|
I took some time to try to reverse engineer the tBone SC140 PCB. The board itself has 797 Audio printed on it, so there's no doubt as to who makes the mic. I had searched for the original schematic and couldn't find the one by Khron, but I found a post that listed the Cap values. They seemed to match so I filled them in, however I haven't verified them yet. I have double checked my work the best that I could, though it's very possible that mistakes exist. I took measurements of the PCB and the housing, and got the following dimensions:
PCB is 16.4 by 107mm and 1mm thick
Bottom of PCB is 6.5mm from the OD of tube, or 1.25mm offset from the center.
Body Tube ID is 17.5mm
The body capsule threads are 19mm diameter, 2mm long and 0.5mm pitch
The body is 20mm diameter and 132mm long (without the capsule installed) 146mm with capsule.
Capsule pogo pin is 2.5mm diameter, 17mm long, with 3mm travel minimum (I'm not having any luck finding suitable replacement pogo pins)
The capsule pin is inset 3mm.
The switches are 61mm and 81mm centered from the XLR connector side of the PCB
The vent plate on the back of the capsule is the common 8 hole type, looks cast but is really high quality with straight holes and very flat.
Q1, Q3, and D4 appear to be made by Galaxy Microelectronics, and the JFET Q2 by Fairchild.
D3 and D1 have 0F marking on tip, with a sideways J.
It is assumed that the capsule is biased with a negative voltage.
It's an interesting design and I'm trying to understand how it works.
I don't have a test setup to check the voltages yet. I'm looking for a high impedance multimeter or circuit that I can use.
-Stephanie
|
Thanks for putting all the work into reverse engineering the SC140 and for sharing, Stephanie! I haven't checked all the details (yet), but here are some remarks and additions from my side:
- In the schematic, you forgot to draw R1 (15k) between XLR pin 2 and the R2/C4/D1/Q1 node.
- D2 is indeed a 24V Zener diode. The actual voltage measured in my SC140 equals 10.5V. I guess they used 24V just to maximize the drain voltage of Q2 and protect it from electrical overstress. Just nit-picking: the D2 symbol was drawn as a regular diode, but it is actually a Zener diode. Slightly different symbol.
- D1 and D3 are 6.8V Zener diodes. Something like FDZ6.8 from First Silicon maybe. And same as D2, the symbol should actually be that of a Zener.
I double-checked the PCB dimensions. The actual thickness is indeed 1.0mm, not 1.2mm as I said earlier. I took the thickness from my SC140 PCB design, not from the original PCB. With 1.2mm thickness, the gap between PCB surface and the XLR becomes smaller, which makes soldering a bit easier, but the PCB may not fit into the slot of the nose cone. I also double-checked the width. Mine measures 16.2mm. I'm afraid 16.4 may not fit. Can you check yours again? I can also confirm the center locations of the switches you provided are right. I'm afraid I made a measuring error. My bad. One more thing I'd like to share, which you may already know: when you want to design your own PCB, put ceramic SMT parts (capacitors and resistors) perpendicular to the length direction of the PCB to avoid cracking. Especially the bigger caps (>0402) might crack when inserting the PCBA into the tube or when the microphone drops or vibrates. 797 Audio did a good job in that respect. Wherever possible, I did that too in my design as pictured below. Btw, I changed insights and I never ordered and built this PCBA. I will redesign it later. And to those hawk-eyed people who think I used 1MOhm resistors as JFET and capsule biasing resistors: the resistor bodies are depicted are just a generic 3D model of an axial leaded resistor. The actual resistor value will be either 1G or 2G. Jan 
|
Another thing that struck me: C1 and C2 do not have the same value. In an impedance-balanced circuit, I would have expected them to be of equal value. I am not going to remove the SMT caps from the board of my SC140 to measure the value as I don't want to accidentally lose or damage them. I have only one SC140 and I want to keep the original PCBA as a reference when comparing it to my own circuit.
|
Thanks for the verification and tips. I attached the updated schematic. I don't have a tool to verify zener diodes yet, so I only saw the forward voltage drop of 0.7v. Looking at getting the cheap GM328a, it seems to be popular over at EEVBLOG.
I pulled the SMD cap values from an old forum post so I haven't checked any yet. I can't find my fancy LCR meter else I'd remove the capacitors and verify. I bought 3 SC140 mics, 1 to take apart and play with so I'll have a working set at all times - so let me know if you need anything else measured.
