The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Superlux S502 modified VS MK12 Condenser Microphones
Old 23rd March 2012
  #1
Gear Head
 

Superlux S502 modified VS MK12

Hello, I have modified the internal preamplifier of a Superlux S502 and perhaps, will test it later with a non modified one.

MK12 is in ORTF position with cardio capsules, the 2 takes are differents but levels are similar and fireface UC inputs were used.

Piano 1.wav

Piano 2.wav

Answer :

Piano 1 = Superlux S502
Piano 2 = Oktava MK12


Superlux modification is not easy because of the small place, but the epoxy board is very strong, so you can easily make tests.
- input capacitor changed from 470 pF ceramic to 560pF plastic film
- capsule wire shorten
- best isolation of the capsule wire wich was too close to the body
- capacitor 220nF 50V plastic film added in parallel of the 10µF 50V chemical capacitor (to replace with 47µF 63V) to decrease noise level. (not very usefull in fact and uneasy to find the place for it)
- capapacitor 10µF 50V removed and replaced by 47µF 63V, to have no problem with low frequency response (without that, you may have difference between chanels)
- All 4 ceramic output capacitors removed

following modifications were without effect (for my ears)
- 220nF capacitor added in parallel of 1Gohm/capsule
- 220nF capacitor added in parallel of the 2.2Kohm/FET/2.2Kohm 1st stage amplifier
theoricly may improve noise level

a small modification to avoid diaphony is to get straight the 5 wires from the board to the 5 pins XLR, originally, they are twisted together.

other very difficult improvements, tested but no place :
- 2 tantalum capacitors 0.1µF between 1st stage FET amplifier and the 2 PNP output current drivers, may be replaced by 2 film capacitors, seems to improve high frequencies response (but not shure)

After, just remember the very low price, but the difference, is that a high price mike does not need any modification, so if you want to spare money, just buy a soldering iron.

It needs very good eyes !!!!, very very small

The 2 takes are for me at the same technical quality level, it's just a question of artistic choice, but you will have some interesting opinions.

I will test also with a folk guitar.
Old 27th March 2012
  #2
Gear Head
 

Other tests (I had a problem of bad output transistor, so just a little more noise at the right of Superlux S502).

01 MK12 Take1 Fireface UC.wav
02 Superlux S502 Take1 M-Audio DMP3.wav

Here, inputs are exchanged, it's also to test the differences between Fireface UC mike inputs and M-Audio DMP3
03 Superlux S502 Take2 Fireface UC.wav
04 MK12 Take2 M-Audio DMP3.wav


Panoramic tests
05 MK12 Take3 Fireface Pano.wav
06 Superlux S502 Take3 M-Audio DMP3 Pano.wav

The 2 pairs are very close, no mask effect, so I don't hear great differences unlike other tests, I don't know if it is the modification of precise placement wich is responsable.

Just after this test, I had again noise problem (1 transistor was KO), so I simplified the circuitry making a non symetrical output stage like Oktava and I saw that there was a great problem with the value of a capacitor (10µF 50V to replace with 47µF 63V).

The problem is that the low frequencies may be different from one chanel to the other, the value is too low and bass restitution is good only if the 4 transitors are matched and absolutely identicals. So Superlux S502 may seems to have best sound, but it's only artificial phase problems, like a different EQ on channels. Now that my circuit is good, I will test it again.

Superlux S502 has obiously a good price, capsules seems interesting, but electronic circuitry needs to be improved

DMP3 gave me a good surprise with less diaphony than mic inputs of Fireface UC. My DMP3 was also modified with a iron plate between the 2 voices, higher capacitors for power supply and I cut also the wires of the light of the vu-meters wich needs too much power.
Old 27th March 2012
  #3
Lives for gear
 
didier.brest's Avatar
 

Quote:
Originally Posted by Jean30 View Post
I had a problem of bad output transistor, so just a little more noise at the right of Superlux S502
Wavelab measures the level of the starting silence 1 dB lower on right.
On the first test, I prefer the Superlux , which I feel being more neutral
Old 28th March 2012
  #4
Gear Head
 

Many thank and I'm OK with your opinion. I made new modifications and reduced symetric output to 1 transistor asymetrical output, like Oktava. I had more place to replace a tantalum capacitor by film larger value model and also solve the problem for bass frequencies (47µF better than 10µF), because original model works well if transistors are perfectly matched, if not, low frequencies are lost with not same response for left or right chanel.

