its not about what your dac is at, but what is 'there' at the very input of the pga pins. by THAT point, are you sure its low? across cables and stuff?
some things just are not great unless they get a 'local push' (lol). chip-amps are like that, too; they work best when they have a local driver pushing them.
I think the MAXIM chip (forget the part #, its 64db and 1db steps controllable by up/down buttons) also wants to have a pre-stage for it. its common, I think.
So the PGA wants a little current on its input eh ? Thanks for posting that. Hmmm...not such a convenient chip after all....and it makes the board all the more inadequate for the job.
The best output on my DAC is its class A head amp and the cables are 30cm long Neotech UPOCC. That shouldn't be a problem.
Connecting it up to the Onkyo soundcard might be more interesting because this will be doing the rear surround and that uses 5532 op amps and 1 to 2 metre cables.
I could either modify these output channels, except I don't particularly enjoy working on soundcards, or I could change the input resistors on the PGA board and add a unity gain op amp/buffer into the circuit.
For simplicity, I need an opamp/buffer designed for +/-5V rails, reasonable current output, and happy with low z load. Does anyone know a low voltage version of LME49720HA or just a buffer chip like a low voltage version of LME49600 ? Something with better rail-to-rail performance....a higher voltage AD8656 maybe ?
Unfortunately, the a20 on AMB's site aren't going to fit in my case and they want a higher voltage supply too.
Thanks for the help guys !
The best output on my DAC is its class A head amp and the cables are 30cm long Neotech UPOCC. That shouldn't be a problem.
Connecting it up to the Onkyo soundcard might be more interesting because this will be doing the rear surround and that uses 5532 op amps and 1 to 2 metre cables.
I could either modify these output channels, except I don't particularly enjoy working on soundcards, or I could change the input resistors on the PGA board and add a unity gain op amp/buffer into the circuit.
For simplicity, I need an opamp/buffer designed for +/-5V rails, reasonable current output, and happy with low z load. Does anyone know a low voltage version of LME49720HA or just a buffer chip like a low voltage version of LME49600 ? Something with better rail-to-rail performance....a higher voltage AD8656 maybe ?
Unfortunately, the a20 on AMB's site aren't going to fit in my case and they want a higher voltage supply too.
Thanks for the help guys !
Last edited:
It wouldn't in this case, but most people want multiple inputs, you can't guarantee that some piece of equipment you buy in the future is 600R or less.
w
OPA2134 is unity-gain stable.
Last edited:
Thanks, nice op amp, but that is limited to 5v same as AD8656.
Actually, the 5532 aren't so bad and perhaps I should just measure the output impedance before I start making things harder for myself. The soundcard has 300ohm resistors iirc and the 5532 has an output impedance of less than 1 ohm so I probably don't need a buffer anyway.
However, it is worth considering what would be best - and I'm thinking something like the BUF602 ? Anyone have any experience with this ? Is there a lower distortion version ?
Tom
Actually, the 5532 aren't so bad and perhaps I should just measure the output impedance before I start making things harder for myself. The soundcard has 300ohm resistors iirc and the 5532 has an output impedance of less than 1 ohm so I probably don't need a buffer anyway.
However, it is worth considering what would be best - and I'm thinking something like the BUF602 ? Anyone have any experience with this ? Is there a lower distortion version ?
Tom
Last edited:
Not in my experience, and my latest project will use up to a dozen of them.
As voltage out DACs typically have an opamp internally to do the I/V, and it must at least be a reasonably decent design not to limit the overall DAC performance, having an opamp, to buffer an opamp to power the PGA seems more than a little dubious to me without some empirical and by that I mean measured, evidence.
do they ?
I have only seen the recomendation of source Zout < 600ohms.
A pre-buffer seems to me like a solution to make sure this is the case whatever source is used.
If you KNOW (by design or other mean) that the source output impedance is lower, what would be the reason to add an addtional chip in the signal path ?
Hardly convincing evidence to me I'm afraid.
No it doesn't.
Edit: I was referring to the 2310 datasheet, so i downloaded the 2311 and there is a slightly different wording in the analogue I/O section of the datasheet. Refer to the last graph on p5 of the 2311 datasheet for THD differences with different Zsource. It does vary, but not enough to worry about or be audible, especially in the DAC instance mentioned earlier in the thread. Drive it from a tube source for example with a (typically) much higher Zout, and it might make a difference to the 2310/2311 THD performance, but it would still be lower than the tubes anyway.
