U87 (Revision A) in BM800 microphone donor body

[Potti18]

Finally got mine finished today

You had to do some adjustments on that SC400 body or the PCB fit in perfectly? Starting mine very soon and can get one for like 10€.

 

[mihi_fuchs]

Hi @mihi_fuchs



Yes, I will do.




I have checked the U87 Building Guide and for my understanding, to refine RV10 for the best sweet spot is to generate a tone via J4. How can I do that without special equipment?

If not, which cheapest kind of machine should I be looking for?

Or preferable a portable digital tone generator or portable digital oscilloscope will do the trick for this and bullet proof for future projects?

A low distortion sine wave generator. Maybe you can use your sound card for that too. I never tried. But you can give it a go. Just be careful and start very low in volume.

I used GND on Capsule/body which i took a cable from XLR pin1 and J9, is that correct?

that's perfectly fine. Values look good too.

Good idea, let's see if I can figure it out.

----

| QUESTIONS |

1. I must give a lot of gain to have a reasonable level of sound, is this normal?

Nope, make sure you have the PAD jumper in the right position. Check with guide, there is a description for it.
Additionally make sure you have put the Transformer wires in the right connection holes. If you changed the direction of the transformer that can also result in low signals.

2. When I activate phantom power and start giving gain, I hear some crackling which goes away after a few seconds ON. Is this anything with the transformer? Does it need to break in on order to have normal results?

No, a crackling noise is fine or the moment switching it on. But not for too long. The capsule is fine, is it?
The transformer has little to do with the phantom. It is more the DCDC converter. Voltages around the Transistor on the DCDC board are stable, are they?

---

Always very helpful. Thank you so much @mihi_fuchs

Best Regards,

RB

 

[RBeTT]

Here you can find some good info about soldering
If i were in the US I'd try the Jli capsule you can find it here TSC-2
Never tried out for my build as I'm from Italy and shipping makes it inconvenient, but they're not so expensive so worth a try in my opinion if you are in the US.

About the cleaning process, I didn't wash the pcb as you see in the video you mentioned, I've just used some Isoprophyl alchool on a towel and a toothbrush, or cotton fioc for the HiZ junction, and the mic is incredibly silent.

One thing I receltly implemented are a pair of 22nF caps at the XLR, as I was having some RF noise.. now it works perfectly.

@antocu ,I'm kind of having the same problem with RF noise...which part specifically [on Mouser] did you use as 22nF caps specs for the XLR?

Thanks

 

[antocu]

@antocu ,I'm kind of having the same problem with RF noise...which part specifically [on Mouser] did you use as 22nF caps specs for the XLR?

Thanks

Hi, i think I had these caps K223K15X7RF5UL2

 

[mrgrooves666]

What kind of noise you're getting? I have built three of these, and two have sort of a hiss noise, more of a fizzy kind, not smooth, and kind of over the normal noise floor, but its only present with certain preamp combinations, and only noticeable at very high gain settings, otherwise not noticeable under normal circumstances. I think it might be RF interference too, since it was mentioned somewhere on this thread, mihi omitted some inductors that seem to act as RFI filters too. Boards are clean, and so is hiz section too, and mics sound wonderful. I'm going to give a try with the RFI OPA filter too, thanks for the tip....

 

[MicMaven]

This is a live video recorded with fuchs' u87. . . . This is the 8th single by Korean singer-songwriter Jo Jin-young.

What a treat. Lovely voice, pretty song, and very relaxing and pleasant. I do not notice the microphone, so it must be OK. A moment of respite and escape - thank you. James

 

[RBeTT]

What kind of noise you're getting? I have built three of these, and two have sort of a hiss noise, more of a fizzy kind, not smooth, and kind of over the normal noise floor, but its only present with certain preamp combinations, and only noticeable at very high gain settings, otherwise not noticeable under normal circumstances. I think it might be RF interference too, since it was mentioned somewhere on this thread, mihi omitted some inductors that seem to act as RFI filters too. Boards are clean, and so is hiz section too, and mics sound wonderful. I'm going to give a try with the RFI OPA filter too, thanks for the tip....