I plan on 3D printing a replacement nose cone so I don't think PCB thickness will be an issue. Standard PCB thickness is 1.2 to 1.6mm - 1.6 being a bit nicer for structural strength.
I double checked the PCB width in multiple locations and still measured 16.4mm, however that includes the leftover material from the V cut, if I carefully measure the width around it, I get 16.1mm minimum. So taking the difference we get 16.2 to 16.3mm. You can get away with 16.4mm but you might need to file the edges a little due to the kerf.
I also double checked the capsule thread with my thread gauge and can confirm it's 0.5mm pitch. For comparison of 19mm thread capsules, I found a thread that states the Rode NT5 has 0.4mm pitch and the Behringer B5 is 0.45mm pitch - bummer. The ID of the capsule retaining ring thread is 18mm so it's probably 18.5mm*0.4 to 0.5mm - maybe I'll have some luck finding a capsule housing this can go into instead. (I really should learn how to use a lathe and just make this stuff myself).
Speaking of PCB's I have a question for the group. I was thinking about how the body ground is only connected to 1 point - the set screw on the XLR connecter. This means that the return signal path for the capsule must go all the way around the body, through that screw, and up the length of the PCB back to the JFET bias circuitry.
I was wondering if it would be any better to add edge spring contacts to the edge of the PCB to contact the body. This would remove the 200mm+ loop that the ground would have to travel from the capsule. Something like this??edge mounted sideways.
-Stephanie
|
Has anyone measured an SC140 to either confirm or deny the 9 dB self noise?
On Sun, Jun 25, 2023 at 12:57?PM Stephanie < Qwertys@...> wrote: Thanks for the verification and tips. I attached the updated schematic. I don't have a tool to verify zener diodes yet, so I only saw the forward voltage drop of 0.7v. Looking at getting the cheap GM328a, it seems to be popular over at EEVBLOG.
I pulled the SMD cap values from an old forum post so I haven't checked any yet. I can't find my fancy LCR meter else I'd remove the capacitors and verify. I bought 3 SC140 mics, 1 to take apart and play with so I'll have a working set at all times - so let me know if you need anything else measured.
I plan on 3D printing a replacement nose cone so I don't think PCB thickness will be an issue. Standard PCB thickness is 1.2 to 1.6mm - 1.6 being a bit nicer for structural strength.
I double checked the PCB width in multiple locations and still measured 16.4mm, however that includes the leftover material from the V cut, if I carefully measure the width around it, I get 16.1mm minimum. So taking the difference we get 16.2 to 16.3mm. You can get away with 16.4mm but you might need to file the edges a little due to the kerf.
I also double checked the capsule thread with my thread gauge and can confirm it's 0.5mm pitch. For comparison of 19mm thread capsules, I found a thread that states the Rode NT5 has 0.4mm pitch and the Behringer B5 is 0.45mm pitch - bummer. The ID of the capsule retaining ring thread is 18mm so it's probably 18.5mm*0.4 to 0.5mm - maybe I'll have some luck finding a capsule housing this can go into instead. (I really should learn how to use a lathe and just make this stuff myself).
Speaking of PCB's I have a question for the group. I was thinking about how the body ground is only connected to 1 point - the set screw on the XLR connecter. This means that the return signal path for the capsule must go all the way around the body, through that screw, and up the length of the PCB back to the JFET bias circuitry.
I was wondering if it would be any better to add edge spring contacts to the edge of the PCB to contact the body. This would remove the 200mm+ loop that the ground would have to travel from the capsule. Something like this??edge mounted sideways.
-Stephanie
|
Looking at the schematic, components used and the capsule size. I can’t see how published spec of self noise is achieved. ? Best Regards,
Jules Ryckebusch?
toggle quoted message
Show quoted text
On Jun 25, 2023, at 15:45, eric benjamin <ericmbenj@...> wrote:
? Has anyone measured an SC140 to either confirm or deny the 9 dB self noise? On Sun, Jun 25, 2023 at 12:57?PM Stephanie < Qwertys@...> wrote: Thanks for the verification and tips. I attached the updated schematic. I don't have a tool to verify zener diodes yet, so I only saw the forward voltage drop of 0.7v. Looking at getting the cheap GM328a, it seems to be popular over at EEVBLOG.