Next test with new modification and compared with the Oktava MK12 to have the same reference.
Old 28th March 2012
  #5
Gear Addict
 
Bibster's Avatar
 

Jean,

Did you happen to draw a schematic of the amp? Eventually with the modifications you propose?

I don't quite get what you're trying to say about the 10uF cap.

Paul
Old 28th March 2012
  #6
Gear Head
 

Original circuitry is a copy of Schoeps and common to a lot of Chinese microphone production. 1st stage FET amplifier and 2nd stage current driver balanced output with 2 classic PNP transistors. In Superlux S 502, there is a capacitor of 10µF 50V wired from the collectors of the transistors to the ground, his fonction is to filter alternative current for the voltage supply of the FET and capsule polarization and other important function is to create a virtual ground for the transistors. Too low value will attenuate low frequencies of microphone signal. Theorically, the 2 transistors are matched and current summation gives 0, but in reality, there is always differences and capacitor need at least 47µF value to avoid that. You will find this minimum value for all circuitry.

I will scan my schemas in a few minutes, but I don't want some problem with Superlux, anyway, it's not new and I think it's now in public domain.
Old 28th March 2012
  #7
Gear Head
 





There, I had no choice because 1 PNP transistor was out of order and strange mark (2L), not easy to find, so rather than replace the 4 transistors, I prefered this more simple circuit. I used this opportunity to change the 0.1µF 50V tantalum capacitor with a 0.22 µF 63V plastic film.

About the 10µF, it's really critical. Minimal improvements are also input ceramic 470 pF capacitor to change with 560pF or more (1nF) 63V in plastic film type.

The first 2.2µF 50V tantalum in parallel with original 10µF 50V is absolutely stupid. 150K resistor can be 1M and following capacitor 0.22µF without the second 2.2µF capacitor.

In my modification with unbalanced outputs, notice that original 6.8K resistor, becomes 8.2K (1/8W if not MSC), so, FET drain has the same direct current, but 47µF capacitor allow more gain on 2.2K resistor. That's partially compensate the loss of 6dB because of the unbalance output.

I just also invert the phase (simply exchanging the XLR outputs), to have the same phase of my MK 12

I have now a larger and more regular response in low frequencies. High frequencies are clear and stereo image is better. Noise is the same than my Oktavas MK 12.

Next tests shortly
Old 28th March 2012
  #8
Gear Head
 

other views :






Old 29th March 2012
  #9
Gear Addict
 
Bibster's Avatar
 

Bonjour Jean!

The 4 output caps, are the ones marked L1..4,right? On the same side as the 10uF (1st photo)
They need to come off, and both 10uF will need to be replaced by 47uF ones.

Maybe replace the 1st 2.2uF (the one before the 150k in the polarisation circuit) by something to bypass the 10uF (Well, 47uF... :-) ) ?

I'm not that confident in my steady hand to replace the 150k resistor, the 2u2 caps are already small enough!

Paul
Old 29th March 2012
  #10
Gear Head
 

No, L1 to L4 are 47 ohm output resistors. Output capacitors are very small surface mounting components C17 C18 C19 C20, close to the output wires, on the other side.

When 10µF is replaced by 47µF, the 1st 2.2µF in parallel can be removed, you will have more place to exchange the 2 tantalum 0.1µF with plastic film 0.22 µF 63V.

You don't need to change the 150K resistor. I think this value is rather small because of the tantalum capasitor after. In other circuitry, I saw 1M to 30Mohms and 10nF

Every work need a very thin soldering iron and a loupe on the eye
Old 29th March 2012
  #11
Gear Head
 

Rachmaninov

Piano Rachmanoniv 1.wav

Piano Rachmanoniv 2.wav

I'm not in the good axis but great care was took for level, no effect on the signal.

All modifications were made on the Superlux circuitry. Now, outputs are unbalanced (because of the transistor out of order) and no tantalum in the signal path. For my opinion, I can hear small difference in the frequencies response, but very small, noise level are equals.

