Universal Audio SC-1 Hemisphere Mic Modeling Is a Behringer

[kingkorg]

The latest modeling mic is basically a Behringer B-1. It uses the same capsule, it is made by 797 audio, they share the similar body, and measure 100% the same. Not sure if the circuit is exactly the same, but they are both flat circuits, and all the correction happens in the box. Good news is that this type of headbasket is more transparent, and nails the models 99% accurately FR wise, better than the original Sphere. The bad news is it's cardioid only, so no off-axis modeling, or multi pattern.


https://www.uaudio.com/uad-plugins/sc-1-standard-condenser-microphone.html

 

[wanderingengineer]

Hi kingkorg, thanks for posting this. Interesting isn't it? The UAD SC-1 gives the illusion of being "better" because of the cream painted body and lack of slightly fussy details on the B1, plus UAD's marketing of course. Which specific 797 capsule is it using, do you know?

Lagging behind you by many years (and probably many brain cells!) I am trying to build some L22/DLX/LX clones. I have an L22 from a few years ago and think it's decent, so I have all versions of the Sphere software. I have read your teardown and posts on this with great interest.

I am proposing to try to replicate the L22 (without the de-emphasis circuit) and use an EQ plugin before the Sphere plugin. My plan is therefore to use a 797 CY002 and a dual Alice OPA (to Jules Ryckebusch's design), in a Chinese 47 body with a 5 pin output. Do you have anything to suggest to get the best result?

One of the most expensive components is the capsule. USD 50 for each CY002. Not massive but. Do you think any other cheaper capsules might achieve similar results? Or am I best sticking with the CY002 in your view?

Do you think this still makes sense to build this? I am conscious of your comment that the SC-1 and Hemisphere is better at nailing the emulations than the L22.

Finally just to say thanks for your contributions, which I really enjoy!

Best
Adam

 

[kingkorg]

Hi kingkorg, thanks for posting this. Interesting isn't it? The UAD SC-1 gives the illusion of being "better" because of the cream painted body and lack of slightly fussy details on the B1, plus UAD's marketing of course. Which specific 797 capsule is it using, do you know?

Lagging behind you by many years (and probably many brain cells!) I am trying to build some L22/DLX/LX clones. I have an L22 from a few years ago and think it's decent, so I have all versions of the Sphere software. I have read your teardown and posts on this with great interest.

I am proposing to try to replicate the L22 (without the de-emphasis circuit) and use an EQ plugin before the Sphere plugin. My plan is therefore to use a 797 CY002 and a dual Alice OPA (to Jules Ryckebusch's design), in a Chinese 47 body with a 5 pin output. Do you have anything to suggest to get the best result?

One of the most expensive components is the capsule. USD 50 for each CY002. Not massive but. Do you think any other cheaper capsules might achieve similar results? Or am I best sticking with the CY002 in your view?

Do you think this still makes sense to build this? I am conscious of your comment that the SC-1 and Hemisphere is better at nailing the emulations than the L22.

Finally just to say thanks for your contributions, which I really enjoy!

Best
Adam

Hi Adam. Most of what you asked have i already posted somewehere on the forum. You can absolutely use EQ in post. Here's the curve.

https://app.box.com/s/nq2a3rsi961z8dimr6rra0t5fpm88by8

This is Takstar cts-2 capsule, it is just 1db different than 797 002. Not sure what the price is, you can contact them.

https://groupdiy.com/threads/whats-...ctured-capsules-direct-797.78212/post-1043198

I use Sphere software in literally every project, i find it invaluable for ambience because of all the pattern variations, and endless combinations regardless of the model accuracy. I even have one mic with CK12 capsule. Very similar to OpaAlice, but with true CK12.

 

[wanderingengineer]

Thanks very much. Apologies if I have missed stuff written elsewhere on the forum. I know that's can be a bugbear sometimes. I will try to absorb everything I can.

As to the CTS-2, I think the only challenge is sourcing them reliably but that's probably also been covered elsewhere.

 

[kingkorg]

Thanks very much. Apologies if I have missed stuff written elsewhere on the forum. I know that's can be a bugbear sometimes. I will try to absorb everything I can.

As to the CTS-2, I think the only challenge is sourcing them reliably but that's probably also been covered elsewhere.

Directly from Takstar, drop them an inquiry.

 

[wanderingengineer]

Apologies in advance if this should be in another thread but I wanted some help in my quest as I test some capsules out for dual output purposes.



I received some 797 CY002s and tested them with Jules Ryckebusch's Alice Dual OPA circuit, which seems to be de rigeur. Impressed with both of them to be honest!

Do you know whether 80V bias is within tolerance for them? I read that Townsend were using these with the Sphere L22 at 60V. The OPA circuit defaults to 80V so I wondered if that was ok? I have read that 6 micron mylar capsules will often tolerate 75-80V but not 3 micron (for which the considered limit seems to be 60V).