I pulled the SMD cap values from an old forum post so I haven't checked any yet. I can't find my fancy LCR meter else I'd remove the capacitors and verify. I bought 3 SC140 mics, 1 to take apart and play with so I'll have a working set at all times - so let me know if you need anything else measured.
I plan on 3D printing a replacement nose cone so I don't think PCB thickness will be an issue. Standard PCB thickness is 1.2 to 1.6mm - 1.6 being a bit nicer for structural strength.
I double checked the PCB width in multiple locations and still measured 16.4mm, however that includes the leftover material from the V cut, if I carefully measure the width around it, I get 16.1mm minimum. So taking the difference we get 16.2 to 16.3mm. You can get away with 16.4mm but you might need to file the edges a little due to the kerf.
I also double checked the capsule thread with my thread gauge and can confirm it's 0.5mm pitch. For comparison of 19mm thread capsules, I found a thread that states the Rode NT5 has 0.4mm pitch and the Behringer B5 is 0.45mm pitch - bummer. The ID of the capsule retaining ring thread is 18mm so it's probably 18.5mm*0.4 to 0.5mm - maybe I'll have some luck finding a capsule housing this can go into instead. (I really should learn how to use a lathe and just make this stuff myself).
Speaking of PCB's I have a question for the group. I was thinking about how the body ground is only connected to 1 point - the set screw on the XLR connecter. This means that the return signal path for the capsule must go all the way around the body, through that screw, and up the length of the PCB back to the JFET bias circuitry.
I was wondering if it would be any better to add edge spring contacts to the edge of the PCB to contact the body. This would remove the 200mm+ loop that the ground would have to travel from the capsule. Something like this??edge mounted sideways.
-Stephanie
|
I have no personal interest in knowing the actual values of the capacitors as I'm not going to modify the original circuit anyway. But for those who want to do component swaps, it would help to know the original values and also dielectric material. My personal interest consists only of knowing the circuit on a conceptual level.
I recognize the grounding issue you mentioned and I too am looking for commercially available spring contacts which can be used. The concept you describe is used in Neumann SDC, e.g. the KM184. You can see the internals here:??. I was thinking of using nickel-plated steel tags used for batteries. But the ones I could find are actually pure nickel and too soft. I know Nickel plated stainless steel tags do exist (they are used in my company's rechargeable products), but I haven't been able to get hold of them in the right dimensions on e.g. Aliexpress. So for the time being, my best option is this:??. You may want to clip the + terminal. The two "ears" of the CR2032 battery retainer push against the inside of the mic tube. Haven't tried it yet. Is anyone having a better idea?
Alternatively, at the XLR side, the pin 1 connection to the tube can als be established by using an XLR connector with an integrated grounding tab. I use these:??. They slip nicely into the tube and instead of a single, questionable ground connection to the tube, you'll get a double connection. Btw, due to the thicknes of the solder cups of this connector, only 1mm or 1.2mm PCBs will fit. You cannot push a 1.6mm PCB into the space between the pins. Other XLR connectors might accept 1.6mm PCBs, but not this one. For 1.6mm PCBs, consider these:??. But these are without grounding tabs.
Another thing to think about when using 1.6mm PCBs is the following: When you insert/pull the PCBA into/from the tube, then in my experience you'll have to bend the PCB somewhat in the switches area by pushing on the switches. This might prove to be a little (too) difficult with 1.6mm PCBs. I don't know, just guessing...
I once tried to screw Rode NT5 capsules on my SC140. The thread is indeed slightly different and you'll not be able to screw it all the way down to the end stop. It would have been nice if one could have other capsules than the Cardioid capsule on the SC140, which would make it a universal, cheap workhorse SDC. So I mailed 797 Audio with a request for Omni and HAC capsules which would fit on the SC140, but never received an answer. I even mailed them for a 2nd time but to no avail. While others seem to be successful in contacting 797 and even sourcing capsules from them directly, 797 Audio didn't care to reply :-(. Maybe someone else would like to give it a try and maybe we can do a group buy? Seems they take customer orders for at least 100 pcs. Will be difficult to reach such high numbers in a group buy, but who knows...
