Schematic Rode NT2-A

[LevinGuitar]

And yet it works.
NT2-a is different from NT1 black.
I put a 0.51nF capacitor (511) from the jFet source (pin C11) to ground.

I just tried 470pf and 4.7nf there, not much difference, just a hair less of top end

 

[micolas]

I just tried 470pf and 4.7nf there, not much difference, just a hair less of top end

Exactly what I said.
That's what I followed.
The 510pF value worked perfectly for me.
Already with 1nF it seemed more closed than I like.

 

[kingkorg]

And yet it works.
NT2-a is different from NT1 black.
I put a 0.51nF capacitor (511) from the jFet source (pin C11) to ground.

Glad if it works for you, source impedance is quite low, i can't see how 0.51nF would do anything, but i could be missing something.

 

[micolas]

Glad if it works for you, source impedance is quite low, i can't see how 0.51nF would do anything, but i could be missing something.

Kingkorg,
I approached the topic in the most pragmatic way. I soldered a 10nF capacitor to ground and went the reverse audio path to the input. I thought it was better to filter at the beginning of the audio chain. It worked after C12, between C11-C12, before C11. After some tweaking I decided to remove the treble that was bothering me, but with the desire to still keep a sound profile close to the U87ai (at least graphically, because I didn't have one at hand). I used Superlux 681 headphones, it's true that they have expanded highs. At the value of 511G (I hope I said 0.51nF correctly), I liked it. Radio-ready sound.
Please, If you have the possibility, in an NT2a of yours, make comparative measurements between stock and my mod. I'm curious if I attenuated about 0.5..0.7 dB after 8..9kHz, that's what I was hoping for.

 

[micolas]

I just tried 470pf and 4.7nf there, not much difference, just a hair less of top end

Did you try my mod simultaneously with your mod, or on the stock circuit?

 

[LevinGuitar]

I tried first yours one only, then both. Anyway none of them makes a real change. A small piece of foam under the capsule in my NT1000 did a bigger change.

Today I was wondering about desoldering C11, pick there the signal there and send it through a cap and a low? ratio transformer into the output and see if it works (to add later a classic deemphasis)

You don't like my ways, but I do learn somehow

 

[Khron]

I don't suppose you ever considered adding a capacitor in parallel with the JFET's drain resistor, have you?

 

[LevinGuitar]

No :/
Tomorrow will try!

 

[micolas]

I tried first yours one only, then both. Anyway none of them makes a real change. A small piece of foam under the capsule in my NT1000 did a bigger change.

Today I was wondering about desoldering C11, pick there the signal there and send it through a cap and a low? ratio transformer into the output and see if it works (to add later a classic deemphasis)

You don't like my ways, but I do learn somehow

I also thought about using the input part of the circuit (which ensures low noise) and a transformer or BJT+transformer (for some magnetic saturation). I'm still studying.
In another context I thought that a jumper over C11 would increase the equivalent capacitance (from about 41.1 nF now, (47nF in series with 330nF) to 330nF and lower the frequency response (but would invert the output phase probably xlr pins 2 and 3 should be reversed and would affect the Hi-Pass.
For treble, as I didn't want a classic attenuation of several dB, like for a very bright k67 capsule, in this case I was satisfied, the stock microphone is very close as FR to the U87ai, only a small peak above about 9khz more , it bothered me a bit.
I thought that changes towards the output would introduce greater distortions, phase changes (although passive multi-way filters in acoustic enclosures exist and work).
And one more thing, if the impedance at the input of C11 was too low, then for a good response to the bass, the values of C11 and C12 had to be several microFarads, and the resistance from them, component of the filter, put to ground would be should have been a few kohms (not 940kohms for 'flat').
The next mod will be a U87 headbasket for the NT2a (the one received on the HL-77).

 

[micolas]

I don't suppose you ever considered adding a capacitor in parallel with the JFET's drain resistor, have you?

Khron, yes.
I use it in the classic Shoeps circuit. Sometimes in combination with the mod used by MicParts.

 

[micolas]

Hey guys,
does anyone know if there is any similarity between the circuit diagram of NT2a and NT2000?

