Better Circuit for the Original Capsule from the MCA SP-1?

Shop deals on ebay
shop deals on amazon
Advertise with us
GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

Paul678

Well-known member
Joined
Mar 3, 2013
Messages
97
Ok, installed my new RK-47 (along with almost all the Jim Williams mods), and....

It sounds great. Much fuller low end, and fuller low-mids. Less
piercing highs. Much smoother overall. I can hear the 2-8kHz
bump, but it's not too bad. It's definitely a better mic, and sounds
damn good to me.

Having said that, I still think the original mic could be useful in
certain situations. The original capsule with the upgraded
electronics was more shrill, but that too, could be useful. The
excess brightness or presence could be something that you
would end up EQing in anyways, to cut through a mix.

The original capsule is a small diameter one, right? I was thinking
it would be fun and interesting to build another circuit for it, and
find another body for it, and make yet another mic. Perhaps the
same Schoeps circuit, but with caps optimized for lower frequencies
(C3, C4 at 1uF), and more high frequency roll-off (maybe larger
values for C15 and C16), and maybe the capsule would be
more useful, and flatter. Although I'm sure there's a limit to how
much low frequencies you can get out of a small diameter capsule
versus a large one.

Perhaps someone has a better idea about how to use the original
capsule? A circuit that would match it better than the Schoeps?

Thanks for your time.....
 
There has always been a lot of discussion about increasing the values of the capacitors between the FET and the PNP transistors.
In fact, the transistors are nothing more than emitter followers, where the emitter resistors are the 6.8K phantom power resistors.
The input impedance of an emitter follower is Hfe * Re.
Let's assume a very low Hfe of 100. (Not likely!) Then the input impedance of the transistor is 100 * 6.8K = 680 K.
With the 150 K resistor on the base in parallel this would be (680 * 150) / (680 + 150) = 123 K.
To find the low - 3 dB point, we have to calculate the frequency for a reactance of 123K.
This equals 13 Hz...
So, unless you want to record subsonic audio, not very much improvement is to be expected.
It can even have the unwanted effect of 'blocking' the input amplifier, if you have wind noise (or air from breathing) in the microphone.

(Yes, I know, there are two capacitors and two transistors... But even then the low -3 dB point won't be higher than 20 Hz.)
 
RuudNL said:
There has always been a lot of discussion about increasing the values of the capacitors between the FET and the PNP transistors.
In fact, the transistors are nothing more than emitter followers, where the emitter resistors are the 6.8K phantom power resistors.
The input impedance of an emitter follower is Hfe * Re.
Let's assume a very low Hfe of 100. (Not likely!) Then the input impedance of the transistor is 100 * 6.8K = 680 K.
With the 150 K resistor on the base in parallel this would be (680 * 150) / (680 + 150) = 123 K.
To find the low - 3 dB point, we have to calculate the frequency for a reactance of 123K.
This equals 13 Hz...
So, unless you want to record subsonic audio, not very much improvement is to be expected.
It can even have the unwanted effect of 'blocking' the input amplifier, if you have wind noise (or air from breathing) in the microphone.

(Yes, I know, there are two capacitors and two transistors... But even then the low -3 dB point won't be higher than 20 Hz.)
Click to expand...

Ok, I followed you up until calculating the frequency for a reactance of 123K. 

How did you get 13Hz? 

Oh yes, I meant to say C15 and C14 for the high frequency roll off caps.

 
Xc = 1/ (2 * Pi * f * C)  =>  f = 1 / (2 Pi * Xc * C)

f = 1 / (6.28 * 123 K * 0.1 uF)    Because 0.1 uF is the 'Schoeps' value in most designs.

f = 1000000 / (6.28 * 123000 * 0.1) = > 1000000 / 77244 =  12,945989332504790016053026772306 Hz.

In case your microphone has capacitors with a lower value than 0.1 uF, it may help to increase the value, but not more than necessary!
You don't want a 'flat' response down to 1 Hz!
Also the Hfe is a worst case scenario! Most transistors have a much higher Hfe.

You will notice that increasing the value of C14 and C15 does not change very much, they are mainly there to prevent (V)HF amplification.
You will have to go as high as 47 nF (or so), to get an audible HF rolloff.
 
Nico Veth, the Dutch guy who made the well-known MCA SP-1 and MXL 990 mod site (site is down at the moment) uses the stock capsule, too.

Another Dutch guy, who services my vintage mics, has been doing an MCA SP-1 mod from day one. He makes a new PCB, but uses the stock the capsule.

While I'm at it, a third Dutch guy is our own Ruud. He's a rising star in mod land and, as can be seen in reply #3, he knows a thing or two.  8)
And I can tell you he's a nice guy to boot.

