Universal Audio SC-1 Hemisphere Mic Modeling Is a Behringer

[sumitagarwal]

Here you are! Should work in Space Designer. Let me know how you get on.

Phenomenal! I'm getting a B-1 as soon as I'm back from Thanksgiving travel. For now to test out the workflow and proof-of-concept I'm using a Rode NT1-A which I don't remember buying...

Over on the UAD forums I seem to have gotten confirmation that the Hemisphere software comes with the models for all of their source mics, not just the source mic that was registered. This seems to mean that a $100 SD-3 would unlock all of that modeling (including B-1 modeling) plus you get a free dynamic mic with it

Makes me wonder what other UA mics may be simple rebrands of existing ones. From what I'm seeing UA doesn't list specs like diaphragm size, which makes narrowing things down trickier.

Will keep you posted on how the IR works, thanks again!

 

[SubSpec]

Makes me wonder what other UA mics may be simple rebrands of existing ones.

https://groupdiy.com/threads/300-universal-audio-sd-1-is-the-60-t-bone-bc500.79887/
https://groupdiy.com/threads/disassembling-a-ua-sp-1.87004/

 

[sumitagarwal]

https://groupdiy.com/threads/300-universal-audio-sd-1-is-the-60-t-bone-bc500.79887/
https://groupdiy.com/threads/disassembling-a-ua-sp-1.87004/

Hah, I'm a dummy. How did I not find this before?

On the plus-side: now we know all it takes is a cheap used SD-3 to make any of those other cheap mics perform a bit better.

 

[sumitagarwal]

Following on that last thought, it seems like for folks who want to use UA/Townsend's modeling tech to get in the ballpark of a lot of classic mic sounds there are a few different options:

1. Sphere LX - $800: dual diaphragm, most advanced option. UA claims you need the DLX to get stereo processing but that doesn't make sense to me. Couldn't you just run it into two separate tracks, run two instances of the Sphere plug-in, and configure one plug-in for the front diaphragm and the other for the rear?

2. SD-3 plus a 797 Audio mics - $450: The SD-3 is the cheapest Hemisphere mic and unlocks access to the other sources as well. So far it's believed that three other mics are suitable replacements for UA mics: $180 MXL-603S pair ~= SP-1, $80 BC 500 ~= SD-1, $80 B-1 ~= SC-1. Netting out 5 total modeling mics of various specialization averaging $90/mic

3. DIY CY002 capsule build + Townsend 1.5.2 plug-in: ~$100 + ~8 hours of effort? Should be roughly comparable to Sphere LX results, vastly cheaper. And, as a bonus, geekier too!

Are there considerations I'm missing, or options I've overlooked?

Obviously, when factoring in used pricing all of the above looks significantly different.

 

[wanderingengineer]

This is my version of your option 3. I am deeply indebted to this forum’s grandees for the necessary pointers that helped me put it together and it’s not really original because of the work they’d already done. With @kingkorg and @Wordsushi deserving special mentions, as well as Jules Rykebusch for the circuits.

So it’s dual OPA Alice, CMOS inverter at 80v on a 797 CY002 in an “Ali ‘47” housing. I have also used kingkorg’s / Wordsushi’s foam mod at the base of the capsule, around the pedestal too because there is perhaps a touch of resonance in this headbasket.

I own the OG Townsend L22 since a few years but I now no longer have to bring it between my studio and my home because I’ve got another option! It sounds almost identical.

Costs: Capsule at £40, body about £60. Circuits built for about £40 (can be done for less I expect). Misc connectors, wiring and adaptation for fitment brings it to about £150 plus my own labour.

View attachment IMG_4942.jpeg

This is closest to the L22 / DLX in my view. From what has been discovered by kingkorg, the LX no longer has a hardware de-emphasis built in because the UAD version of the latest plugins does that work in software. Therefore, you need to run the EQ curve before the old plugin to get the best emulation accuracy, I think. If you already have access to the UAD plugins, you can just select the LX as the input mic and you're off to the races. The view seems to be that a slightly smaller "LX" sized body would make a better match for this, though.

Inspired by those with incredible knowledge on this forum, I am spurred on to try more ambitious projects. A K-47 based mic next I think. Probably all old hat to the trailblazers but exciting territory for me and hopefully a distinct flavour of microphone compared to what I have already.

 

[kenwallace]

Wow, that would be amazing! Much appreciated, and I think likely a very useful resource for anyone looking to use non-Townsend mics with the free Townsend plug-in, regardless of DAW.

Hello i have Townsend mic but i don't have the plugin .. of you have can you please send it to me.. thank you KW

 

[wanderingengineer]

Hello i have Townsend mic but i don't have the plugin .. of you have can you please send it to me.. thank you KW

I believe it’s referenced on the thread if you look carefully. It was a freely available plugin before the UA acquisition but I would still be careful about publicly sharing it directly in case of falling foul of licensing. However if you look at the UAD forums link, it’s been posted by someone else.

The great thing is it’s the latest 1.5.2 version which was Apple Silicon compatible. Unfortunately no AAX for our Pro Tools friends due to Apple Silicon issues at that time.

