonoclone noise data, linked above, plotted using the conventional reference. The measurement threshold is about -120 dB.
The noise at 1 kHz is -103 dB. Input refered, gain = 58 dB, is -161 dB. The datasheet noise for the op27 is -167 dB. That these numbers match, with the measured figure slightly higher, tells us that everything is ticking over as it should.
Attachments
MJL21193 said:
Two remarks.
First, to make level referencing easier it is useful to record a commonly available reference signal, like the 300Hz 0dB track of the HFN test record, and then refer the noise floor to it.
The actual measurement is easy to do in Audition. First reduce the sample rate to 44k1, thus limiting the bandwidth to 20kHz (which is standard for measurements). Then select a few seconds of reference track, and do Analyse > Statistics. Note the average RMS over that segment. Then to the same for a segment of noise. The difference between the results is a good measure for the unweighted signal to noise ratio, provided the reference track corresponded to 7cm/s lateral 0dB LP level.
If you want to have it weighted you first have to apply the weighting filter(s) to the noise segment.
Werner said:
Hi Werner,
A good tip, thanks.
I did this, results attached. First window is with input to the preamp grounded. The second is with a track playing. Record level is unchanged.
Richard and I have been discussing this outside this thread and I don't want to clutter this thread up with my work, so I won't post anymore about mine, but feel free to PM me if you have any comments.
I have a Phonoclone 3 build of my own planned - a reference for comparison to de-bug my own design.
I applaud Richards work here - it's truly difficult to successfully design a preamp as quiet as his.
Attachments
Bit of an OT post I'm afraid, but worth posting I think.
Have a look at this thread:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=137089
I tried this alignment method tonight and got very good results.
On-topic bit:
Given that the P3 is pretty revealing, its worthwhile knowing that everything else is set up correctly and not limiting the performance. Mine is giving great sounds now.
For those who are about to build, don't forget not to rush to judgement. Allow a good 2 weeks continuously powered on and an hour ot 2 a day of music playing before really assessing it.
Also, watch those caps in C3. Has anyone out there got a few nichicon muse ES in the right value that they would like to part with? Even temporarily - I would like to try them - if they are sitting on your shelf doing nothing if you post them to me, I will try them and post them back to you at the end. Or where can I buy them? Anyone got a favourite vendor?
While I am very lucky to have gotten a pair of the blackgate Nx I would like to put my VSPS back together with something respectable in that position. Is there anything else that works as good as the blackgates in there?
Fran
Have a look at this thread:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=137089
I tried this alignment method tonight and got very good results.
On-topic bit:
Given that the P3 is pretty revealing, its worthwhile knowing that everything else is set up correctly and not limiting the performance. Mine is giving great sounds now.
For those who are about to build, don't forget not to rush to judgement. Allow a good 2 weeks continuously powered on and an hour ot 2 a day of music playing before really assessing it.
Also, watch those caps in C3. Has anyone out there got a few nichicon muse ES in the right value that they would like to part with? Even temporarily - I would like to try them - if they are sitting on your shelf doing nothing if you post them to me, I will try them and post them back to you at the end. Or where can I buy them? Anyone got a favourite vendor?
While I am very lucky to have gotten a pair of the blackgate Nx I would like to put my VSPS back together with something respectable in that position. Is there anything else that works as good as the blackgates in there?
Fran
I think size does matter - the value of the cap in conjunction with the input impedance of the following stage (ie your preamp) determines what the frequency cut off is. RJM sent me this a while back and its very useful:
30 Ohms 1000 uF
300 Ohms 100 uF
600 Ohms 47 uF
1k 22 uF
2.2k 10 uF
4.7k 4.7 uF
10k 2.2 uF
22k 1 uF
47k 0.47 uF
100k 0.22 uF
220k 0.1 uF
500k 0.047 uF
1M 0.022 uF
In short I don't think bigger is any harm at all. Just you can get away with lower is all. I think they must be non-polar though.
Fran
30 Ohms 1000 uF
300 Ohms 100 uF
600 Ohms 47 uF
1k 22 uF
2.2k 10 uF
4.7k 4.7 uF
10k 2.2 uF
22k 1 uF
47k 0.47 uF
100k 0.22 uF
220k 0.1 uF
500k 0.047 uF
1M 0.022 uF
In short I don't think bigger is any harm at all. Just you can get away with lower is all. I think they must be non-polar though.
