Apex 460 Noise

[Bo Deadly]

It could be tricky to try and feed signal in directly because the input is such high impedance. Meaning you cannot actually connect a probe to the grid without affecting the impedance.

Place the mic 1 meter away from a monitor speaker and play a pure tone into it. Try and do it in an isolated / quiet place (make a couch cushion "isolation chamber" maybe). Raise the monitor output until the mic output starts to clip (use lower mic pre settings to make sure it's not the mic pre clipping). Look at the fft / spectrum in your DAW and raise the monitor level until you see the 2nd and 3rd harmonics jump. That's the clipping point. Then increase mic pre gain until you get full-scale digital on the ADC. Then, without touching anything else, turn off the tone and record / listen to the silence critically and compare to decide if you really have a problem.

If you still think you have a problem, you will need to characterize the noise more descriptively (hiss, whistle, hum, etc). Post a screenshot of the spectrum / FFT (use the highest sample rate + averaging if your DAW supports specifying that sort of thing so that the resolution is better).

 

[radardoug]

Its not a problem to feed a signal into the grid, however the generator will have low output impedance. Put a .1 uf cap in series with the hot lead from the generator, and this will keep the d.c. conditions on the grid the same.  Note though that the generator will load this point from a noise perspective, and so noise figures would be misleading with the generator connected but turned down.
If you use a 1M grid resistor, and the unit is quiet, but when you go back to the 200M resistor its noisy, then it could be the 200M resistor, or just the difficult high impedance conditions the tube is working in.

 

[bernatvm]

Thank you all for your suggestions. I don't have any signal generator other than my small audio interface at home, nor I have any monitors. Next week I'll go to my college electronics lab and do all the tests, electronic and acoustic. Will pots all screenshoots here so you can help me.

I think the problem I have right now is solid transmision of low frequency (i have 20db of <20Hz over my noise floor). Only way to prove that is by injecting some tone to the circuit, as the solid transmision should not be there if there's no capsule to pick it up.
I suspect that because when I hold the mic with my hand this 20db become 40 ??? . I also have a mesurement mic, so I will do a bi-channel freq and phase analysis too.

 

[Bo Deadly]

I suspect that because when I hold the mic with my hand this 20db become 40 ???

Sounds like a grounding issue. Check resistance between pin 1 of the far end of the cable and mic body and basket. Be aware that it can be difficult to find a good connection on anodized or smooth plating surfaces so you might need to fidget with the probes a little until you get a properly low reading. The resistance should be 0 of course (or whatever your meter displays when the probe tips are shorted together).

 

[bernatvm]

How are you supposed to deal with the heater ground or "return" and the ground of the b+/signal? Where should they meet. I didn't mod any of this but maybe stock mic was wrong wired in the first place?

 

[Bo Deadly]

The rule is that currents should run anti-parallel so that magnetic fields created by AC tend to cancel. Meaning the heater ground wire should return right next to the supply wire. If the cable actually has separate wires for pins 4 and 7, you could try using them to separate the two grounds. Meaning separate 4 and 7 and connect 4 (and 3?) to pin 9 of the tube. But then in the power supply circuit you will need to add a largish resistor between the two grounds just to hold them at the same DC potential (elevating the heater supply can also significantly reduce noise).

However, I don't think any of that is the problem because a) your heater supply is already pretty seriously filtered DC so it should be quiet and b) the circuit doesn't have enough gain to amplify the heater noise that much (compared to a guitar amp for example). Unless you find a heater wire running close to the grid wire, I think your problem is elsewhere. Note that tube stuff is inherently a little noisier then something like a JFET circuit.

If you're just listening to silence at full volume, you will hear some hum. Make a recording of 10 seconds of vocalizing or guitar or whatever followed by 10 seconds of silence. Put a solid signal into the mic and adjust the mic pre level so that you're recording at +4dBu or whatever you normally do. Then listen to the recording with 10 seconds of sound and then 10 seconds of silence. Can you still hear the hum during the 10 seconds of "silence"?