 

[LevinGuitar]

I don't suppose you ever considered adding a capacitor in parallel with the JFET's drain resistor, have you?

Tried today, it works a bit too, but cannot find a nice combination between all/some of them

 

[greg_is_not_a_bot]

In case anyone is still watching this discussion or comes across it, I have an issue I'd like to run by the group.

I had some issues with random popping and noise in my NT2a, replaced the capsule (was going to anyway) with an RK-47 (no change in the noise) and then proceeded to replace what I figured were all the caps in the high-impedance section before the JFET. (See photos) I used Wima FKPs where I could, but the two 0.1μF are MKPs due to space limitations. The popping is gone (nice!) but I have a weird problem. In cardioid the mic sounds fine, but in figure-8 there's a big difference in the sound from the front and rear of the mic. The front side sounds scooped, while the back side sounds mid-hyped. I haven't done any measurements, but it seems like the same frequencies attenuated on one side are enhanced on the other. In omni, the mic is quite dark. All 3 patterns work as expected from a directional standpoint.

Any clues as to where I should be looking? I have zero experience working on LDCs, so I'm a bit out of my depth here. I know the logical thing would be to try a different capsule, switch capsule leads, etc, but with the change of capacitors it's a tightly packed circuit board now and I'd prefer not to do too much desoldering / resoldering on those little smd pads just to get access.

Oh, and I used what I believe are the correct capacitor values, based on the schematic that @Drosselmeier supplied earlier in this thread. I wasn't sure about one of them that appears to be listed as 4700pF, since the numbers are a little fuzzy. Could that be my issue? Is that the wrong value for the cap?

Thanks in advance for any help, in case anyone actually ends up finding this post after so long.

 

[Khron]

The popping is gone (nice!) but I have a weird problem. In cardioid the mic sounds fine, but in figure-8 there's a big difference in the sound from the front and rear of the mic. The front side sounds scooped, while the back side sounds mid-hyped. I haven't done any measurements, but it seems like the same frequencies attenuated on one side are enhanced on the other. In omni, the mic is quite dark. All 3 patterns work as expected from a directional standpoint.

Was it really worth blindly "shotgunning" all those capacitors, without (seemingly) measuring anything? Here's to hoping it won't be used in vibratory environments though - that's quite some cantilevered mass, relying only on the adhesion between the copper pads and the fiberglass-epoxy substrate...

Onto more practical issues though, what bias voltages are you getting, between ground and each of the two square pads near the output of the voltage multiplier?

I see you've... resoldered / replaced most of the components? Some seem to be barely hanging on...

 

[greg_is_not_a_bot]

For those unfamiliar with the practice, I can certainly understand the alarm at seeing large(ish) through-hole components on a surface mount board. I assure you, though, it works quite well if it's done with a bit of care and the board wasn't poorly made to begin with. I've put several hundred of these little Wimas as well as small electrolytics in place of SMD components with not a single failure to date. I've been doing it for years, and I've come to trust the track record. As far as a high-vibratory environment goes, I treat my LDCs almost like I treat my ribbon mics and honestly any hit that'll dislodge a pad from the PCB will have probably pulled awfully hard on the center electrode of the 3 micron diaphragm. I'm very new to LDCs, but not to practical electronics work.

Shotgunning is not really what's going on here, to be fair. I replaced a couple of caps that I suspected of causing my noise issues and that problem is indeed solved. (Measuring those faulty caps didn't strike me as particularly useful.) I replaced the other 3 audio path caps preventatively, since I was already in there. I generally don't like testing cap values in-circuit, and I didn't completely trust the SMD cap values after they'd been heated for removal. (At the time, I was still working with a conventional solder station - I have a dedicated tweezer-style SMD tool now, and even though I generally dislike doing SMD work it seems unavoidable on anything made in the last 20-ish years.)

As far as the rest of the components, I suppose you'll have to take that matter up with the Australians. I've owned this mic since new, and nothing else on the board has been touched. If you're seeing something I've missed, I would very much like to know so I can rework anything the factory got wrong. I know some people cite re-flowing certain components as advisable but I don't know which ones might be the usual suspects on an NT2a.