Fran
I uploaded a couple of new phonoclone 3 recordings this evening, 24 bit 96 kHz .flac format
Led Zeppelin : Ramble On
Carpenters : Ticket to Ride
The files are about 90 MB each.
Led Zeppelin : Ramble On
Carpenters : Ticket to Ride
The files are about 90 MB each.
Just to let you know I have been playing with some different caps for C3 over the last while. First up were some GE caps, 2 x 0.68uf in parallel giving 1.36uF total.
So I need to give them a little time to settle, but off the blocks they sound pretty indistinguishable from the blackgates. If this continues it would be very good news for a whole load of equipment!
Oh yeah, the best bit: They only cost about 30c from madisound!
linky
Fran
So I need to give them a little time to settle, but off the blocks they sound pretty indistinguishable from the blackgates. If this continues it would be very good news for a whole load of equipment!
Oh yeah, the best bit: They only cost about 30c from madisound!
linky
Fran
Onkyo SE-200PCI Soundcard
Some people were asking, so:
About the sound card I'm using to record and measure the phonoclone.
Onkyo SE-200PCI
http://www.jp.onkyo.com/wavio/se_200pci/index.htm
It's the big sister of the output-only SE-90PCI (which I also have used and found to be very good). The SE-200 adds 24-96 recording capability, digital input, and 7.1 channel output. There is even a "limited edition" version of the SE-200PCI with a copper shield and custom "audio" capacitors.
All cards will play back 24 bit 192 kHz data.
I've checked all the modes with Rightmark Audio Analyzer and apart from a slightly annoying channel imbalance of 0.3 dB it tests very well.
The 2448 test is also provided from Onkyo directly, you can see the results here:
http://av.watch.impress.co.jp/docs/20061211/dal262d.htm
My copy matches that performance, within error. (See attached reports)
It also sounds really good as a digital source, and I'd recommend it to anyone who might be able to get their hands on one (it's Japan only). The only real annoyance is the drivers. It is not nearly as convenient as it should be to set and check input and output sample rate and bit depth, for starters. The number of time I started doing a test at 44 kHz and accidentally left the hardware at something else...
In fairness the ASUS Xonar D2 board does a fair bit better at Rightmark, and costs about the same. I'd trust Onkyo before Asus to make a better sounding output stage, though.
Some people were asking, so:
About the sound card I'm using to record and measure the phonoclone.
Onkyo SE-200PCI
http://www.jp.onkyo.com/wavio/se_200pci/index.htm
It's the big sister of the output-only SE-90PCI (which I also have used and found to be very good). The SE-200 adds 24-96 recording capability, digital input, and 7.1 channel output. There is even a "limited edition" version of the SE-200PCI with a copper shield and custom "audio" capacitors.
All cards will play back 24 bit 192 kHz data.
I've checked all the modes with Rightmark Audio Analyzer and apart from a slightly annoying channel imbalance of 0.3 dB it tests very well.
The 2448 test is also provided from Onkyo directly, you can see the results here:
http://av.watch.impress.co.jp/docs/20061211/dal262d.htm
My copy matches that performance, within error. (See attached reports)
It also sounds really good as a digital source, and I'd recommend it to anyone who might be able to get their hands on one (it's Japan only). The only real annoyance is the drivers. It is not nearly as convenient as it should be to set and check input and output sample rate and bit depth, for starters. The number of time I started doing a test at 44 kHz and accidentally left the hardware at something else...
In fairness the ASUS Xonar D2 board does a fair bit better at Rightmark, and costs about the same. I'd trust Onkyo before Asus to make a better sounding output stage, though.
Update on the C3 story.
I had installed those GEC caps mentioned back up a few posts. Very nice, to be honest they are so similar to the blackgates that I can't really distinguish a difference.
So being happy with that, tonight I installed another cap, also from madisound, this time a bennic 1.5uF one: click for bennic at madisound
So, like the last one, these are pretty cheap, 1USD each (although the last one was only 30c each!). So I'll let it run in for a while and pass some comments later in the week on them.
Fran
I had installed those GEC caps mentioned back up a few posts. Very nice, to be honest they are so similar to the blackgates that I can't really distinguish a difference.
So being happy with that, tonight I installed another cap, also from madisound, this time a bennic 1.5uF one: click for bennic at madisound
So, like the last one, these are pretty cheap, 1USD each (although the last one was only 30c each!). So I'll let it run in for a while and pass some comments later in the week on them.
Fran
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.