LDC capsule self-noise?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
ricardo said:
Yes please Abbey.  If you could post your LTspice model too, that would be much appreciated.

BTW, if you are just injecting your signal to the 'Test Point' doesn't give the 'real' EQ.  You need your source floating with the capsule. 
The voltage source is properly connected in my simulation.
 

Attachments

  • U87 sch+plt.jpg
    U87 sch+plt.jpg
    129 KB · Views: 44
And the sim file.
You have to rename it as an .asc file.
Thanks for this Abbey.

-15dB down @ 20kHz is a bit drastic.  :eek:

The other sources I looked at circa 2011 (from now defunct prodigy-pro) suggest
  • Gyraf thinks it's just -3dB(or -4dB) @ 16kHz from Neumann specs.
  • SSLtech actually measures the electronics, which shows that, but as a shallow shelf centred around 3kHz.
  • David Satz suggests comparing U87 response specs with SM69 (which doesn't have the EQ) but that shows -2dB narrowly centred at 10kHz
I'll have to look further into this.
 
ricardo said:
-15dB down @ 20kHz is a bit drastic.  :eek:
Yes, definitely. I re-checked my sim. In fact that was an old file, that is a cross between U87 and U87ai. I made a number of simplifications (particularly the bias circuit) that I thought would not make a significant difference.
Feel free to check between my published schemo and the official ones; I may have simplified too much... One thing that may differ is my use of a 2N3819 instead of the undivulged original item, but there is no way it could make a big difference.
One thing I know I missed is the 47n (33n in the U87ai) cap in series with the 6.8k. That makes a significant difference at LF; none at HF.
 
Don't trust the jfet models in the sims some of the tube models you can find seem to off as well in microphone circuit.
Often they give way off bias adjustment values.

I have been using LT spice for some time with microphone circuits.
 
Gus said:
Don't trust the jfet models in the sims some of the tube models you can find seem to off as well in microphone circuit.
Often they give way off bias adjustment values.

I have been using LT spice for some time with microphone circuits.
FET and tube Spice models are generally either utterly wrong or good enough for gain and impedance evaluation. Where they are not adequate is regarding noise, partly because manufacturers and scientists are still debating over noise. Most of the models are good at HF applications, but not so much at audio frequencies.
The electrical noise spectrum must be convoluted with the auditory response for significant results.

Often they give way off bias adjustment values. 
Spice models cannot be more accurate than the physical part they are representing. Tolerance on FET's pinch voltage is often -50/+100%
 
abbey road d enfer

To add to my post I bought 1000 of one jfet number in one batch. I measured most of the batch most converged around an  IDSS and Vgs. That jfets model in LT Spice does not sim close at all the how the real parts measure in a real circuit. Now it is a sample size of most of the 1000 jfets I bought maybe a 100 left to measure.

I am aware jfets specs can have a range
 
Gus said:
abbey road d enfer

To add to my post I bought 1000 of one jfet number in one batch. I measured most of the batch most converged around an  IDSS and Vgs. That jfets model in LT Spice does not sim close at all the how the real parts measure in a real circuit. Now it is a sample size of most of the 1000 jfets I bought maybe a 100 left to measure.

I am aware jfets specs can have a range
There is a number of unverified Spice models floating around. I've found the 2N3819 model in LTspice to be reasonably close to the real world; I mean the simulation was wrong by only a factor 2. I don't expect more from a simulation.
 
"What I'm talking about is large midrange dip that would compensate for an undamped diaphragm. Typically, that technique of reducing self-noise involves about 20dB midrange cut."
Which commercially produced mic would that be?
 
"I'm not writing about first NT1 at all, which was schoeps circuit adaptation by Jim Williams "
What was the adaptation? It was not a simple copy?
 
bockaudio said:
"What I'm talking about is large midrange dip that would compensate for an undamped diaphragm. Typically, that technique of reducing self-noise involves about 20dB midrange cut."
Which commercially produced mic would that be?
B&K 4179, GRAS 40HF, PCB Piezotronics 378A04, all specialized measurement microphones.
Some have sub-zero self-noise.
I've read only one comment about the 4179 used for recording and the tester said it was inadequate. I guess the transient response might be the cause; two second-order resonators fighting each other predict a troubled transient behavior.
 
abbey road d enfer said:
Another technique has been developed, which consists in applying less damping and compensating with electronic EQ. This allows reducing the intrinsic noise to sub-zero values.

This makes perfect sense. In terms of linearization of frequency response there is no difference between making it acoustically, or electronically. Since the curve is quite predictable, electronic EQ'ing is by far easier, more accurate, and preferable. That is, less diaphragm damping means larger volume of back chamber, which in turn, by itself would have less noise, to start with. Larger chamber also means less ringing on the top and less HF chamber resonances. The entire system will have much better compliance, so the diaphragm can 'breath', reducing effect of compression intrinsic to condenser capsules, as opposed say, to natural sound of ribbon microphones...

Best, M
 
In terms of linearization of frequency response there is no difference between making it acoustically, or electronically

..and then again, this would impose an unwanted emphasis/deemphasis on everything in behavior not linear in nature - like breakup resonances and other nonlinear sources of error. These may be much more predictable if you control it early, won't they..?

Jakob E.
 
gyraf said:
..and then again, this would impose an unwanted emphasis/deemphasis on everything in behavior not linear in nature - like breakup resonances and other nonlinear sources of error. These may be much more predictable if you control it early, won't they..?

Jakob E.
The issue there is that the capsule's resonance and damping are environment-dependant; I would think these measurement microphones have somewhat restricted operating conditions.
Actually, the B&K 4179 datasheet indicates compensation coefficient for pressure and temperature that seem to affect sensitivity only.
And the PCB piezotronics manual adds humidity compensation coefficient.
 
gyraf said:
..and then again, this would impose an unwanted emphasis/deemphasis on everything in behavior not linear in nature - like breakup resonances and other nonlinear sources of error. These may be much more predictable if you control it early, won't they..?

Jakob E.

Jakob,

Unwanted by whom?:))) Depends who you ask--acoustical, or electronics engineer:))) The resistance controlled system is no fun, to start with and in both cases we are dealing with the same LCR. If dealing acoustically, we rely on precision shims, or high tolerance machining, if electronically--just simple trim pots will do it. So in the end it is where you are willing to make compromises. Since the undamped capsule can have up to 60dB peak at the resonant frequency the real issue to watch is overloading...

abbey road d enfer said:
The issue there is that the capsule's resonance and damping are environment-dependant; I would think these measurement microphones have somewhat restricted operating conditions.
Actually, the B&K 4179 datasheet indicates compensation coefficient for pressure and temperature that seem to affect sensitivity only.
And the PCB piezotronics manual adds humidity compensation coefficient.

Abbey,

Especially true for measurement mics, which often should be used in harsh environments. The recording mics most invariably used in much more controlled environment, so much less of the issue here. Pressure?--of course--after all those are pressure mics! Vented mics should not be sensitive to it. I would think the underdamped capsule might be less sensitive to thermal expansion, so might be another benefit here.

Best, M
 
ln76d said:
I'm not writing about first NT1 at all, which was schoeps circuit adaptation by Jim Williams only about NT1a and NT1 (current version, also named NT1, black body finish). For the current revision i doubt there's even unofficial schematic over web, if there's any please share the link.
Electric measurement of NT1a circuit didn't showed me any midrange dip, but clearly showed low end rolloff.
1) What is the difference between the Schoeps circuit and Rode, besides perhaps the fet and compont mfgs?
2) That circuit doesn't have anything resembling ANY mid EQ. 
 
Back
Top