I'm still working on them. I need to set the gain very high, and I have a bit too much floor noise, too. I will do what Mihi said and see how things work out. I'll keep you posted.

 

[jp8]

Hi, i think I had these caps K223K15X7RF5UL2

These are X7R caps, generally frowned upon when used in the audio signal path for the distortion they add and their microphonic behavior. Admittedly, I'm not sure whether that will be audible. For RF purposes, they are at least much better than film caps.

In many mic designs, you'll find these caps as RFI filter. The 22nF value originates from an era without cell phones and in many cases prove to be ineffective against cell phone interference, which is the most common source of RFI nowadays. Even when these caps are mounted on the XLR's. I've done RFI tests on various mics with various 22nF locations. It was a hit or miss whether it worked or not. And combined with the capacitive load of long XLR cables, it could also cause early top-end fall-off.

The KM84 has a similar RFI susceptibility. I was able to effectively suppress the cell phone RFI issue, but only by adding a SMT component RFI filter, combined with a XLR insert with short pin 1 to chassis connection (Being short is important here. Literally every mm of wire between RFI filter caps and chassis ground reduces the effectiveness. That's also the reason wired components are less effective than SMT parts.) It's the same XLR insert as used in the Takstar CM-63, which has very effective RFI suppression.

I've copied the RFI filter to a small PCB that should be mounted on the XLR insert. See attached picture. I've tested it in several LDCs and works quite effectively when tested with an RF jammer between 500MHz and 1.2GHz. Will make the board available through PCBway in due time.

See also here: Post in thread 'Is there a way to reduce cell phone interference in DIY KM-84?' https://groupdiy.com/threads/is-the...-interference-in-diy-km-84.86707/post-1160686

Jan

 

[MicMaven]

problem with RF noise...which part specifically ... did you use as 22nF caps specs for the XLR?

Serving Suggestion 28737 -- Do less to protect the microphone - and more to mitigate the source of RFI/EMI.

Rationale: I am a funny sort of guy. Rather than install capacitors on all my microphones, or have a studio/shack festooned with a gazillion clamp-on ferrite beads, etc., I identify, isolate, and mitigate the source of interference, instead.

Many amateur radio operators sustain RFI in the shack, and cover the problem with various tricks, especially clamping ferrite beads on every audio and RF patch cable. I believe these issues are best resolved with judicious routing of cables, cinching up loose connectors, and employing best installation and operating practice. I have a pile of audio gear stacked next to a 1500 watt RF deck with no issues. Moving cell phones away from sensitive equipment is nearly always sufficient mitigation. Except for ordering more beer and pizza, I cannot imagine why anyone would need a cell phone in a recording studio . . .

Just is just MY take ... I am not being critical, argumentative or judgmental - your mileage may vary! / J

 

[joulupukki]

yes they do. but you isolate them via the teflon part. the hole in the pcb fits the teflon ring leading to no contact to the pcb. top and bottom part are conducting, but no contact to the pcb is there as the teflon parts isolates.

Hmm. I’m not seeing a hole large enough on the PCB for the 11218 teflon to fit through. Is it supposed to be right where my pencil is pointing in this picture? Should I drill it out so the teflon will fit in there?

EDIT: I see in the build guide that it was drilled out. I just drilled it out and the turret fits fine now.

 

[mihi_fuchs]

Hmm. I’m not seeing a hole large enough on the PCB for the 11218 teflon to fit through. Is it supposed to be right where my pencil is pointing in this picture? Should I drill it out so the teflon will fit in there?
View attachment 138994
EDIT: I see in the build guide that it was drilled out. I just drilled it out and the turret fits fine now.