Last edited:
do they ?
I have only seen the recomendation of source Zout < 600ohms.
A pre-buffer seems to me like a solution to make sure this is the case whatever source is used.
If you KNOW (by design or other mean) that the source output impedance is lower, what would be the reason to add an addtional chip in the signal path ?
if you KNOW, then you don't need the extra stage.
it seems like good insurance to me, though. I'm not at all worried if I use a good buffer; and for the times that I might switch this to some other input, I like knowing that the pga is being driven 'no matter what'.
now, if I was building a dac in the same box and there was no external input, you really may not need a pre-driver. but as a general purpose preamp where you may not know what's coming ahead of you, the buffer seems wise and there is really little down side to it.
slightly different wording?
just simply look at the general descr, page7. says right there:
"for optimal performance, its best to drive the pga2311 with a low source impedance. a source impedance of 600ohms or less is recommended."
if you already meet that, fine. but they do say it and its pretty clear they expect a low z-ish source driving this chip.
ymmv but I take this bit of advice seriously.
Hi guys,
I had a quick look at the RMAA software in a bit of downtime at work. Unfortunately it seems I'll need the pro version (for kernel streaming support) or otherwise the good ole Windows kernel mixer will mask any chance of meaningful tests. Did I get this right ? It was just a quick look....
This is the result using the pro software on a standard card I believe :
http://av.watch.impress.co.jp/docs/20061211/dal262a.htm
Frequency response (from 40 Hz to 15 kHz), dB: +0.04, -0.30 Very good
Noise level, dB (A): -99.5 Excellent
Dynamic range, dB (A): 99.2 Excellent
THD, %: 0.0017 Excellent
IMD + Noise, %: 0.0039 Excellent
Stereo crosstalk, dB: -99.6 Excellent
IMD at 10 kHz, %: 0.0044 Excellent
So it seems the card, even in standard trim, is suitable as a test mule.
The BG N caps have recently been replaced with NX HiQ and bypassed with Vishay MKP.
Tom
I had a quick look at the RMAA software in a bit of downtime at work. Unfortunately it seems I'll need the pro version (for kernel streaming support) or otherwise the good ole Windows kernel mixer will mask any chance of meaningful tests. Did I get this right ? It was just a quick look....
This is the result using the pro software on a standard card I believe :
http://av.watch.impress.co.jp/docs/20061211/dal262a.htm
Frequency response (from 40 Hz to 15 kHz), dB: +0.04, -0.30 Very good
Noise level, dB (A): -99.5 Excellent
Dynamic range, dB (A): 99.2 Excellent
THD, %: 0.0017 Excellent
IMD + Noise, %: 0.0039 Excellent
Stereo crosstalk, dB: -99.6 Excellent
IMD at 10 kHz, %: 0.0044 Excellent
So it seems the card, even in standard trim, is suitable as a test mule.
An externally hosted image should be here but it was not working when we last tested it.
The BG N caps have recently been replaced with NX HiQ and bypassed with Vishay MKP.
Tom
Last edited:
I looked at both datasheets before I posted this morning as my previous post indicates. The manufacturer's own measurements show that 2k source, the highest in the graph shown still does not significantly degrade performance. So with a (voltage O/P) DAC O/P of ~200R, there is zero need for any further buffering.
Very few modern SS sources will have a Zout above 600R.
This card is indeed excellent
I find it interesting to test my own device, it's so easy to do. The windows audio management will not create THD or other noise. The soundcard hardware is the weak point to test. At least, you will have the above test to compare with The main advantage of RMAA Pro is to enable use of the ASIO drivers which are usually provided with the pro cards. These drivers give access to device features that standard RMAA will not see. A typical example is digital in/out at higher freq.
Hi,
I did a quick test. When I get the PGA up and running I'll post the results with it in the loop at 0dB gain.