What do you prefer ? 1 or 2 ???
Old 2nd April 2012
  #12
Gear Head
 

Noboy have an opinion ???

Bad news, in europa (Tho.... in germany), price increased from 98 euros to 125 eurso !!!!

Next modifications: 1nF input capacitor instead of 560pF
exchange of the 4 unknown output PNP transistors with 2N 5087 SMD version wich have certainly best quality and low noise specifications
Old 2nd April 2012
  #13
Lives for gear
 
didier.brest's Avatar
 

My opininion is the same here like on Audiofanzine: I prefer 1 and I am thinking that it is the Oktava. Nice take anyway!
Old 3rd April 2012
  #14
Gear Addict
 
Bibster's Avatar
 

Well, I read on audiofanzine that 1 is NOT the oktava :-)

I will soon pass at the electronics-shop (elecdiff in TLS), hope they have small form-factor 47uF/63v capas as well as small plastic film 560u of 1nF's.

Pity I only have ONE S502.... would be nice to have something to compare to. No nice rooms & only a yamaha U1a here... :-(

Paul
Old 3rd April 2012
  #15
bee
Lives for gear
 
bee's Avatar
 

Thanks for the clips. I'm really impressed with the S502 for the money. I need to pick one up.
Old 3rd April 2012
  #16
Gear Head
 

Exact, Piano 1 is Superlx S502.
Oktava have directivity problem in ORTF couple, certainly also some particular resonnances. I prefer the omni capsules in AB takes, they are more linear.

For Bibster : before going to your favorite electronics shop, just take care to the type of component. I put 2 X 560pF capacitor (polystyrene film), but before, I tried a 1nF polyester film like this one :
B32529C102K - EPCOS - CONDENSATEUR 1NF 63V | Farnell France
I only had one for the test, but it seemed better because of the very high isolation (15Gohms) and bass frequencies are more present without excess. It's smaller than 560pF polystyren and in polyester film, you have only 470pF or 1nF value, so take 1nF.

All components have not the same qualities for each use.
For the 47µF63V (X2), this one have a small volume :
UPM1J470MPD - NICHICON - CONDENSATEUR 47UF 63V | Farnell France
8mm diameter and 11.5mm high

You can by also this transistors (X4), I didn't tried them already but I thing they are certainly better than originals and it can help you if you have a problem in the future (it's very low price) :
MMBT5087LT1G - ON SEMICONDUCTOR - TRANSISTOR, PNP, SOT-23 | Farnell France

To replace, the 2 tantalum 0.1µF, take this one (X4), not more, because of the very low volume available :
B32529C224K - EPCOS - CONDENSATEUR 0.22UF 63V | Farnell France
There is a mistake on specifications , it's 7.2mm X 6mm X 2.5mm, you can see that in data sheet. If you need more place, The first 2.2µF tantalum capacitor in parallel with the original 10µF can be removed when you will have the 47µF instead. that will allow you to place one of the 2 X 220nF to remplace the 0.1µF tantalum capacitor.

The more delicate replacement is about the transistors wich are very small surface mount devices. Wait for my test before doing that, I'have no choice because of my transistor malfunction but I will see if result is better with 2N 5087.

With internet shopping, you will pay shipping and minimum numbers of each elements, but with direct shopping, you will need only a few dollars.

Just remember that you have to take care to the capsule wire, very thin.
Check also the noise of each chanel, if one is bad, it may be like me a bad output transistor or worse, the input transistor.

You can simply know it putting a capacitor of 100µF 63V in parallel between the XLR 1 et 2, and after, XLR 1 and 3. Do that before putting phantom power on and wait after stopping phantom to unwire the capacitor, it needs to be discharged. Doing that test, you make a short circuit on 1 the 2 symetric outputs, level is 6dB lower, but you can hear if there is more noise with XLR 2 or 3. If you have a lot of noise in the 2 cases, the problem is the input FET transistor, if not, it's only one transistor of the output stage.
Old 4th April 2012
  #17
Gear Head
 

Components received today and all new and definitive modifications made. Better than ever.
Ouput stage transistors have to be replaced (with 2N5087G), noise is a little lower but the most important is the best presence and definition for high frequencies.
1nF 63V polyester film gives more low frequencies.
The microphone becomes warmer and more detailed. I'm happy !!!

