Sony C-37A questions

[Marik]

[quote author="soundguy"]AKG C60, C12a and lomo 19a9 are mics I have that are cathode follower (I thnk). They tend to sit behind regular tube mics really nicely. They dont have the most presence in the world and I can see why people might think they are noisy or not the best idea, but used strategically, they can make a mix fall into place with little effort. That might be one advantage to a CF design.

dave[/quote]

Dave,

Yes, you are right, they are all CFs. I worked on all of them many times (and had 19a9). The problem with presence is that they have old 'n sh*tty electrolytic caps from transformers. Change it to nice films and you'll see what they are capable of. Is your C12A still with an old brass style CK12? If yes and capsule is good, with nice cap and good parts it should be a terrific mic.

 

[starfish]

Hi, I'm new to the forum and somewhat new to DIY. I love this place and have learned (lurked?) a lot in the last 6 mo, but still feel like I'm scratching the surface. I've modded a pair of 2001's with an elam type circuit that I'm very pleased with(thanks Duka, Gus), and done a few other small projects-and I have 100 questions, but I'm going to start out with some info. A friend loaned me his C37a and I got to check out the power supply. The secondaries on the power transformer are 250-0-250@5ma / [email protected] for the rectifier / [email protected] for the heater (the light is tapped off the 6v side. I think the resistors are 3/4 watt wirewound (they have 3/4 written on them) except for the 45ohm resistor. This is actually four resistors in series-15-10-10-10. These have 5L 10ohm written on them. The only info I have on the audio transformer is that it's a Tamradio L1-D (pat# 200573, 412764). Does anyone know the ratio? Thanks for all the great info everyone :thumb:

 

[codered]

Anybody know the correct ratio for the tamura.I'm trying to build one using Dale's capsules

 

[zebra50]

[quote author="codered"]Anybody know the correct ratio for the tamura. I'm trying to build one using Dale's capsules[/quote]

No, but I've built a handful of K-followers with 2:1 ratio working very well. This is higher B+ than some, so you may need to go higher. I'd probably start at 3:1 or 4:1 - whatever you have. Maybe play with a 1+1:2+2 for lots of options.

Depends what you're feeding, of course.

z50

 

[zebra50]

Have we talked about the PSU here?

I was thinking about the rectifiers. One is solid state, the other is tube. I was wondering if this was to engineer a slow-start to the head amplifier? So that it sees the filament supply before B+.

Also note that half x B+ is referenced to half of f+.

A lot of the DIY mics we see here have F- and the audio earth connected. It's nice to look at a different approach.

Stewart

 

[Rossi]

I also got two of Dale_s Sony-style capsules that I hope to try out soon. Does anybody know the polarization voltage that Sony used on their capsules? I see that they used a DC-DC converter, but I can't find a voltage rating on the schematic.

As Dale said, the Sony capsules are quite low output, so a quiet head amp and optimal capsule polarizations are a must for decent S/N.

 

[zebra50]

Rossi,

I hooked up one of Dales capsules this week to a standard cathode follower with polarisation simply through grid leakage. It works a treat & sounds very smooth.

Yes, the output is noticeably lower than the same circuit with one of the 747 Neumann clone capsules, but the signal to noise is not a problem at all. And that's without any effort at opimisation.

I may try a close C37 clone circuit next week. I built a 6AU6a mic a few years ago but just used a regulated G7 style PSU.

 

[Rossi]

Thanks, looks easy enough, I even have a few 5840s lying around. :thumb:

Still, I'm curious about the original polarization voltage. I'd like to experiment with a FET circuit and a DC-DC converter & see what's possible regarding noise performance.

 

[Gus]

http://www.rcrowley.com/SonyC37/index.htm

we can make a guess.

If you look at the CF there is a 3K to bias the tube a 6au6 might want say 2V drop for .66ma
then the 100K and .66ma would give about 66V

I have not measured or used 37 so the above is a guess

 

[zebra50]

Cheers Gus - it's a sensible guess.

(I should have noted in the schematic above that in this case I added B+ and f+ stabilisation caps in the mic itself.)

 

[CJ]

you need the rein narma circuit for that.

 

[MatthewF]

Hi everyone,

I've made a few attempts at working out figures for B+ / plate current / capsule polarisation voltage in the C37A and keep ending up with some pretty high figures. Please correct me if I'm wrong (I'm a tube newbie) but from what I can tell B+ sits about 230v, with idle plate current at 1mA which sets capsule polarisation at a rather high 100v?

I arrived at the above conclusions after noting starfish's observation that the HT secondary on the power transformer is 250-0-250v. If we take my guess of 230v B+ then we see 2.3mA flowing through the two 50k series resistors which are used to reference heater supply to 1/2 B+. That combined with my estimated 1mA quiescent plate current suggests a total draw on B+ of 3.3mA.

The 6X4 rectifier, with 250vac on its plates and a current draw of 3.3mA, will give ~330vdc (see excerpt from the 6X4 datasheet):

3.3mA through the 30k of B+ filter network resistors will drop 100v, hence my suggestion of 230v B+. With the suggested 1mA quiescent plate current, the bias at the 6AU6's grid would be -3v. A brief load line analysis of the tube stage is shown below, with saturation current at 230v / 103k = 2.2mA and cutoff voltage at 230v. The (imaginary) -3v grid bias line looks to me like it would intersect the load line at around 1mA plate current, hence my original suggestion of a 100v capsule polarisation.