 

[bernatvm]

I measured the headgrill to the far end on the cable. I get 2.6ohm on pin 4 and 1.7 ohm on pin 7. While checking your explanation i realized the psu schematic on first post is wrong. Here's the actual psu

 

[Bo Deadly]

I measured the headgrill to the far end on the cable. I get 2.6ohm on pin 4 and 1.7 ohm on pin 7.

The fact that you get different readings even through the two pins are connected together at the other end means that your probes are simply not making good contact. Keep trying ...

 

[bernatvm]

I went to the lab earlier today and took at the voltages where the cable join the mic pcb with the oscilloscope. Didn't have time for any of the other tests. Here are some pics of the readings, I think there are no problems.

Heater

B+

I also attach an FFT of the noise. Snapshot is taken in a silence moment on a speech track. The gain is what I would usually use in that situation.

 

[bernatvm]

The fact that you get different readings even through the two pins are connected together at the other end means that your probes are simply not making good contact. Keep trying ...

I keep trying and getting similar results. I'm gonna recheck all the wiring between the 7pin and the pcb in the mic. Also will try with a new cable, stock is bad quality anyway...

 

[Bo Deadly]

I keep trying and getting similar results. I'm gonna recheck all the wiring between the 7pin and the pcb in the mic. Also will try with a new cable, stock is bad quality anyway...

Measure without the cable between pin 1 on the mic and the basket. Measure the cable separately. Divide and conquer. What does your meter read with the probes directly shorted? Mine reads 0.3 ohms which seems to be the limit of my DMM. But I have nice probes. If your probes are corroded or dirty, you might have to clean them to get good readings. If you really are getting 2-3 ohms between the basket and earth ground at the wall, that's not good. That could be a source of noise. Try grounding the basket separately using a jumper to something that you know has a solid ground. Is the noise reduced / eliminated?

 

[bernatvm]

Measure without the cable between pin 1 on the mic and the basket. Measure the cable separately. Divide and conquer. What does your meter read with the probes directly shorted? Mine reads 0.3 ohms which seems to be the limit of my DMM. But I have nice probes. If your probes are corroded or dirty, you might have to clean them to get good readings. If you really are getting 2-3 ohms between the basket and earth ground at the wall, that's not good. That could be a source of noise. Try grounding the basket separately using a jumper to something that you know has a solid ground. Is the noise reduced / eliminated?

My meter is reading 0.5 ohm with probes shorted. Cable mesurements gives me 2.4 - 0.5 = 1.9 ohm on pin 4 and 1.3 - 0.5 = 0.8 ohm on pin 7. From both pin 4 and 7 on the mic conector to the grill I get 0~0.1 ohm unstable reading.

I did a test today and unsoldered the capsule. I still get the same exact noise profile, so grounding should be the issue. It seems that the 7 pin cable is not ok...

 

[Bo Deadly]

Sounds like a wonky cable. Are the cable pin / wire connections soldered? If yes, try re-wetting with a little bit of new solder. If not, and you have the right tool for removing the pins from the housing, you could try removing them and re-crimping the pins.

 

[bernatvm]

Recived the new cable today and still same isues. It looks like the microphone is amplifying the lows and sub a lot more than the rest of the frequencies and the noise gets a lot louder on those...

I'm a bit lost, don't know what more to do.

 

[radardoug]

Where is your studio located? Are you in a high traffic area? That spectrum curve looks typical for a noisy environment.
Thats what the high pass filter is for, to clean off all the low frequency crud you dont need.

 

[bernatvm]

I doubt it's ambient noise what the mic is picking, I'm prety sure it's its self noise

 

[Rocinante]

I worked out of studio that had 'dirty' electricity. There were nice surge protectors to protect and clean up some of the electrity but there were a few sockets that always created noise. Tube mics were effected the Most.  When the mic was plugged into a clean socket it was great but when using one of the 'dirty ' sockets it was a hum machine.

 

[bernatvm]

I've checked the mic in different places (different plug, different rooms and different towns) and I have the same problems everywhere

 

[radardoug]

Does the 0 dB on your spectrum display indicate true maximum signal level? Then the LF noise is 50 dB down. Higher frequencies are 90 dB down, which is probably as good as it will get. Are you sure its not power supply noise or bad shielding?

 

[bernatvm]

I checked psu with oscilloscope and seemed right. About shielding, what could I do to check the shielding?