The fourth Dutch guy in this story would be me. And I ............ um...... love microphones.  ;D
 
I have 2 SP-1's that I took to Jim Willams personally to have Modded.
They both have the orginal capsules with his modded Circuit.
I was told not to change the capsule do to the fact that
Jim modded the circuit for the orginal capsule.
They sound great.
 
RuudNL said:
Xc = 1/ (2 * Pi * f * C)  =>  f = 1 / (2 Pi * Xc * C)

f = 1 / (6.28 * 123 K * 0.1 uF)    Because 0.1 uF is the 'Schoeps' value in most designs.

f = 1000000 / (6.28 * 123000 * 0.1) = > 1000000 / 77244 =  12,945989332504790016053026772306 Hz.

In case your microphone has capacitors with a lower value than 0.1 uF, it may help to increase the value, but not more than necessary!
You don't want a 'flat' response down to 1 Hz!
Also the Hfe is a worst case scenario! Most transistors have a much higher Hfe.

You will notice that increasing the value of C14 and C15 does not change very much, they are mainly there to prevent (V)HF amplification.
You will have to go as high as 47 nF (or so), to get an audible HF rolloff.
Click to expand...

Ok, that's what I thought you meant, 100nF for C4 and C3.

Didn't you solve for the 1/2 voltage divider, or the 20*Log(0.5)= -6dB  point?
Or is it just close enough to -3dB?

The stock values on my SP-1 was 330nF, like the ZAPNSPARK mod schematic for
generic "china" condenser, on Yahoo micbuilders group.

And additionally, some mods increase C4 and C3 to 680nF or even 1000nF.  Don't know if
anyone has complained about a problem.

Ok, so in theory, we could use C14 and C15 to get high frequency roll off.  What about the 22nF
C1 and C2?  They are shunts to ground, so they look like they could do the same job, right?

Also, how would you set up an equivalent circuit where you can simulate the effect of these caps,
that I could put into a simulator?  I have Microwave office, which is overkill, but I've done passive audio
work in it before.  I assume you would make the node between R3 and R4 an AC short to ground?
Can you also simulate the effects of the bipolars as well?  Try different transistors?

Thanks so much for your advice.....
 
micaddict said:
Nico Veth, the Dutch guy who made the well-known MCA SP-1 and MXL 990 mod site (site is down at the moment) uses the stock capsule, too.

Another Dutch guy, who services my vintage mics, has been doing an MCA SP-1 mod from day one. He makes a new PCB, but uses the stock the capsule.

While I'm at it, a third Dutch guy is our own Ruud. He's a rising star in mod land and, as can be seen in reply #3, he knows a thing or two.  8)
And I can tell you he's a nice guy to boot.

The fourth Dutch guy in this story would be me. And I ............ um...... love microphones.  ;D
Click to expand...

Ok, I could use the new PCB, and a new body, and a new headbasket, etc!  I only have the
original capsule of the SP-1!  Where can I buy the new PCB?  What is the schematic, if
it's not Top Secret?  :p 

Or are their mods very similar to the Jim Williams mods (replace T3 w/J305, BC559C for T1 and T2, etc)?

 
Winetree said:
I have 2 SP-1's that I took to Jim Willams personally to have Modded.
They both have the orginal capsules with his modded Circuit.
I was told not to change the capsule do to the fact that
Jim modded the circuit for the orginal capsule.
They sound great.
Click to expand...

I'm sure he didn't tell you what he changed, eh?  Maybe you could open it up and
look.......    :D
 
Right, the capsule is now a leftover.
In fact it's the same thing that's in several  Chinese sdc (pencil) mics. So would removing the ring and turning it into a pencil mic be an option? Or would that make the hardware (housing) trickier? In any case it would "improve" the off axis behavior. Just thinking out loud. Ignore if stupid.

Oh, and yes, Jim Williams fits some SP-1s with an RK-12, some an RK-47 and in some he leaves the stock capsule. This way he created a locker for himself, too.
Unfortunately, he deleted most of his posts concerning this matter.
 
micaddict said:
Right, the capsule is now a leftover.
In fact it's the same thing that's in several  Chinese sdc (pencil) mics. So would removing the ring and turning it into a pencil mic be an option? Or would that make the hardware (housing) trickier? In any case it would "improve" the off axis behavior. Just thinking out loud. Ignore if stupid.

Oh, and yes, Jim Williams fits some SP-1s with an RK-12, some an RK-47 and in some he leaves the stock capsule. This way he created a locker for himself, too.
Unfortunately, he deleted most of his posts concerning this matter.
Click to expand...