Yes. You have to drill it out. My apologies. On initial orders i did it before shipping. However now i do kot do it, so you have the option to go as you prefer.

 

[joulupukki]

Yes. You have to drill it out. My apologies. On initial orders i did it before shipping. However now i do kot do it, so you have the option to go as you prefer.

Makes sense. That’d be good tidbit to add to your build guide, unless you already have and I’ve got an old version of it.

 

[mihi_fuchs]

Makes sense. That’d be good tidbit to add to your build guide, unless you already have and I’ve got an old version of it.

Included in the building guide. Online already. V 1.2. Thanks for the hint, best regards and nice Sunday!

Michael

 

[joulupukki]

I just finished up my build (thanks @mihi_fuchs) and so far the mic is sounding great! Here's a little REW capture (totally non-scientific because I'm just placing it right in front of my monitor speakers and playing the test tone) comparing it with the DIY U87 I made with the Dany Bouchard PCB in an Aurycle A460. The Bouchard vintage U87's output is lower (as expected).

Just now I'm realizing that I don't have a shock mount that works with the HL-95 mic body. Doh!



DIY U87A (Fuchs PCB)
HL-95 Mic Body
Arienne Audio K87 Capsule (v1)
3U Audio GZT-87 Transformer

DIY U87 (Bouchard PCB)
Aurycle A460 Mic Body
JLI TSC-2 Capsule
3U Audio GZT-87 Transformer

AKG P420
Modded with 470pF added in parallel with the existing 220pF for a total of 690pF. I've gigged with this mic as our center mic in an acoustic bluegrass band and it is stellar).

Once I find a good shock mount (open to suggestions), I'm guessing this new DIY U87A (Fuchs) will be my go-to for gigs.

 

[MicMaven]

Here's a little REW capture (totally non-scientific because I'm just placing it right in front of my monitor speakers and playing the test tone)

Novice question No 20986

Acknowledging the less than ideal test protocol, what do these graphs tell us about the microphones, given the apparently consistent test test conditions and similar results? Note - I am not complaining or asking about about the test protocol - I am asking ONLY about what we can take away from the graph.

Considering how closely parallel the three lines are, and match each other, albeit at different sensitivity levels, do they suggest ...?

the capsules produce similar results,
the capsules and boards produce similar results,
the different bodies/grilles have little impact on the results,
the test source is consistent over time,
the capsules are equal, with differing sensitivity,
all or none of the above, and/or
something completely different, if so what?

Again, I am merely curious what the data indicates, not challenging the test procedure. The lines are (I think) remarkably similar across the entire range, so I wonder whether to make of it. It almost suggests the TC-2 and Arienne capsules are close enough to be drop in replacements for each other, and yet we KNOW BETTER. Or do we? Perhaps I an be making too much of the data. Perhaps it has a very limited, yet useful purpose I am missing. I ask because we have seen similar graphs before, and I was not sure what all I can make of them, either.

Thanks. James, (recklessly revealing his ignorance!)

 

[mrgrooves666]

For starters, green is based on poctop's boards, wich don't offer dc to dc converter, hence much less sensitivity due to less available polarizing voltage.
The other two do have that, so sensitivity is higher. Now I believe I've read somewhere in the forum that AKG is using JLI capsules in these mics, or at least they look like them, so that would explain the similarity in the high end with the Poctop build. The smaller difference in sensitivity between the two mics with higher polarization voltage, could be just that, the polarization voltage, or capsule capacitance?

 

[Khron]

The smaller difference in sensitivity between the two mics with higher polarization voltage, could be just that, the polarization voltage, or capsule capacitance?

... Or JFET tolerances, or the bias resistors around them (setting the gain to whatever it ends up as, depending on the JFET; no guarantees it'll be the same even between two mics of the same model).

 

[mihi_fuchs]

... Or JFET tolerances, or the bias resistors around them (setting the gain to whatever it ends up as, depending on the JFET; no guarantees it'll be the same even between two mics of the same model).