On the more pertinent subject of bias voltage, I opted to check it at the terminals of the pattern switch since that looks like the last stop before the resistor network. I get 77.7V on one and 77.5V on the other, which seems well matched to me but maybe these capsules are pickier than I realize. I think measuring elsewhere would likely exceed the abilities of any of my meters due to their 10M impedance. Is that a false assumption? Any insight would be much appreciated, especially regarding the cap values that seem to be missing or seemingly contradictory in the schematics I have. I may have just made mistakes reading them, and I certainly appreciate the efforts of the people who provided some readable layouts here at least. It sounds like you understand the inner workings of LDCs quite well, which is territory I'm just making an entry into.

 

[Khron]

For those unfamiliar with the practice, I can certainly understand the alarm at seeing large(ish) through-hole components on a surface mount board.

It wasn't the fact that through-hole components were employed in a surface-mount fashion, but the HOW alarmed me. As in, free-standing, as opposed to at least laid flat on the PCB (ideally with some double-sided tape or something). On a side-note, that first line "feels" a bit passive-aggressive or patronizing or something...

I think measuring elsewhere would likely exceed the abilities of any of my meters due to their 10M impedance. Is that a false assumption?

That is indeed true.

Any insight would be much appreciated, especially regarding the cap values that seem to be missing or seemingly contradictory in the schematics I have.

Most (remaining) values are not really critical, especially in the power filtering sections.

As far as the rest of the components, I suppose you'll have to take that matter up with the Australians. I've owned this mic since new, and nothing else on the board has been touched.

If i look at, and zoom into, the last photo you posted (the overall view), I'm not sure i can see any resistors that have factory-soldering, and a couple of the ones between the JFET / low-cut capacitors and the side of the PCB are nowhere near the surface of the board.

All the voltage multiplier capacitors have added solder (also betrayed by the flux leftovers around them), and a couple of the double diodes are far from straight. If that's really how it came from Rode, it must be some late-friday-afternoon hack-job...

But glossing over all of that, are you still experiencing the very different sound of the new capsule's two sides?

 

[greg_is_not_a_bot]

Khron - no aggression intended. I never know what people have done / seen before, and I do agree that sometimes stabilization is a good idea. In this case, I just had reasons not to and I've never found it to be problematic in gentle-use scenarios. An amp that's going on the road? Yes, that's going to get a different approach.

Looking more closely at the stock component work, I do see what you mean about a less than meticulous job of component placement at the factory. Maybe there was some Fosters involved that day. I really hadn't paid attention to that area, since the mic had been working fine for nearly 20 years and I was focused on the circuitry ahead of the JFET. I don't see the need to correct any of the factory work since it functions well, but I won't say it was done with a lot of pride in workmanship.

I haven't resolved anything with the sound yet, no. I'm pretty certain it's a matter of capacitor values in the rear diaphragm circuit. (It sounds nice in cardioid, just what I'd expect from the RK47 capsule, so front diaphragm related values are evidently correct.) My cap values were taken from one of the diagrams here, but they could be incorrect. It's unlikely that the resistor values are out of spec, since the mic sounded fine before the capsule and capacitor replacement but I suppose stranger things have happened. So I'm probably going to have to simply experiment with different cap values until things sound right, but I figured if someone knew the circuit well enough they might already have the information. To be honest, I'm not clear on why the front and rear capsules appear to have different R/C networks in the first place since they're meant to sound identical. As I said, I'm new to LDCs.

 

[micolas]

@greg_is_not_a_bot

How does the RK-47 capsule sound in the NT2A headbasket (with shape/dimensions closer to the 87 style than the 49) which favors mids, mids-highs? Doesn't that sound edgy,pokey?

 

[Khron]

I haven't resolved anything with the sound yet, no.

Are you 10000000000% certain the three capsule wires of the new one, ended up in the corresponding correct spots? Say, if the backplate wire got swapped with the rear diaphragm one? I'll double-check my two stock ones tomorrow...

I'm pretty certain it's a matter of capacitor values in the rear diaphragm circuit.

That assumption is based on.......? Or rather, why capacitor values?