@kingkorg - thanks to your suggestion, I am also in touch with Takstar. They say their CTS-2 will only tolerate 80V if I "upgrade" to the capsule (I think the one used in the Lewitt 640) (rated at 100V) being a "high pressure resistance" capsule as opposed to "low pressure resistance" for the standard CTS-2.

They offer a CTS-2-01 and a CTS-2-640 accordingly but the price differential is > USD 20 (50 as opposed to 30).

Any views from the community?

 

[kingkorg]

Apologies in advance if this should be in another thread but I wanted some help in my quest as I test some capsules out for dual output purposes.

View attachment 138730

I received some 797 CY002s and tested them with Jules Ryckebusch's Alice Dual OPA circuit, which seems to be de rigeur. Impressed with both of them to be honest!

Do you know whether 80V bias is within tolerance for them? I read that Townsend were using these with the Sphere L22 at 60V. The OPA circuit defaults to 80V so I wondered if that was ok? I have read that 6 micron mylar capsules will often tolerate 75-80V but not 3 micron (for which the considered limit seems to be 60V).

@kingkorg - thanks to your suggestion, I am also in touch with Takstar. They say their CTS-2 will only tolerate 80V if I "upgrade" to the capsule (I think the one used in the Lewitt 640) (rated at 100V) being a "high pressure resistance" capsule as opposed to "low pressure resistance" for the standard CTS-2.

They offer a CTS-2-01 and a CTS-2-640 accordingly but the price differential is > USD 20 (50 as opposed to 30).

Any views from the community?

I don't thing there should be any issue with those and 80V.

 

[sumitagarwal]

Very new and inexperienced here and so probably a dumb-dumb question: does this mean that if somebody has the UA Hemisphere software and the B-1 that they would get the intended performance out of that software. And, when simulating mics in their cardiod setting, would outperform the Sphere?

I'm imagining a scenario where a person buys a cheap used UA SD-3 for the Hemisphere software (and maybe they need a kick mic) and then loads up on B-1's for most other applications.

 

[wanderingengineer]

I am probably not far off your inexperience level but my own research on this forum and in other places suggests that the answer is probably along the lines you suggest.

My focus has been on recreating a dual output “sphere clone” to benefit from the axis and polar pattern controls but given what @kingkorg has said, I think you could expect decent results from hemisphere using the approach you describe.

Obviously, it necessarily entails purchasing a microphone that gives you a license for the hemisphere plug-in. When the L 22 was produced by Townsend labs, their version of the plug-in was free and unrestricted up to version 1.5.2. This enabled someone to build a dual output/flat circuit microphone and use it with the plug-in and the HF de-emphasis EQ provided by @kingkorg and potentially get equivalent results. Or to use it with the free Austrian Audio or Lewitt plugins, if polar pattern jiggery was your goal instead of modelling other mics.

 

[sumitagarwal]

I am probably not far off your inexperience level but my own research on this forum and in other places suggests that the answer is probably along the lines you suggest.

Thanks for the feedback. Although looking at it some more I may be wrong in the detail that one could just buy as SD-3 as it the specific Hemisphere models may be gated by which mic serial numbers have been registered.

It's a shame that UA doesn't sell the software separately.

 

[wanderingengineer]

You could try a Behringer B1 with the free Townsend Labs 1.5.2 plugin and just ignore the “single output” pop-ups. That plugin can still be found at if you look hard enough. You might still want to use the EQ curve to de-emphasise the HF response before the plugin though.

Or build yourself something with a flat circuit and a 797/Takstar capsule!

 

[sumitagarwal]

You could try a Behringer B1 with the free Townsend Labs 1.5.2 plugin and just ignore the “single output” pop-ups. That plugin can still be found at if you look hard enough. You might still want to use the EQ curve to de-emphasise the HF response before the plugin though.

Or build yourself something with a flat circuit and a 797/Takstar capsule!

Thanks! I had no idea about that, tracking that down now. Not sure what you mean about the "single output" pop-ups... I assume it's that the plugin expects the two channels of the L22 and will complain when working with a single channel, but will ultimately apply the correct modeling anyway?

I would try building something but I'm already doing a bad job of taking on more build+repair hobbies instead of dropping ones. I need to rebuild my Harley, need to do a lot of repairs on cracks/etc of vintage guitars I've acquired, and I live in a very old house...

When buying used UA mics does anyone know exactly how the licensing process works? UA customer service registers the serial number to your account for free to make the models available and removes and disables the models from the previous owner's account?

 

[kingkorg]

Thanks! I had no idea about that, tracking that down now. Not sure what you mean about the "single output" pop-ups... I assume it's that the plugin expects the two channels of the L22 and will complain when working with a single channel, but will ultimately apply the correct modeling anyway?

I would try building something but I'm already doing a bad job of taking on more build+repair hobbies instead of dropping ones. I need to rebuild my Harley, need to do a lot of repairs on cracks/etc of vintage guitars I've acquired, and I live in a very old house...