I don't know if there's anyway to improve the test or if I've done something wrong, but anyway, here's the results of my card :
RightMark Audio Analyzer test report
Testing device Envy24 Family Audio (WDM)
Sampling mode 24-bit, 48 kHz
Interface DirectSound
Testing chain External loopback (line-out - line-in)
RMAA Version 6.2.3
20 Hz - 20 kHz filter ON
Normalize amplitude ON
Level change 0.2 dB / 0.2 dB
Mono mode OFF
Calibration singal, Hz 1000
Polarity correct/correct
Summary of my card :
Frequency response (from 40 Hz to 15 kHz), dB +0.04, -0.30 Very good
Noise level, dB (A) -97.8 Excellent
Dynamic range, dB (A) 98.2 Excellent
THD, % 0.0020 Excellent
THD + Noise, dB (A) -88.8 Good
IMD + Noise, % 0.0038 Excellent
Stereo crosstalk, dB -97.0 Excellent
IMD at 10 kHz, % 0.0040 Excellent
General performance Excellent
Summary from the web test :
Frequency response (from 40 Hz to 15 kHz), dB: +0.04, -0.30 Very good
Noise level, dB (A): -99.5 Excellent
Dynamic range, dB (A): 99.2 Excellent
THD, %: 0.0017 Excellent
IMD + Noise, %: 0.0039 Excellent
Stereo crosstalk, dB: -99.6 Excellent
IMD at 10 kHz, %: 0.0044 Excellent
My card details :
Frequency response
From 20 Hz to 20 kHz, dB -0.52, +0.04
From 40 Hz to 15 kHz, dB -0.30, +0.04
Noise level
Left Right
RMS power, dB -97.3 -97.5
RMS power (A-weighted), dB -97.7 -98.0
Peak level, dB FS -82.6 -82.0
DC offset, % -0.0 -0.0
Dynamic range
Left Right
Dynamic range, dB +97.5 +97.7
Dynamic range (A-weighted), dB +98.0 +98.4
DC offset, % +0.00 +0.00
THD + Noise (at -3 dB FS)
Left Right
THD, % +0.0020 +0.0019
THD + Noise, % +0.0034 +0.0034
THD + Noise (A-weighted), % +0.0037 +0.0036
Intermodulation distortion
Left Right
IMD + Noise, % +0.0038 +0.0038
IMD + Noise (A-weighted), % +0.0034 +0.0034
Stereo crosstalk
Left Right
Crosstalk at 100 Hz, dB -96 -93
Crosstalk at 1000 Hz, dB -96 -96
Crosstalk at 10000 Hz, dB -92 -92
IMD (swept tones)
Left Right
IMD + Noise at 5000 Hz, 0.0042 0.0041
IMD + Noise at 10000 Hz, 0.0037 0.0037
IMD + Noise at 15000 Hz, 0.0043 0.0041
I did a quick test. When I get the PGA up and running I'll post the results with it in the loop at 0dB gain.
I don't know if there's anyway to improve the test or if I've done something wrong, but anyway, here's the results of my card :
RightMark Audio Analyzer test report
Testing device Envy24 Family Audio (WDM)
Sampling mode 24-bit, 48 kHz
Interface DirectSound
Testing chain External loopback (line-out - line-in)
RMAA Version 6.2.3
20 Hz - 20 kHz filter ON
Normalize amplitude ON
Level change 0.2 dB / 0.2 dB
Mono mode OFF
Calibration singal, Hz 1000
Polarity correct/correct
Summary of my card :
Frequency response (from 40 Hz to 15 kHz), dB +0.04, -0.30 Very good
Noise level, dB (A) -97.8 Excellent
Dynamic range, dB (A) 98.2 Excellent
THD, % 0.0020 Excellent
THD + Noise, dB (A) -88.8 Good
IMD + Noise, % 0.0038 Excellent
Stereo crosstalk, dB -97.0 Excellent
IMD at 10 kHz, % 0.0040 Excellent
General performance Excellent
Summary from the web test :
Frequency response (from 40 Hz to 15 kHz), dB: +0.04, -0.30 Very good
Noise level, dB (A): -99.5 Excellent
Dynamic range, dB (A): 99.2 Excellent
THD, %: 0.0017 Excellent
IMD + Noise, %: 0.0039 Excellent
Stereo crosstalk, dB: -99.6 Excellent
IMD at 10 kHz, %: 0.0044 Excellent
My card details :
Frequency response
From 20 Hz to 20 kHz, dB -0.52, +0.04
From 40 Hz to 15 kHz, dB -0.30, +0.04
Noise level
Left Right
RMS power, dB -97.3 -97.5
RMS power (A-weighted), dB -97.7 -98.0
Peak level, dB FS -82.6 -82.0
DC offset, % -0.0 -0.0
Dynamic range
Left Right
Dynamic range, dB +97.5 +97.7
Dynamic range (A-weighted), dB +98.0 +98.4
DC offset, % +0.00 +0.00
THD + Noise (at -3 dB FS)
Left Right
THD, % +0.0020 +0.0019