Test in a few days.
Old 10th April 2012
  #18
Gear Head
 

Last tests on my Seagull folk guitar. Results are very similar. 2 takes were made because of my differrent inputs, DMP3 M-Audio is ecellent and RME mike input have a small diaphony.
I put again the ouptut capacitor C17/18/19/20 with ceramic SMD 10nF 63V. I ignored the original value, but sometimes it's 22nF, 10nf seems to me better.

10 MK12 Take A Fireface mike input.wav

11 Superlux S502 Take A DMP3 input.wav

12 MK12 S502 Take B DMP3 input.wav

13 Superlux S502 Take B Fireface mike input.wav

Superlux S502 has symetric output again and level is 2.5dB over the Oktava MK12, noises are identical, just a little better for the Oktava. Low frequencies response is larger for Oktava and Superlux modifications reduce the difference.

Taking a guitar in a small room with stereo ORTF system is not ideal, but some distorsions of the Superlux are removed, so, you can do some treatments after the record if you want but always keep a very straight original record. Without the modifications, there was a special sound usefull in certains ambiences, but not good for my opinion. The modifications reduces noise level, increases precision and bass response, sound becomes professional.

Capsules of Superlux are very linear and smaller than Oktava's. Cardioid response is larger than Oktava and takes more ambience room sound, the main difference is here. Oktava is an excellent choice with more uses, because Superlux is a fixed configuration, but, you can also use it with 1 microphone and for exemple a figure of 8 to do Mid/Side takes, it works perfectly.

Price increased froom 98 Euros to 125 for France and Germany, you can find it at 98 Euros for Neederland and Belgium but no more countries, here :
Superlux S 502 kopen? | Studio Microfoons - Superlux S 502
Old 20th April 2012
  #19
Lives for gear
 

Hello Jean -

Thanks for posting this. I found these at Mouser in the U.S.:

http://www.mouser.com/Search/Product...FR5471J63J11L4
Old 20th April 2012
  #20
Gear Head
 

It seems to have large dimensions, difficult to find a place with more than 3mm thickness. Insultion resistance is 500 Go, very good, but I prefer 1nF value and 100V because dimensions are the same.

The problem is to find all components and take more than you need (or for other projects), because you have often a minimum invoice, shipping fee and minimum value for some components (like SMD).

A recent record on my piano gave me very good result with the Superlux S502, I tried M/S with Oktava MK12, figure of 8 cardios and omni for the middle, it was good but there was a difference of sensitivity between the 2 cardios capsules, rectified with a Plugin (Wave S1 imager) using symetry adjustment. It works, but not the most easy system, it stays excellent for acoustic guitar.
Old 16th June 2012
  #21
Gear Nut
 

Quote:
Originally Posted by Jean30 View Post
Components received today and all new and definitive modifications made. Better than ever.
Ouput stage transistors have to be replaced (with 2N5087G), noise is a little lower but the most important is the best presence and definition for high frequencies.
1nF 63V polyester film gives more low frequencies.
The microphone becomes warmer and more detailed. I'm happy !!!

Test in a few days.
Hi Jean,

I would like to try these changes but I am getting a little lost the exact changes due to the way this thread is laid out, sorry!

Could you list the exact changes with the components in one final post to follow? Would really appreciate it!

Last edited by Simsy; 16th June 2012 at 01:32 PM.. Reason: Stupid iPhone corrective text!!!
Old 18th June 2012
  #22
Here for the gear
 

Quote:
Originally Posted by Jean30 View Post
It seems to have large dimensions, difficult to find a place with more than 3mm thickness. Insultion resistance is 500 Go, very good, but I prefer 1nF value and 100V because dimensions are the same.

The problem is to find all components and take more than you need (or for other projects), because you have often a minimum invoice, shipping fee and minimum value for some components (like SMD).