PS, I thought I introduced myself back in 2021 when I signed up for Groups.io. However, I didn't but will do so later.
Jan
|
Forgot to mention the pogo pin. The closest match I was able to find is this one:?https://www.aliexpress.com/item/4001294383416.html?spm=a2g0o.cart.0.0.fd0238danIFNlV&mp=1 . (M1648 version, 3.0x18.5 5Amps) But there zillions of pogo pins on Aliexpress and from distributors, so there might be better options available.
I consider to 3D print nose cones as well, but haven't decided on the material yet. Which material would have the best properties in terms of surface and bulk resistance and moist absorption, if those matter at all for the given dimensions? Anyone having any experience? And what should be the maximum gap between tube and cone? I guess a wide gap will cause FR anominalies.
Jan
|
I found my LCR meter and confirmed that C1 is 15nF but also found C2 is 15nF (I accidentally broke C1 in the process so it's a good thing I got a measurement first). I think I'll need to get tweezer probes before I attempt to measure any other caps.
Those XLR connectors with the tabs look pretty nice and a worthwhile upgrade.
Regarding materials to 3D print the nose cone with, I found this (PDF Link)??which is a good start. PTFE would be ideal, but it is expensive and you'll need to outsource it. I found a PC/PTFE blend filament which might be promising. Polycarbonate looks like it would work well enough, but you'll need to test with an insulation meter. You can raise resistance a little by proper design. You'd want it SLS or resin printed for highest accuracy, FDM would need post processing on the cone otherwise you might get weird effects.
I'm disappointed enough with the PCB of the t.Bone SC140 that I'm unsure if I want to proceed with a rebuild. It's quite obvious that they're lying about the noise floor - the dirty PCB, transistor, and ceramic capsule capacitor are enough evidence for me. That and being unable to find replacement bodies or capsules make it less desirable to mod even if the stock capsule is nice quality. I couldn't find a noise figure on the Fairchild BF245A but if it's anything like the NXP, the 1.5 DB isn't bad at all.
I think I'll move onto the Takstar CM60 and CM63 which seem to have a nicer quality body - and replacement capsules will be easier to find.
|
even if the stock capsule is nice quality I have not tested this specific brand/model of SDC capsule, but I’ve tested 50+ common Chinese SDC capsules — MXL, Apex, Nady, etc — and found all of them to have a very colored response. See attached PDF. It shows 5 MXL capsules selected randomly from a giant bag of leftover parts from a decade’s worth of mic upgrades. There was a 4dB spread in sensitivity across these 5 capsules, which I’ve normalized here to make the frequency response comparison easier.? All 5 have thin bass. These sweeps were done at 10 inches, so you’d expect to see a lift in the bass due to proximity effect. All five exhibit a response dip above 5kHz, some extreme (-3dB within 1kHz) followed by a fairly sharp rise (+6dB over about 3kHz). The best mic circuit in the world won’t make these capsules sound good. In fact, in this test these capsules were connected to a high-grade implementation of the Schoeps CMC5 design, which is linear (within 0.5dB) from below 20 Hz to 20kHz.? The point of this is that while it’s always rewarding to dissect, reverse-engineer, and upgrade mic circuits, the real-world value of that is sometimes limited to the educational benefit. If your goal is to make a great-sounding microphone, there’s no point at all in touching the circuit unless and until you have a nice capsule… because the capsule largely determines the sound of the microphone. The circuit can do some limited EQ; in practice that’s usually limited to rolling off the top or bottom. There is no circuit EQ that will fix the monstrous 6kHz notch exhibited by capsules #3, #4, and #5 in the attached sweep. Also shown on the PDF is a frequency sweep of one of my cardioid SDC capsules, just to illustrate that there are smooth-response SDC capsules available for DIY upgrades. Omitting proximity effect, this capsule is within 1dB of flat to ~11kHz. These won’t fit every possible cheap donor mic; we designed them to fit the common 22mm MXL/CAD/Apex/Nady SDCs, as those are ubiquitous and usually inexpensive.? — matt.
|
Thanks for the PDF. That is useful to see. The SC140 is very popular at Thomann but there's very little measurements to verify their spec sheet. They state a 9dBa noise floor and a pretty flat response which would be incredibly impressive if it were true (but very likely is not). Michael Joly once touted the impressiveness of these mics after his mods, but since he closed up shop all of that information is lost to the abyss. So there's a possibility that they are actually as impressive as they state but there are some major factors in the way: no alternative bodies or capsules (that we know of) are available for this mic.