THD + Noise, % +0.0034 +0.0034
THD + Noise (A-weighted), % +0.0037 +0.0036
Intermodulation distortion
Left Right
IMD + Noise, % +0.0038 +0.0038
IMD + Noise (A-weighted), % +0.0034 +0.0034
Stereo crosstalk
Left Right
Crosstalk at 100 Hz, dB -96 -93
Crosstalk at 1000 Hz, dB -96 -96
Crosstalk at 10000 Hz, dB -92 -92
IMD (swept tones)
Left Right
IMD + Noise at 5000 Hz, 0.0042 0.0041
IMD + Noise at 10000 Hz, 0.0037 0.0037
IMD + Noise at 15000 Hz, 0.0043 0.0041
Last edited:
Here are the test results with the PGA kit in the loop set at 0dB (actually the kit lost 0.6dB so I boosted the recording voluume 0.5dB to get the test levels the same as before). As you can see, it seems the card is the limiting factor here and the volume control is having no obvious effect on the test results. Great eh ?
But frankly, I'd take this with a small salt mine. I've heard a major improvement by swapping out the op amps on the card and upgrading the output caps, but the RMAA tests for my card and the standard one tested on the web are almost the same.
I also have another soundcard - Asus Essence ST. It seems this tests better than the Onkyo if you do some searches for results, but the Onkyo sounds better imo. I modded the ST heavily and it improved enormously (especially by changing the caps in the buffer feedback loop and upgrading the xo), but not enough to beat the Onkyo to my ears. Maybe my ears need upgrading.
Anyway, it seems the kit from Jims-Audio is a good one despite the pcb being less-than-top-notch.
RightMark Audio Analyzer test report
Testing device Envy24 Family Audio (WDM)
Sampling mode 24-bit, 48 kHz
Interface DirectSound
Testing chain External loopback (line-out - line-in)
RMAA Version 6.2.3
20 Hz - 20 kHz filter ON
Normalize amplitude ON
Level change 0.3 dB / 0.2 dB
Mono mode OFF
Calibration singal, Hz 1000
Polarity correct/correct
Summary
Frequency response (from 40 Hz to 15 kHz), dB +0.04, -0.30 Very good
Noise level, dB (A) -97.5 Excellent
Dynamic range, dB (A) 97.7 Excellent
THD, % 0.0021 Excellent
THD + Noise, dB (A) -88.2 Good
IMD + Noise, % 0.0039 Excellent
Stereo crosstalk, dB -96.3 Excellent
IMD at 10 kHz, % 0.0040 Excellent
General performance Excellent
Summary of my card as above :
Frequency response (from 40 Hz to 15 kHz), dB +0.04, -0.30 Very good
Noise level, dB (A) -97.8 Excellent
Dynamic range, dB (A) 98.2 Excellent
THD, % 0.0020 Excellent
THD + Noise, dB (A) -88.8 Good
IMD + Noise, % 0.0038 Excellent
Stereo crosstalk, dB -97.0 Excellent
IMD at 10 kHz, % 0.0040 Excellent
General performance Excellent
Frequency response
From 20 Hz to 20 kHz, dB -0.53, +0.04
From 40 Hz to 15 kHz, dB -0.30, +0.04
Noise level
Left Right
RMS power, dB -97.2 -97.3
RMS power (A-weighted), dB -97.4 -97.6
Peak level, dB FS -81.6 -81.0
DC offset, % -0.0 -0.0
Dynamic range
Left Right
Dynamic range, dB +97.3 +97.4
Dynamic range (A-weighted), dB +97.6 +97.8
DC offset, % +0.00 +0.00
THD + Noise (at -3 dB FS)
Left Right
THD, % +0.0020 +0.0022
THD + Noise, % +0.0035 +0.0036
THD + Noise (A-weighted), % +0.0038 +0.0039
Intermodulation distortion
Left Right
IMD + Noise, % +0.0039 +0.0039
IMD + Noise (A-weighted), % +0.0036 +0.0036
Stereo crosstalk
Left Right
Crosstalk at 100 Hz, dB -96 -98
Crosstalk at 1000 Hz, dB -95 -95
Crosstalk at 10000 Hz, dB -89 -87
IMD (swept tones)
Left Right
IMD + Noise at 5000 Hz, 0.0042 0.0042
IMD + Noise at 10000 Hz, 0.0038 0.0038
IMD + Noise at 15000 Hz, 0.0041 0.0041
But frankly, I'd take this with a small salt mine. I've heard a major improvement by swapping out the op amps on the card and upgrading the output caps, but the RMAA tests for my card and the standard one tested on the web are almost the same.