A recent record on my piano gave me very good result with the Superlux S502, I tried M/S with Oktava MK12, figure of 8 cardios and omni for the middle, it was good but there was a difference of sensitivity between the 2 cardios capsules, rectified with a Plugin (Wave S1 imager) using symetry adjustment. It works, but not the most easy system, it stays excellent for acoustic guitar.
A little off-topic... I have the S502 and am curious about making my own longer version of the supplied slitter cable, which is only about 1 meter. Ideally, a single cable about 100-ft. or 30.48 meters would be good to have for studio or on-location. I've found some special cable used by the audio-visual field to control devices. It is called a DMX cable and can be supplied with XLR 5-pin male and female connectors at the ends of the 100-ft. run. What I don't know is if anything gets twisted within the cable that would change the connections downstream. Also, will phantom power travel through this DMX cable as needed by the capsules? Has anyone tried using a DMX 5-pin cable for this type of break-out?
Old 18th June 2012
  #23
Here for the gear
 

Quote:
Originally Posted by soundcheck606 View Post
A little off-topic... I have the S502 and am curious about making my own longer version of the supplied slitter cable, which is only about 1 meter. Ideally, a single cable about 100-ft. or 30.48 meters would be good to have for studio or on-location. I've found some special cable used by the audio-visual field to control devices. It is called a DMX cable and can be supplied with XLR 5-pin male and female connectors at the ends of the 100-ft. run. What I don't know is if anything gets twisted within the cable that would change the connections downstream. Also, will phantom power travel through this DMX cable as needed by the capsules? Has anyone tried using a DMX 5-pin cable for this type of break-out?
I should add that the plan would be then to connect the 5-pin splitter at the equipment end of the 100-ft. DMX cable to convert to the two 3-pin XLRs to input into recording gear. Anyone who has laid out (and then coiled up) cable on location will appreciate the quest to reduce two cables into just one.
Old 18th June 2012
  #24
Here for the gear
 

Quote:
Originally Posted by soundcheck606 View Post
I should add that the plan would be then to connect the 5-pin splitter at the equipment end of the 100-ft. DMX cable to convert to the two 3-pin XLRs to input into recording gear. Anyone who has laid out (and then coiled up) cable on location will appreciate the quest to reduce two cables into just one.
OK. I've located information about the pin outs for DMX controller cables:
http://tinyurl.com/82skd52

Now, just need to know if this will match up with the pin outs of a typical 5-pin XLR audio connector like that used in the Superlux S502. Anybody?
Old 10th July 2012
  #25
Gear Nut
 

5 pin xlr = 5 pin xlr. application is up to the user.
Old 22nd November 2012
  #26
Gear Addict
@Jean30: could you post pictures of the modded pcb?
Old 28th January 2013
  #27
Lives for gear
 

Let's see if we can make sense of this thread.

He has posted two schematics. One is the "original" version with some of his modifications shown:

http://imageshack.us/a/img26/1773/01s502original.jpg

and one is for his unbalanced version:

http://imageshack.us/a/img191/4108/0...ncedoutput.jpg

Quote:
Superlux S502 modification is not easy because of the small space but the epoxy board is very strong, so you can easily make tests.

INPUT CAPACITOR
Plastic-film, polystyrene-film and polyester-film capacitors were tried. Polyester-film was best.
Input capacitor changed from 470 pF ceramic to 560pF plastic film. Minimal improvement with 560pF or 1nF 63V plastic film type.
I used 2 x 560pF capacitor (polystyrene film), but before, I tried a 1nF polyester film like this one: B32529C102K - EPCOS - CONDENSATEUR 1NF 63V | Farnell France.
I only had one for the test, but it seemed better because of the very high isolation (15 Gohms) and bass frequencies are more present without excess.
It's smaller than 560 pF polystyrene and in polyester film you have only 470pF or 1nF value, so take 1 nF.
1nF (63V) polyester film capacitors give more low frequencies.
The microphone becomes warmer and more detailed.
- Replace the 470 pF ceramic input capacitor with 1nF 100V 5% polyester-film capacitor: B32529C1102J EPCOS | Mouser


- Capsule wire shortened

- Better isolation of the capsule wire which was too close to the body

- Capacitor 10µF 50V removed and replaced by 47µF 63V, to have no problem with low frequency response (without that, you may have a difference between chanels)

- All 4 ceramic output capacitors removed

- A small modification to avoid diaphony is to straighten the 5 wires from the board to the 5-pin XLR; originally they are twisted together.