The ability to swap 22mm capsules is partially the reason why I'm looking at the Takstar SDC mics. I bought a MXL 991 thinking I could use the body, but it's quite poor quality and is pretty much useless in all respects. The Takstar CM60/63 are noticeably higher quality, and I just completed reverse engineering the circuit and it's much better than the T.Bone SC140 with the classic Schoeps CMC5 circuit with a few modifications (I'll be posting it shortly).
I'm curious with your capsule comparisons, I was looking at the 3u audio capsules which are pricey for singles but someone found they'll get quite reasonable at quantity of 10-30.
Over on the groupdiy forums there is a user Chrisfromthepast that has an excellent thread with a lot of useful comparisons of 22mm capsules. It looks like the Takstar capsules have a pretty flat response, but I don't think they sell their mics in matched sets. I don't have the equipment to do any frequency response tests so I'm in a bit of a bind there.
You're right it's rewarding to dissect these circuits and so far I'm learning a lot by doing it, plus I get to share what I learn because I'm sure I'm not the only person who wants an SDC < $200 that's decent quality. And I love to make it open source because not everyone has the money or tools required to do this stuff.
-Steph
|
On 7/1/23 04:34, Stephanie wrote:
[snip]
... because I'm sure I'm not the only person who wants an SDC <
$200 that's decent quality. And I love to make it open source
because not everyone has the money or tools required to do this
stuff.
-Steph
_._,_._,_
Which leads us to the Alice and and all her variants here - I
believe it's still rather expensive to get the capsules in Europe,
but it's a really great SDC for well under $200 and not a lot of
fancy tools either.
-Scott
--
---- Scott Helmke ---- scott@... ---- (734) 604-9340 ----
"I have ceased distinguishing between the religious and the secular,
for everything is holy" - Joe Henry
|
And for a slightly larger capsule the TSB2555B is seriously good. ?Scott refers to the TSB160 as a SDC. ?It is really good as well. ? Best Regards,
Jules Ryckebusch?
toggle quoted message
Show quoted text
On Jul 1, 2023, at 09:27, Scott Helmke <scott@...> wrote:
?
On 7/1/23 04:34, Stephanie wrote:
[snip]
... because I'm sure I'm not the only person who wants an SDC <
$200 that's decent quality. And I love to make it open source
because not everyone has the money or tools required to do this
stuff.
-Steph
|
TSB-165A is the classic Alice capsule.? 16.5mm or "MDC" if there is
such a term.
-Scott
On 7/1/23 09:36, Jules Ryckebusch
wrote:
And for a slightly larger capsule the TSB2555B is seriously good.
?Scott refers to the TSB160 as a SDC. ?It is really good as well.
?
Best Regards,
Jules Ryckebusch?
On Jul 1, 2023, at 09:27, Scott Helmke
<scott@...> wrote:
?
On 7/1/23 04:34, Stephanie wrote:
[snip]
... because I'm sure I'm not the only person who wants an SDC
< $200 that's decent quality. And I love to make it open
source because not everyone has the money or tools required to
do this stuff.
-Steph
--
---- Scott Helmke ---- scott@... ---- (734) 604-9340 ----
"I have ceased distinguishing between the religious and the secular,
for everything is holy" - Joe Henry
|
Yep! 165 mis quoted my self lol Best Regards,
Jules Ryckebusch?
toggle quoted message
Show quoted text
On Jul 1, 2023, at 09:43, Scott Helmke <scott@...> wrote:
?
TSB-165A is the classic Alice capsule.? 16.5mm or "MDC" if there is
such a term.
-Scott
On 7/1/23 09:36, Jules Ryckebusch
wrote:
And for a slightly larger capsule the TSB2555B is seriously good.
?Scott refers to the TSB160 as a SDC. ?It is really good as well.
?
Best Regards,
Jules Ryckebusch?
On Jul 1, 2023, at 09:27, Scott Helmke
<scott@...> wrote:
?
On 7/1/23 04:34, Stephanie wrote:
[snip]
... because I'm sure I'm not the only person who wants an SDC
< $200 that's decent quality. And I love to make it open
source because not everyone has the money or tools required to
do this stuff.