You mean remove the (brass?) ring on the front?  That has a mesh on it, so it helps protect the diaphragm,
but the large diameter capsules don't have this, so maybe it's ok to remove it. 

Yes, someone informed me that Jim currently prefers the RK-12 in his SP-1s.  He supposedly has 12 of them, with
different flavors and colors.  Too bad he decided to cash in on his knowledge...I guess I would have done the same!    :p

EDIT:  It looks like taking the capsule apart is very risky, and will likely
ruin the capsule:

        http://tech.groups.yahoo.com/group/micbuilders/message/19520
 
Nico Veth, the Dutch guy who made the well-known MCA SP-1 and MXL 990 mod site (site is down at the moment) uses the stock capsule, too.
Click to expand...

Nico Veth is 'still alive', but the URL of his website has changed.
This is the new URL.

In the meantime I have done some 'real life' measurements with different capacitors.
Rsource = Rdrain = 2200 ohm.

With capacitors of 22 nF there is almost no effect on the frequency response: -1 dB at 20 KHz.

With 47 nF I measure:
11 KHz: -1 dB
20 KHz: -2 dB

With 100 nF the situaltion is:
6 KHz: -1 dB
11 KHz: -2 dB
16 KHz: -3 dB
20 KHz: -4 dB

If you really want to hear a difference, connect a 100 nF capacitor between drain and source:
4 KHz: -1 dB
6 KHz: -2 dB
9 KHz: -3 dB
11 KHz: -4 dB
14 KHz: -5 dB
15 KHz: -6 dB
18 KHz: -7 dB
20 KHz: -8 dB

 
RuudNL said:
Nico Veth, the Dutch guy who made the well-known MCA SP-1 and MXL 990 mod site (site is down at the moment) uses the stock capsule, too.
Click to expand...

Nico Veth is 'still alive', but the URL of his website has changed.
This is the new URL.

In the meantime I have done some 'real life' measurements with different capacitors.
Rsource = Rdrain = 2200 ohm.

With capacitors of 22 nF there is almost no effect on the frequency response: -1 dB at 20 KHz.

With 47 nF I measure:
11 KHz: -1 dB
20 KHz: -2 dB

With 100 nF the situaltion is:
6 KHz: -1 dB
11 KHz: -2 dB
16 KHz: -3 dB
20 KHz: -4 dB

If you really want to hear a difference, connect a 100 nF capacitor between drain and source:
4 KHz: -1 dB
6 KHz: -2 dB
9 KHz: -3 dB
11 KHz: -4 dB
14 KHz: -5 dB
15 KHz: -6 dB
18 KHz: -7 dB
20 KHz: -8 dB
Click to expand...

    Wow, thanks Ruud.  The 22nF caps you are talking about are the C1 and C2 shunts just on the
output pins of the XLR, correct?  Do you use some sort of FFT analyzer to measure this?

    It looks like it really works then.  But which one gave you the best overall frequency response,
to your ears?  The 47nF or the 100nF?  Or did you prefer the stock value?

    Putting a cap across the drain  and source of the FET sounds a bit radical, but I'll try
it if increasing C1 and C2 is not enough for me.

    Thanks much!!!
 
The capacitors I am talking about, are the ones in parallel with the Collector-Base resistors. (Usually 150K.ohm)
In fact for AC they are in parallel with the drain and the source resistor.

Just for fun, I have tested a MCA-SP-1 with a 100 nF capacitor between drain and source. (So: -6 dB at 15 KHz.)
The result was a very 'flat' sound, lacking a lot of treble.
Certainly NOT the type of microphone I would prefer!
The original capsule might be a bit on the bright side, but I prefer that really over a microphone that has a reduced high end.

The capacitors over the outputs are only there to prevent (V)HF entering the circuit.
Personally I would use the capsule as it is, and not modify the frequency response.
IMHO there is nothing wrong with the Schoeps schematic. Those guys knew what they were doing!
The only thing you have to take care of (especially if you are using a transformer input), is to match the transistors.
I you don't do that, there may be a DC offset between the two wires going to the amplifier, causing core magnetisation of the transformer.
A couple of millivolts is not a problem, but I have seen values of 150-200 mV with commercial microphones!
(Electronically balanced inputs are much more tolerant for DC offsets.)

I have seen many MCA-SP1 microphones, but  NONE of them had the FET bias set right!
No surprising, they use the same resistor values for all microphones!
To do it right, you should adjust the bias for each individual FET.
Personally I use a trimpot and a distortion meter. At a relative high signal at the gate of the FET, I then adjust the pot for minimal distortion. Then I measure the resistance of the trimpot and put in resistors with the values I measured.
You can also use an oscilloscope, but with a distortion meter you will get more exact results.
 