The following modifications were without effect (for my ears)

- 220nF capacitor added in parallel of 1Gohm/capsule

- 220nF capacitor added in parallel of the 2.2Kohm/FET/2.2Kohm 1st stage amplifier theoretically may improve noise level

Other very difficult improvements, tested but no place:
- Capacitor 220nF 50V plastic film added in parallel of the 10µF 50V chemical capacitor (to replace with 47µF 63V) to decrease noise level. (Not very useful and difficult to find a place for it)

- 2 tantalum capacitors 0.1µF between 1st stage FET amplifier and the 2 PNP output current drivers may be replaced by 2 film capacitors; seems to improve high frequencies response (but not sure)

UNBALANCED OUTPUT

Just after this test, I again had a noise problem (1 transistor was KO), so I simplified the circuitry making an unbalanced output stage like Oktava and I saw that there was a great problem with the value of a capacitor (10µF 50V to replace with 47µF 63V). The problem is that the low frequencies may be different from one channel to the other, the value is too low and bass restitution is good only if the 4 transistors are matched and absolutely identical. So Superlux S502 may seem to have the best sound, but it's only artificial phase problems, like a different EQ on channels. Now that my circuit is good, I will test it again.

I made new modifications and reduced balanced output to 1 transistor unbalanced output. I had more space to replace a tantalum capacitor with a larger value film capacitor and also to solve a problem with bass frequencies (47µF better than 10µF), because original model works well if transistors are perfectly matched; if not, low frequencies are lost with different response for left and right channels.

I had no choice because one PNP transistor was out of order and strange mark (2L), not easy to find, so rather than replace the 4 transistors, I opted for this simpler (unbalanced) circuit. I used this opportunity to change the 0.1µF 50V tantalum capacitor with a 0.22 µF 63V plastic film.

The first 2.2µF 50V tantalum in parallel with original 10µF 50V is absolutely stupid. 150K resistor can be 1M and following capacitor 0.22µF without the second 2.2µF capacitor.

In my modification with unbalanced outputs, notice that original 6.8K resistor, becomes 8.2K (1/8W if not MSC), so, FET drain has the same direct current, but 47µF capacitor allow more gain on 2.2K resistor. That's partially compensate the loss of 6dB because of the unbalanced output.

10µF CAPACITOR

<< I don't quite get what you're trying to say about the 10uF cap. >>

About the 10µF, it's really critical. When 10µF is replaced by 47µF, the 1st 2.2µF in parallel can be removed, you will have more space to exchange the 2 tantalum 0.1µF with plastic film 0.22 µF 63V.

The original circuitry is a copy of Schoeps and is common to a lot of Chinese microphones. 1st stage FET amplifier and 2nd stage current driver balanced output with 2 classic PNP transistors. In Superlux S502 there is a capacitor of 10µF 50V wired from the collectors of the transistors to ground. Its function is to filter AC for the voltage supply of the FET and capsule polarization. Another important function is to create a virtual ground for the transistors. Too low a value will attenuate low frequencies. Ideally the 2 transistors are matched and current summation gives 0, but in reality there are always differences and the capacitor needs to be at least 47µF to avoid that. You will find this minimum value for all circuitry.

For the 47µF 63V (X2), this one has a small volume:

UPM1J470MPD - NICHICON - CONDENSATEUR 47UF 63V | Farnell France 8mm diameter and 11.5mm high

I also inverted the phase (simply exchanging the XLR outputs) to have the same phase as my MK012.

I now have a larger and more regular response in low frequencies. High frequencies are clear and stereo image is better. Noise is the same as my Oktava MK012's.

<< The 4 output caps are the ones marked L1 through L4, right? On the same side as the 10uF (1st photo)? They need to come off, and both 10uF will need to be replaced by 47uF ones. Maybe replace the 1st 2.2uF (the one before the 150k in the polarisation circuit) by something to bypass the 10uF (47uF)? >>

No, L1 to L4 are 47 ohm output resistors. Output capacitors are very small surface-mount components C17, C18, C19, C20, close to the output wires on the other side.