Some of us (including myself) will hopefully soon have new C37 capsules from Dale so I thought this might be a good time to bring this topic back up. Maybe Dale could comment on his findings with regards to the ideal polarisation voltage for his capsules?

As I said before, I'm new to this stuff so please correct me where necessary!

Cheers,

Matthew

 

[rodabod]

Bump.

That looks like it adds up to me assuming the grid bias is indeed 3V which looks like it would sit fairly neatly in the middle of the load line.

What would be very handy obviously is if someone could meter inside the mic when powered.

The high polarisation voltage would, I suppose, be explained because of the use of a lower sensitivity capsule. But then is that the only reason for using a high B+? I can't see any other reason. I thought it would be simpler to use a lower B+ to avoid having to raise the heater and then derive polarisation from that and couple to the grid with a cap. Still, nice to save on using a cap though. I wonder if a larger load resistor with lower B+ would increase distortion in the same degree as with plate-out stages. But then maybe factoring-in the NFB this might not be such a problem.

Might be worth listening to see if bypassing the bias resistor changes the sound in any good way.

Roddy

 

[dale116dot7]

Normally, 60-80 volts is good for almost all capsules. Some Neumanns (K67) and the AKG C12's will tend to suck in much above 60V, or if you 'pop' the mic, the diaphragm will become stuck to the backplate and require a mic power-down to unstick it. Normal diaphragm tension should release the diaphragm. In extreme cases, the diaphragm can 'stick' to one of the holes in the mic but that's rare. I'd run the Sony at maybe 70 volts or thereabouts. You can 'measure' the capsule voltage in the 'Royer' circuit (an adaptation of the Altec M11 / M20) by measuring the cathode voltage and subtracting the offset based on the published tube curves. That's how the older high-impedance probes worked. It's not quite precise but will get you within a volt or two which is good enough. The Sony C38 used a DC-DC converter because it needed to - it's powered off a 9V battery so even to get 15V you'd need a converter.

The Altec M11/M20 with the 21-series capsule, as well as B+K measurement mics, use around 200V polarizing voltage, but that high of a voltage isn't even used on most SDC mics - many of them are 60V as well. The Altec M30, which, I believe, used an AKG capsule with a PE diaphragm, needed 60V so there was a conversion you could do to the glass capsule's power supply to lower the voltage to what the PE capsule needed.

 

[rodabod]

Thinking about this (and discussing with Matthew), I think that the main issue distortion-wise is possibly the dynamic load presented to the cathode. This maybe explains why Sony used a 4:1 transformer where as you sometimes see as low as 1:1 and 2:1 elsewhere.

The dynamic load might be as low as 16K if using a 4:1 output transformer and 1K input preamp, or worse still, 1K if we were to use a 1:1 transformer. What limits the current drive of the circuit? Is it the maximum level we can see for the original DC conditions load line (1mA) or do we take a new higher value from the dynamic load line?

Roddy

 

[Matt Nolan]

Old thread, I know, but it was useful to me yesterday and I'd like to add stuff to it which will help others in future.

I was helping a friend out with his newly acquired C-37A last night. His has the CP-3B power supply which is different from the CP-2 which you see more widely in the schematics available on the web. The manual at coutant.org has the CP-3B and you will see that, besides having silicon rectifiers on both B+ and heater, a different HPF implementation, plus a high cut and switchable output impedances, it also has a different biasing and polarisation inside the mic (depsite still being nominally a C-37A).

So, there are 2 versions of the C-37A. The earlier has a 3k bias resistor, the later has a 1k bias resistor.  All other components inside the mic are the same, and the supply voltages are nominally the same. The earlier PSU, the CP-2 elevates the heater to B+ * 1/2. The later PSU, the CP-3B elevates the heater to B+ * 2/3.

This tallies with the different cathode voltages. With the 3k bias resistor, the cathode is approximately at B+ * 1/2, as is the heater elevation. With the 1k bias resistor, the cathode is at approximately B+ * 2/3, as is the heater elevation. In this case, B+ was 210V, heater elevation 140V, cathode 140V, grid and capsule polarisation an eye-watering 138.6V

This is surprisingly high for LD capsule polarisation but, looking at the CP-2 version, even that has the capsule at around 100V. This checks out with SPICE too, but my 6AU6 model doesn't like the 1kOhm bias too much - massive LF roll off until the bias resistor gets above about 1k5, then it simulates as you'd expect.

With the later revision, you should get about 3dB more signal off the capsule. But, wow, 138V !!

Best,
Matt.

 

[shabtek]

i recall Marik mention the sony capsule requires a higher bias voltage because the distance from membrane to backplate is greater than most capsules we see.

 

[Matt Nolan]

i recall Marik mention the sony capsule requires a higher bias voltage because the distance from membrane to backplate is greater than most capsules we see.

That would make a lot of sense.

 

[OneRoomStudio]

I know this is an old/dead thread, but it's relevant to what I'm doing now. I'm trying to finally finish up a pair of mics based around Dale's C37 capsules, using a single PSU. Here's what I'm thinking:



My main question is: Is it ok to house R7 (bias resistor) in the PSU, or should it be in the mic. Will the placement effect noise at all? Also, if folks are wondering, I'm regulating the voltages so I can use either one mic or two mics simultaneously with the same supply.

 

[shabtek]

Consider what the length of cable contributes in the 2 configurations

I dont think dales capsule needs as high a bias voltage, also the current sony capsule has an insulating film between diaphragm and backplane