RuudNL said:
The capacitors I am talking about, are the ones in parallel with the Collector-Base resistors. (Usually 150K.ohm)
In fact for AC they are in parallel with the drain and the source resistor.

Just for fun, I have tested a MCA-SP-1 with a 100 nF capacitor between drain and source. (So: -6 dB at 15 KHz.)
The result was a very 'flat' sound, lacking a lot of treble.
Certainly NOT the type of microphone I would prefer!
The original capsule might be a bit on the bright side, but I prefer that really over a microphone that has a reduced high end.

The capacitors over the outputs are only there to prevent (V)HF entering the circuit.
Personally I would use the capsule as it is, and not modify the frequency response.
IMHO there is nothing wrong with the Schoeps schematic. Those guys knew what they were doing!
The only thing you have to take care of (especially if you are using a transformer input), is to match the transistors.
I you don't do that, there may be a DC offset between the two wires going to the amplifier, causing core magnetisation of the transformer.
A couple of millivolts is not a problem, but I have seen values of 150-200 mV with commercial microphones!
(Electronically balanced inputs are much more tolerant for DC offsets.)

I have seen many MCA-SP1 microphones, but  NONE of them had the FET bias set right!
No surprising, they use the same resistor values for all microphones!
To do it right, you should adjust the bias for each individual FET.
Personally I use a trimpot and a distortion meter. At a relative high signal at the gate of the FET, I then adjust the pot for minimal distortion. Then I measure the resistance of the trimpot and put in resistors with the values I measured.
You can also use an oscilloscope, but with a distortion meter you will get more exact results.
Click to expand...

  Oh, I see.  You were changing C15 and C14.

  It looks like 47nF might work for me.

  I agree about a bright mic being preferred over a dull sounding mic.  You can always subtract
high frequencies, but sometimes you cannot add them without major distortion.

  The high-end pro-audio snobs always talk about a "natural", flat frequency response as the
ideal for a microphone.  But in an actual mix, you almost always boost the high frequency presence
of the vocals or guitars ANYWAYS!  Just so you can hear them through the mix!

    But yeah, with the stock capsule, and the Jim Williams mods on the electronics, the SP-1 was
still a bit bright for me.

    I re-biased with the oscilloscope method when I replaced T3 with a J305, as shown by ZAPNSPARK
in the micbuilders group on Yahoo:

        http://farm9.staticflickr.com/8248/8561702277_d2da4f614a_b.jpg

    I didn't try to match the BC559C replacements for the bipolars, but because they
were from the same lot, I'm sure they aren't too different.



       
 
  I didn't try to match the BC559C replacements for the bipolars, but because they
were from the same lot, I'm sure they aren't too different.
Click to expand...

Whohohoooooo!...... I have tested hundreds of transistors from the same lot, and you would be surprised about the differences!
In general I only found 5 or 6 identical pairs out of 100 transistors!

 
RuudNL said:
  I didn't try to match the BC559C replacements for the bipolars, but because they
were from the same lot, I'm sure they aren't too different.
Click to expand...

Whohohoooooo!...... I have tested hundreds of transistors from the same lot, and you would be surprised about the differences!
In general I only found 5 or 6 identical pairs out of 100 transistors!
Click to expand...

When you say "identical", how close are you talking?    You are mainly measuring
Hfe, right?

I'm not sure what kind of pre-amps are in my Tascam US-1641, but so far,
I haven't noticed a problem with the microphone.

 
I test them for Hfe, Ube and DC offset as a pair in a test circuit.
DC offset (across the output) should be < 2 mV in my tests.
Finally I measure them for noise.
 
RuudNL said:
I test them for Hfe, Ube and DC offset as a pair in a test circuit.
DC offset (across the output) should be < 2 mV in my tests.
Finally I measure them for noise.
Click to expand...

Can I see your test circuit, if you have a schematic handy?  And your
procedure for these tests? 

How do you check for noise?  What test equipment you use?

 

Similar threads

Sebass
Finding the best Capsule and the Video: "Where Does The Tone Come From In A Microphone?"
Replies
13
Views
1K
Paul W
P
K
Diy Push-Pull symetrical undamped F8 capsule
2
Replies
36
Views
4K
adamasnan
A
OneRoomStudio
Cleaning a dynamic capsule with magnets
Replies
2
Views
421
frans
frans
A
I want to make a reliable clone of a C800G microphone
4 5 6
Replies
100
Views
9K
Khron
Khron
W
Mic Circuit housed in separate enclosure?
Replies
11
Views
2K
Lampie519
Lampie519

Latest posts

Back
Top