 

[SubSpec]

Hello i have Townsend mic but i don't have the plugin .. of you have can you please send it to me.. thank you KW

You can register the Townsend mic directly on UAs website and get your license for the latest versions straight from the source.

 

[sumitagarwal]

This is my version of your option 3. I am deeply indebted to this forum’s grandees for the necessary pointers that helped me put it together and it’s not really original because of the work they’d already done. With @kingkorg and @Wordsushi deserving special mentions, as well as Jules Rykebusch for the circuits.

So it’s dual OPA Alice, CMOS inverter at 80v on a 797 CY002 in an “Ali ‘47” housing. I have also used kingkorg’s / Wordsushi’s foam mod at the base of the capsule, around the pedestal too because there is perhaps a touch of resonance in this headbasket.

I own the OG Townsend L22 since a few years but I now no longer have to bring it between my studio and my home because I’ve got another option! It sounds almost identical.

Costs: Capsule at £40, body about £60. Circuits built for about £40 (can be done for less I expect). Misc connectors, wiring and adaptation for fitment brings it to about £150 plus my own labour.

View attachment 140326

This is closest to the L22 / DLX in my view. From what has been discovered by kingkorg, the LX no longer has a hardware de-emphasis built in because the UAD version of the latest plugins does that work in software. Therefore, you need to run the EQ curve before the old plugin to get the best emulation accuracy, I think. If you already have access to the UAD plugins, you can just select the LX as the input mic and you're off to the races. The view seems to be that a slightly smaller "LX" sized body would make a better match for this, though.

Inspired by those with incredible knowledge on this forum, I am spurred on to try more ambitious projects. A K-47 based mic next I think. Probably all old hat to the trailblazers but exciting territory for me and hopefully a distinct flavour of microphone compared to what I have already.

That's a beauty. With the legwork already having been done, how much actual labor would it take for someone to follow this lead?

 

[wanderingengineer]

That's a beauty. With the legwork already having been done, how much actual labor would it take for someone to follow this lead?

If I can do it, it can’t be too hard.

You can buy the circuits from JLI Electronics, fully built SMD. Once you have the other parts, it’s basic soldering, a few AWG 24/26 wires and care!

I made my first one in a couple of hours or so.

 

[sumitagarwal]

A lot of interesting options out there. Interestingly, on the SD-3 + MXL + Behringer + T.bone setup there are no overlapping models between them as Hemisphere sources, so in theory it nets out at 23 models (plus the raw outputs of the four mics). That makes for more models than the Sphere LX, but fewer than the DLX. Additionally, all the models for a given source would all have the same response patterns as it they can't model nuanced dofferences in response patterns like the Sphere can. But with their inherent patterns based on their designs (LDC, SDC, dynamic) they're probably decent at modeling mics in their category

Here's what you get, theoretically:
SC-1 = Behringer B-1:
LD-12
LD-47K
LD-67 NOS
LD-87 Vintage
LD-103
LD-251
LD-414 US
LD-800

SP-1 = MXL 603 or Takstar CM60 (more similar housing and no switch on MXL):
SD-54
SD-56
SD-84
SD-86
SD-451


SD-1 = T.Bone BC500:
DN-7
DN-20
RB-77DX Satin
DN-88
DN-441

SD-3:
DN-4
DN-57
DN-409U
DN-545
DN-604

I know this forum is DIY audio and this isn't really DIY...

 

[Khron]

You can buy the circuits from JLI Electronics, fully built SMD.

Just too bad the layouts are... sub-optimal, let's just say. And even some of the circuit ideas

 

[wanderingengineer]

Just too bad the layouts are... sub-optimal, let's just say. And even some of the circuit ideas

I have read on here that some take that view. Whilst I personally have yet to experience a practical issue, I am in no position to doubt these observations. My own boards are based on the through-the-hole original designs with the Schmitt IC and the OPA on the rear of the boards.

@Khron is hopefully considering making his boards available soon which I imagine will not be open to layout or conceptual issues! Excited to get some! Khron, kingkorg and Rog seem to be total pros on this stuff, which is why mere mortals like me can achieve good results!

 

[Khron]

@Khron is hopefully considering making his boards available soon which I imagine will not be open to layout or conceptual issues!

Well, just last week i (also) got a 5pc batch of my version of the dual OpAlice, but i've barely been able to solder up one of the cardioid-only boards, in between the day-job, the house we moved into, some renovations there, and now the recently-arrived winter with the subsequent snow-clearing activities. But worst-worst case, i should have a couple weeks off around Christmas & New-Year's, and i hope to manage to get to assemble & test out some more boards at least then

 

[Naquat]

Nice informative thread. Thank you.

I have 3 older MXL SP-1 and 2 older Cream colored Rode NT-1 mics.
The Rode mics have the Wima cap mods and I did the Jim Williams mods to the SP-1 mics.
But they're all a bit bright.

All of the mics except for one SP-1 have the original capsules.
One SP-1 has a K47????? I purchased via Reverb but I wouldn't say it sounds any better than the originals, and perhaps a bit brighter.