-Steph
--
---- Scott Helmke ---- scott@... ---- (734) 604-9340 ----
"I have ceased distinguishing between the religious and the secular,
for everything is holy" - Joe Henry
|
Good news, I'm in the US and can buy as many as I want.
Though I want to keep the end-addressed pencil form factor. These are going to be in a backpack most of the time, so keeping the whole kit small is one of my goals. I am hoping to keep it within the 22mm body style so I don't need extra clips or anything.
I could do that very pretty copper tube style you did, but I don't know if my hands are the right kind of steady for that.
Part of why i'm getting so detailed on the dimensions and threads of these mics is so I can see what will fit in them. It'd be nice to know what donor bodies can be used.
-Steph
|
That PC/PTFE filament seems promising, indeed. Didn't know it existed. Thanks.
|
Regarding nose cone material, I run the center conductor / spring through a piece of Teflon insulating tubing. The material of the rest of the cone is relatively unimportant.
|
Regarding low cost SDC microphones . . .
____________________________________________
Golly, Stephanie, considering all the cited specifications, I have lost track of the main objective. What are you trying to accomplish? Do you want a low cost SDC microphone? OR, do you want to modify a low cost SDC microphone? If the latter, why insist on modification? I recently poled a number of professional recording engineers, and several recommended the CM-60 as having an excellent value to performance ratio. In what aspect is the original deficient, and what advantage(s) will your realize by modification?
NOTE - I am NOT being critical, negative or snarky. I have simply been distracted by long list of specifications, and wonder why they are important - for example, why should I be interested in circuit board thickness, or the number of ferrite chokes, etc? How is all that material to your purchase decision? THANK YOU for your courtesy and consideration herein. James
|
It's partially curiosity, and partially the desire to modify.
While you might not have a need for all of this information, someone else might. I gathered all of these specifications for my own use, and to share.
I wanted a microphone in the 22mm pencil form factor as replacement capsules are abundant - which makes it a good platform to modify. Knowing the PCB dimensions and other information about the body and capsule allows me to design my own PCB if I wish to do so. Knowing the original circuit design allows me to modify it if I wish to do so.
It also allows me to see what design decisions the original manufacturer had, and evaluate it based on that. Reverse engineering is a learning tool. Seeing the differences in the circuits, like the ferrite and common mode choke in the CM63 makes you think about the design decision there. This is a group called MicBuilders after all, so I think the observation and discussion of microphone circuit design would always be of interest.
As to the end goal, purchasing a low cost microphone and modifying one are towards the same thing - to have a quality microphone at a low cost. How would I know if a microphone would benefit from modifications unless I examine it to make that determination.
In the case of the Takstar CM63 there's very little that can be improved. But it's just as interesting to notice the design decision differences on the Takstar CM60 vs the CM63. Likewise why is the t.Bone sc140 circuit the way it is and how can they claim 9dBa noise which is unheard of in it's class.
Also comparing the Takstar CM63 to the AKG Perception 150 and seeing that the circuit is nearly identical. It's essentially the same mic but the AKG retailed for over 5 times as much (with it's latest revision the P170 at about 2x the cost).
So far I've learned a lot and I hope that this knowledge will be useful to others. I've learned that the Takstar mics are incredible quality for the cost. I've learned that MXL mics aren't worth the raw materials they're made of. I've learned that the tBone mic might have very exaggerated specs. This is all without much expensive equipment (my most expensive tools being a $300 LCR meter and $500 microscope). I could go further if I had an oscilloscope and signal generator.
I come from a maker community where people share openly so others can learn and duplicate or improve on designs. All the information I have shared here so far is in that goal. If someone wants to use this information to make their own PCB, or to find out what capsules will fit, or modify it, I've done all the work for them. The hope is that they might pay the generosity forward. But even if they don't, or they decide to use my work for profit, I'm okay with that too - that just means that more people can have access to whatever product is made from it without the need to understand the designs or make it themselves.
Sorry for the long explanation but to sum it up: I'm selfish and greedy and want the best at the lowest cost. I'm also foolish and will pay whatever it costs to? to get it. And I'm selfless and want to give everything I've learned away.
The makers motto: We do these things not because they're easy or cheap, but because we thought it would be.
-Steph
|