You don't need to change the 150K resistor. I think this value is rather small because of the tantalum capacitor after. In other circuitry I saw 1M to 30 Mohms and 10nF

You can also buy these transistors (X4), I didn't try them already but I think they are certainly better than originals and it can help you if you have a problem in the future (it's very low price) :

MMBT5087LT1G - ON SEMICONDUCTOR - TRANSISTOR, PNP, SOT-23 | Farnell France

To replace, the 2 tantalum 0.1µF, take this one (X4):

B32529C224K - EPCOS - CONDENSATEUR 0.22UF 63V | Farnell France

There is a mistake in the specifications , it's 7.2mm X 6mm X 2.5mm; you can see that in the data sheet. If you need more space, the first 2.2µF tantalum capacitor in parallel with the original 10µF can be removed and you will have the 47µF instead. That will allow you to place one of the 2 x 220nF to remplace the 0.1µF tantalum capacitor.

The more delicate replacements are the transistors wich are very small surface mount devices. Wait for my test before doing that. I have no choice because of my transistor malfunction but I will see if result is better with 2N5087.

Just remember that you have to take care with the capsule wire; it is very thin. Check also the noise of each channel. If one is bad, it may be a bad output transistor like mine or worse, the input transistor. You can simply know it by putting a capacitor of 100µF 63V in parallel between the XLR 1 and 2, and after, XLR 1 and 3. Do that before putting phantom power on and wait after stopping phantom to unwire the capacitor; it needs to be discharged. Doing that test, you make a short circuit on 1 of the 2 balanced outputs, level is 6dB lower, but you can hear if there is more noise with XLR 2 or 3. If you have a lot of noise in the 2 cases, the problem is the input FET transistor, if not, it's only one transistor of the output stage.

BACK TO BALANCED OUTPUT

Components received today and all new and definitive modifications made. Better than ever. Superlux S502 has balanced output again and level is 2.5dB above the Oktava MK12, noise is identical, just a little better for the Oktava. Low-frequency response is larger for Oktava and Superlux modifications reduce the difference. Ouput stage transistors have to be replaced with 2N5087G), noise is a little lower but the most important is the best presence and definition for high frequencies. I'm happy !!! I put again the ouptut capacitor C17/18/19/20 with ceramic SMD 10nF 63V. I ignored the original value, but sometimes it's 22nF, 10nf seems to me better.

Without the modifications there was a special sound useful in certains ambiences, but not good in my opinion. The modifications reduce noise level, increase precision and bass response, sound becomes professional.

Capsules of Superlux are very linear and smaller than Oktava's. Cardioid response is larger than Oktava and takes more ambience room sound, the main difference is here.

It seems to have large dimensions, difficult to find a place with more than 3mm thickness. Insulation resistance is 500 Go, very good, but I prefer 1nF value and 100V because dimensions are the same.
What I get from all of this is:

Disregard the 560 pF value in the schematic.

- Replace 10 µF electrolytic. UPM1J470MPD Nichicon | Mouser

- Capsule wire shortened.

- Better isolation of the capsule wire which was too close to the body.

- A small modification to avoid diaphony is to straighten the 5 wires from the board to the 5-pin XLR; originally they are twisted together.

- If you are inclined to work on surface-mount components, replace four output transistors with 2N5087G.
Old 24th June 2013
  #28
Lives for gear
 
mixerguy's Avatar
has anyone else modified a Superlux S502?
Old 24th June 2013
  #29
Gear Addict
Yes, replaced ceramic input capacitor for polypropylene, resolderd smd caps and put isolation tape on capsule wire.
On recording a choir I know well, it sounded less harsh on the sopranos, so that's a gain to me.
Old 3rd April 2014
  #30
Here for the gear
 

Hey guys. I'm alittle confused about this thread. But I am interested as I have just purchased this mic, which I am quite pleased with except i feel the noise floor is a little high.

Are any of these mods specifically good for lowering the self noise?

And can anyone confirm what chris319 summarized as beeing the important bits as working? if so what does it gain?

thank you guys! Especially Jean30 for starting this and doing these tests.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
Gaston69 / Remote Possibilities in Acoustic Music and Location Recording
370
kirkagur / So much gear, so little time
3
Fhl / Low End Theory
9
A27Hull / Geekslutz forum
11

Forum Jump
Forum Jump