Anyone use the 797 CY002 or Takstar CTS-2 mentioned in this thread with the above mics I listed?

Thank you!

 

[wanderingengineer]

Nice informative thread. Thank you.

I have 3 older MXL SP-1 and 2 older Cream colored Rode NT-1 mics.
The Rode mics have the Wima cap mods and I did the Jim Williams mods to the SP-1 mics.
But they're all a bit bright.

All of the mics except for one SP-1 have the original capsules.
One SP-1 has a K47????? I purchased via Reverb but I wouldn't say it sounds any better than the originals, and perhaps a bit brighter.

Anyone use the 797 CY002 or Takstar CTS-2 mentioned in this thread with the above mics I listed?

Thank you!

From what I can tell through research on this forum and in other places, most of the Chinese capsules whether manufactured by 797 or Takstar or the “lesser” manufacturers - are “bright” or “harsh” in the high frequencies.

I think the reason for this stems from historic development of the Chinese capsule market and the types of experience, skills etc that came from Europe. As a result, the capsules that the Chinese know how to manufacture well are generally those which have a bright character.

That said, there is nothing wrong with many of the capsules per se and some of them are extremely high-quality and consistent.

I think that the problem comes from the mating of these capsules (often the K 67 type) with flat circuits. I think ideally, flat circuits lend themselves better to capsules with a darker or more traditional frequency response such as the K 47 or possibly the C12 type (which I think itself is “brighter” than the k 47)

Consequently, you will find a lot of people who know what they are talking about recommending that you would do better with changing a lot of the capsules in cheap Chinese microphones to a K 47 style capsules instead. This is not necessarily to suggest that the capsules themselves that populate most of these Chinese microphones are poor (although confusingly many of them are!), they are simply not the best match for the circuitry that has been used.

On this forum, everybody raves about the capsules produced by @soliloqueen and I have no reason to doubt that they are anything other than brilliant but they are also fairly expensive (in relative terms), sold out of America which is not ideal for shipping times /taxes etc. for those of us in Europe and perhaps most importantly in such huge demand that there are often lengthy waiting lists.

I would be very interested to learn whether there are any other capsules being manufactured that hold a candle to the Arienne Audio capsules.

As an aside, once you look at changing capsules in otherwise well produced microphones, you also get into difficult territory regarding the correct polarisation voltage to be applied to the capsule itself. For me this is a confusing area. For example, I would like to swap out the manufacturers capsule in a 20-year-old Rode NT1 but it is quite difficult to understand which K 47 capsules might tolerate the bias output from that circuit, although some threads that I have read suggest that the NT1 only outputs about 40 V.

The gold standard is obviously manufacturing your own circuits using some of the excellent schematics and plans advocated on this forum but I appreciate that this can be time consuming and outside of the skills possessed by some aspiring DIYers.

 

[Khron]

For example, I would like to swap out the manufacturers capsule in a 20-year-old Rode NT1 but it is quite difficult to understand which K 47 capsules might tolerate the bias output from that circuit, although some threads that I have read suggest that the NT1 only outputs about 40 V.

If i remember in a couple weeks, when i hope to dig up my mic suitcases from the garage, i could verify that - i have three of the original "hospital blue" NT1's.

 

[kingkorg]

I did an exercise today, and measured a stock Behringer B-1 straight into Hemisphere plugin with all the emulations. Within reason, depending on which specific mics they used for the modelling all the curves look EXACTLY! how the originals should look like. I am stunned by the accuracy! Disregard the weird drop at 50hz and 20K, REW does this sometimes.

C12



C414




C800


Elam251




TLM103




U47



U67nos



U87nos

 

[sumitagarwal]

I did an exercise today, and measured a stock Behringer B-1 straight into Hemisphere plugin with all the emulations. Within reason, depending on which specific mics they used for the modelling all the curves look EXACTLY! how the originals should look like. I am stunned by the accuracy! Disregard the weird drop at 50hz and 20K, REW does this sometimes.

C12

View attachment 140431

C414

View attachment 140432


C800
View attachment 140433

Elam251

View attachment 140434


TLM103

View attachment 140435


U47

View attachment 140436

U67nos

View attachment 140437

U87nos

View attachment 140438

This is awesome! I'm guessing you can't measure polar response, right? Presumably the polar response is reasonably similar to the original LDC's that are modeled here but of course the Sphere plugin shows significant variations in that.

Do you think the Hemisphere plugin is doing anything other than just applying a high resolution EQ curve to achieve this?

 

[kingkorg]

This is awesome! I'm guessing you can't measure polar response, right? Presumably the polar response is reasonably similar to the original LDC's that are modeled here but of course the Sphere plugin shows significant variations in that.

Do you think the Hemisphere plugin is doing anything other than just applying a high resolution EQ curve to achieve this?

Polar response should be pretty close for mics with closely related capsules. So c800, u67, u87 you shouldn't worry about. U47, and AKG mics are different. Hepisphere uses just Eq in this case.