SWEEPS: Cinemag CMMI-5C vs Crimson CT-A10

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

BluegrassDan

Well-known member
GDIY Supporter
Joined
Mar 17, 2009
Messages
507
Location
Elizabethton, TN
Here are sweeps of the Cinemag CMMI-5C compared to the Crimson Audio CT-A10. This is Crimson's new copy of the UTC A-10.

These are hooked into my tube mic preamp. The graph rolls off at 30k, but that is due to the settings of my interface.

First comparison is wired 1:10. There is no load on the secondaries - straight to grid with a 1 meg grid leak resistor.

 

Attachments

  • Crimson vs Cinemag 1-10.jpg
    Crimson vs Cinemag 1-10.jpg
    98.4 KB · Views: 104
The CMMi-5C is an excellent transformer, but I wouldn’t call this necessarily an ad for it. All that the high end resonance on the A-10 says is that it isn’t loaded as it was designed to be. It typically had anywhere from around 47k to 96k across the secondary in classic units, zobel or no.

Take a look at Shabtek’s result with the A-10 with 100k on the secondary.

Good series of tests here, Dan. Thanks for doing them.
 
It's fascinating how sweeps only tell part of the story regarding how a transformer will actually sound. There is something about how these two transformers resolve the midrange differently. The only way I can describe it is the Cinemag has a "metallic" type of sound that comes across as a bit nasal.

I tend to obsess over sound to the point of OCD.
 
looks like Crimson stayed true to the UTC design,

that peak looks to be a result of "foldback" winding,

foldback means you wind a layer of primary wire from left to right, then, instead of going back right to left, you instead bring the wire back to the left side of the coil.

this usually takes about 1 turn. then you add another layer from left to right.

why do this?  you decrease capacitance losses.  when you wind back and forth, the start of each layer has twice the voltage between them than you would have  when using foldback winding.

not too many people nowadays use foldback winding, and it is possible that Crimson is getting that lively peak by some other method if their machine is not able to do foldback winding.

would love to see some 1 KC sq wave tests on the above transormers!
 
Tangent:  1st time I’ve seen this recently, a Collins 6F preamp which specifies multiple tube types for varying gain amounts, one tube type was flat on top before the rolloff, another had a resonant boost. 

The vast majority of old classic transformers show a top boost when measured with a pad, or direct from an interface.  It was used as a feature, counteracting treble rolloffs in long series amp systems.  It was also a feature used with droopy ribbon mic response. 

Loading greater than specified was used somewhat frequently to force broader smiley face response, like wiring a mic transformer meant for 150 source to the 600 source taps, and putting a intermediate (between matching and bridging) secondary load resistor. 

Altec is somewhat unique in their pre-1960 preamps for matching load resistances on secondaries. 
 
BluegrassDan said:
I’m not gonna lie. I actually prefer the sound of the Crimson without the load and the added boost.

Sure.  It’s only up a half dB at 10k, 1.5 at 20k.  Very subtle. 

Now we need you to test an A-10 in your setup. 
 
The saga continues. These show the frequency response when strapped 1:10 in series versus parallel.

Parallel looks more like the Cinemag at 1:10.

Both sweeps are with unloaded secondaries.
 

Attachments

  • Cinemag 1-10 series vs parallel.jpg
    Cinemag 1-10 series vs parallel.jpg
    94.1 KB · Views: 57
BluegrassDan said:
The saga continues. These show the frequency response when strapped 1:10 in series versus parallel.

Parallel looks more like the Cinemag at 1:10.

Both sweeps are with unloaded secondaries.

Yes, that would make sense as the primary in parallel is now expecting a source impedance of 150 ohms which is what you're sending it,  correctly so of course.
The lowest ratio you have when using an A10 as a 150 ohm mic input is as a 1:10.  Gain difference aside, whether you prefer it under-damped into the series connected 600 ohm input is another matter.  Whatever works in that case :)
 
Back
Top