I also have another soundcard - Asus Essence ST. It seems this tests better than the Onkyo if you do some searches for results, but the Onkyo sounds better imo. I modded the ST heavily and it improved enormously (especially by changing the caps in the buffer feedback loop and upgrading the xo), but not enough to beat the Onkyo to my ears. Maybe my ears need upgrading.
Anyway, it seems the kit from Jims-Audio is a good one despite the pcb being less-than-top-notch.
RightMark Audio Analyzer test report
Testing device Envy24 Family Audio (WDM)
Sampling mode 24-bit, 48 kHz
Interface DirectSound
Testing chain External loopback (line-out - line-in)
RMAA Version 6.2.3
20 Hz - 20 kHz filter ON
Normalize amplitude ON
Level change 0.3 dB / 0.2 dB
Mono mode OFF
Calibration singal, Hz 1000
Polarity correct/correct
Summary
Frequency response (from 40 Hz to 15 kHz), dB +0.04, -0.30 Very good
Noise level, dB (A) -97.5 Excellent
Dynamic range, dB (A) 97.7 Excellent
THD, % 0.0021 Excellent
THD + Noise, dB (A) -88.2 Good
IMD + Noise, % 0.0039 Excellent
Stereo crosstalk, dB -96.3 Excellent
IMD at 10 kHz, % 0.0040 Excellent
General performance Excellent
Summary of my card as above :
Frequency response (from 40 Hz to 15 kHz), dB +0.04, -0.30 Very good
Noise level, dB (A) -97.8 Excellent
Dynamic range, dB (A) 98.2 Excellent
THD, % 0.0020 Excellent
THD + Noise, dB (A) -88.8 Good
IMD + Noise, % 0.0038 Excellent
Stereo crosstalk, dB -97.0 Excellent
IMD at 10 kHz, % 0.0040 Excellent
General performance Excellent
Frequency response
From 20 Hz to 20 kHz, dB -0.53, +0.04
From 40 Hz to 15 kHz, dB -0.30, +0.04
Noise level
Left Right
RMS power, dB -97.2 -97.3
RMS power (A-weighted), dB -97.4 -97.6
Peak level, dB FS -81.6 -81.0
DC offset, % -0.0 -0.0
Dynamic range
Left Right
Dynamic range, dB +97.3 +97.4
Dynamic range (A-weighted), dB +97.6 +97.8
DC offset, % +0.00 +0.00
THD + Noise (at -3 dB FS)
Left Right
THD, % +0.0020 +0.0022
THD + Noise, % +0.0035 +0.0036
THD + Noise (A-weighted), % +0.0038 +0.0039
Intermodulation distortion
Left Right
IMD + Noise, % +0.0039 +0.0039
IMD + Noise (A-weighted), % +0.0036 +0.0036
Stereo crosstalk
Left Right
Crosstalk at 100 Hz, dB -96 -98
Crosstalk at 1000 Hz, dB -95 -95
Crosstalk at 10000 Hz, dB -89 -87
IMD (swept tones)
Left Right
IMD + Noise at 5000 Hz, 0.0042 0.0042
IMD + Noise at 10000 Hz, 0.0038 0.0038
IMD + Noise at 15000 Hz, 0.0041 0.0041
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.