… or differences in the transformer windings. Lot of options, but the differences are quite subtle imo Between the two with dcdc conversion.

 

[soliloqueen]

Novice question No 20986

Acknowledging the less than ideal test protocol, what do these graphs tell us about the microphones, given the apparently consistent test test conditions and similar results? Note - I am not complaining or asking about about the test protocol - I am asking ONLY about what we can take away from the graph.

Considering how closely parallel the three lines are, and match each other, albeit at different sensitivity levels, do they suggest ...?

the capsules produce similar results,
the capsules and boards produce similar results,
the different bodies/grilles have little impact on the results,
the test source is consistent over time,
the capsules are equal, with differing sensitivity,
all or none of the above, and/or
something completely different, if so what?

Again, I am merely curious what the data indicates, not challenging the test procedure. The lines are (I think) remarkably similar across the entire range, so I wonder whether to make of it. It almost suggests the TC-2 and Arienne capsules are close enough to be drop in replacements for each other, and yet we KNOW BETTER. Or do we? Perhaps I an be making too much of the data. Perhaps it has a very limited, yet useful purpose I am missing. I ask because we have seen similar graphs before, and I was not sure what all I can make of them, either.

Thanks. James, (recklessly revealing his ignorance!)

the curves are not even remotely similar. the Y axis is just very zoomed out. matched at 1k, there is a nearly 4dB difference between the model with the tsc-2 and the one with mine at 5k. That's like, a 30% difference in volume. This is an absolutely massive difference. That's bigger than the difference between a u87 and a u47. the graphs also very clearly show pretty severe influence from the body type. you have fallen victim to a very common psychovisual phenomena when uncompensated graphs are zoomed out too far in the y axis. at this scale, basically any two graphs will look similar provided they don't have very obvious "lo-fi" nonlinearities

 

[RBeTT]

Hi @mihi_fuchs

Sorry for the late reply, it has been crazy weeks...but I am finally back to answering properly your very kind reply

A low distortion sine wave generator. Maybe you can use your sound card for that too. I never tried. But you can give it a go. Just be careful and start very low in volume.

I'm a bit confused about how to do it. Will I use the studio monitors to produce the sine wave generator? If so, do I have to open the mic every time to adjust the trimmer and do the test again?

Nope, make sure you have the PAD jumper in the right position. Check with guide, there is a description for it.
Additionally make sure you have put the Transformer wires in the right connection holes. If you changed the direction of the transformer that can also result in low signals.

I have confirmed that the PAD and transformer wiring are where they should be. We both used a 3u Audio transformer, too, so I just followed the pictures.

No, a crackling noise is fine or the moment switching it on. But not for too long. The capsule is fine, is it?

The capsule it's new and right off the box. Since it's backplate independent, I have attached 2 black wiring accordingly to the building guide.

Voltages around the Transistor on the DCDC board are stable, are they?

I don't know how to answer you. Can you be specific about where and what should I look for? I'm still a bit fresh on this new world...Sorry!


INFO | QUESTIONS

1. The mic body it's conical and not cylinder, I can feel a bit of no being centered with the bottom screw, can that be a problem? Perhaps if the tall WIMA components touch it? ;

2. Since I didn't know how to calibrate properly, I adjusted to 60v and 10.5v on RV24 and RV10, respectively.

3. Via BOM, it comes 2x as Turrets, which I don't know why. On my first U87 build, I had a turret on J3 [front capsule wiring] like the @Wordsushi had done it, which helped a lot with the hissing noise. Should I keep it or not?

4. I have isolated the rails with electrical tape, and I used m1.6 Screws to attach the PCB's without touching anything.

---

Here are the build pics, Sound spectrum via EQ and the sound recording I made in my last test.

20Nov24 Sound Test | AUDIO SAMPLE [Dropbox]








---

As always I appreciate feedback, time, and your patience.

Best Regards,

RB