When buying used UA mics does anyone know exactly how the licensing process works? UA customer service registers the serial number to your account for free to make the models available and removes and disables the models from the previous owner's account?

Yes, B-1 will perform exactly like the Hemisphere mic. When it comes to L22, no stock mic will outperform it.

Sure hemisphere mic serial is limited to one account?

 

[wanderingengineer]

@sumitagarwal You might also find this thread interesting? https://uadforum.com/community/inde...ware-without-the-mic.61338/page-2#post-506513

 

[sumitagarwal]

@sumitagarwal You might also find this thread interesting? https://uadforum.com/community/inde...ware-without-the-mic.61338/page-2#post-506513

Whoa whoa WHOA! This is absolutely glorious!!

So now I've got it running in Logic Pro. No registration, non proprietary license management, no proprietary hardware. Looks like I can just bang any cheap B-1 into any reasonable interface (got Shure MVX2U... not great noise floor, but crazy portable), and into any DAW and get close to vintage cardioid goodness for almost no money.

You guys rock!

 

[sumitagarwal]

Super interesting to click through the different models while routing the signal primarily to the "rear" input of the plug-in. I'm finding that the plug-in stops complaining if you simply put a 99% pan ahead of it, and then you can basically flip that pan to listen to how the plug-in processes the front capsule versus rear capsule.

Even for cardioid models there can be significant processing and mixing in of the rear capsule. The LD-47K is a good example of this, where you can essentially mute the "front" capsule and still get a significant signal through the "rear". Obviously this type of detail and modeling is what UA/Townsend are referring to when they position L22/Sphere as a unique product.

To get the most out of the dual-channel plug-in from single-channel mics it would be interesting to analyze the average difference in gain and frequency response between the two capsules in a Sphere when used as a conventional cardioid with a unidirectional source. From this it would be possible to come up with a pre-process for duplicating a front channel to a virtual rear channel for the Sphere plug-in to properly process. Maybe the Hemisphere plug-in is already doing pretty much this?

Without that, some of the models that have negligible output from the rear capsule include:

• LD-800
• LD-563 (mostly)
• DN-20
• LD-49K
• LD-67 NOS (mostly)
• LD-87 Modern (mostly)

Considering a B-1 can be had for ~$50, that ain't bad. But clearly a Sphere is needed to get the most out of this.

Of course, this doesn't account for how much the rear diaphragm might be suppressing the response of the front diaphragm. If the B-1/SC-1's single diaphragm is picking up much of the signal that is exclusively captured by the rear diaphragm on the Sphere, maybe the difference is smaller.

 

[kingkorg]

Super interesting to click through the different models while routing the signal primarily to the "rear" input of the plug-in. I'm finding that the plug-in stops complaining if you simply put a 99% pan ahead of it, and then you can basically flip that pan to listen to how the plug-in processes the front capsule versus rear capsule.

Even for cardioid models there can be significant processing and mixing in of the rear capsule. The LD-47K is a good example of this, where you can essentially mute the "front" capsule and still get a significant signal through the "rear". Obviously this type of detail and modeling is what UA/Townsend are referring to when they position L22/Sphere as a unique product.

To get the most out of the dual-channel plug-in from single-channel mics it would be interesting to analyze the average difference in gain and frequency response between the two capsules in a Sphere when used as a conventional cardioid with a unidirectional source. From this it would be possible to come up with a pre-process for duplicating a front channel to a virtual rear channel for the Sphere plug-in to properly process. Maybe the Hemisphere plug-in is already doing pretty much this?

Without that, some of the models that have negligible output from the rear capsule include:

• LD-800
• LD-563 (mostly)
• DN-20
• LD-49K
• LD-67 NOS (mostly)
• LD-87 Modern (mostly)

Considering a B-1 can be had for ~$50, that ain't bad. But clearly a Sphere is needed to get the most out of this.

Of course, this doesn't account for how much the rear diaphragm might be suppressing the response of the front diaphragm. If the B-1/SC-1's single diaphragm is picking up much of the signal that is exclusively captured by the rear diaphragm on the Sphere, maybe the difference is smaller.

View attachment 140279

View attachment 140280

If you are plugging in B1 straight into Sphere plugin, you need to use correction curve before it, not necessary for Hemisphere plugin.

Check this thread, and download CurveEQ correction curve i provided to get correct emulations.

https://groupdiy.com/threads/townsend-labs-sphere-guts-and-basic-analysis.77464/

 

[wanderingengineer]

If you like, I can give you an impulse response I made from kingkorg’s EQ curve which you can load into any convolution plugin before the Sphere plugin, saving you a licence for Voxengo CurveEQ.

 

[sumitagarwal]

If you like, I can give you an impulse response I made from kingkorg’s EQ curve which you can load into any convolution plugin before the Sphere plugin, saving you a licence for Voxengo CurveEQ.

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.

 

[wanderingengineer]

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