symetrix sx202 mod

[Ed Zeppeli]

Sorry to dig up a pretty dead thread but I just picked up one of these Symetrix SX202s for $25.

After digging out the right power supply (The store gave me a 19v DC one) I realize I couldn't get my condenser working on channel 1. It works great on 2 though. I tried an SM57 on channel 1 and it works there just fine.

I don't think it's too presumptuous to assume that the Phantom is not working on channel 1 given these facts.

Would this be a simple fix if I open up this thing?

Thanks,

Warren

[GLouie]

Measure the phantom power voltage on both channels. Normal is about 43vdc, IIRC. Troubleshooting shouldn't be too hard, it's not a complicated device. I've had electrolytic capacitors in these go bad.

Be sure to unlock the XLRs to open the chassis (screwdriver in the center hole of the XLR connector), as the plastic part is attached to the circuit board and the metal XLR frames stay with the sheet metal.

[alex79]

hey warren,

your power supply is wrong!!!
16v dc is the way to go.
a friend of mine used a 18v power supply and the sx202 didn't work correct.
btw, this is the right power supply if i rememer correctly:

SEYMOUR DUNCAN POWER SUPPLY 16V - Thomann UK Cyberstore

cheers,

alex

[Ed Zeppeli]

Thanks guys. I opened it up no problem. I didn't need a screwdriver for the XLRs for some reason. They just slid right out with a little fiddling. I think mine is revision G version IIRC.

I got 37.2 volts on the number 2 input w/ phantom power turned on and 0 on input 1. Although this may seem a little starved, the gain on this unit seems much better than the gain on my Firestudio Project. The sound is very clean as well to my ears and less harsh. I've just recorded an acoustic so far.

The power supply I'm using is 15vac (emphasis AC) and 800ma. The one the store gave me was DC which is incorrect according to the manual and the back panel.

I guess I'll try to trace the phantom to input 1 and see if I can find the short.

Thanks again for the info.

Warren

[Jim Williams]

16 volts AC, not DC. All Electronics in Van Nuys, CA has surplus 16 v AC 1500 ma worts for about 4 bucks.

[Ed Zeppeli]

Originally Posted by Jim Williams

16 volts AC, not DC. All Electronics in Van Nuys, CA has surplus 16 v AC 1500 ma worts for about 4 bucks.

Yep.

[alex79]

ac, of course - sorry!

[GLouie]

IIRC, the zener diode used to set the P48 is actually about 42v, less than spec. Perhaps this changed during revisions. I never experimented with changing the diode. Maybe the power supply voltage makes a difference for the P48. It could conceivably affect a mike sensitive to the P48.

Exercise all the controls (pots and buttons), as they get noisy.

[lancebug]

I know its an old thread, but it comes up near the top if you search google for info on the SX202, so I thought I would post this hear. Just finished modding my first 202, and recorded some a/b samples.

Directory Listing of /preampfiles/

It might not be the most useful comparison, as I don't have access to an actual 312/512 to run it against, but if people are curious about the general tonality, you might find it useful. I just finished my first SX202 mod (7 more pairs to go), so I shot them out at the same time. Hopefully some of you are familiar with those and can use them as a point of reference. All the samples are of the same performance and the signal path goes: AT Pro 37>preamp>MOTU 828>Logic. The files are 96K and 24 bit and levels were generally matched on the way in and subsequently normalized with no other processing or plugins. The Pro 37's might not be ideal for the vocal takes, but they are the only mics that I have 4 of.

The 4 preamps are:
•BLA B12a
•DIY EZ1290 (Pretty faithful Neve 1290/1073 clone, with Carnhill xfrs and Grayhill switches)
•Symetrix SX202 Stock
•Symetrix SX202 w/Monte Mods (new output opamps and different coupling and bypass caps)

The samples are:
•acoustic guitar (cheap fender)
•male vox
•Mono drum Overhead

I may add a few more samples later. The vocals ones might be a little hot and I think I might have some better ones in the session.

[blinkerss]

Originally Posted by lancebug

I know its an old thread, but it comes up near the top if you search google for info on the SX202, so I thought I would post this hear. Just finished modding my first 202, and recorded some a/b samples.

Directory Listing of /preampfiles/

It might not be the most useful comparison, as I don't have access to an actual 312/512 to run it against, but if people are curious about the general tonality, you might find it useful. I just finished my first SX202 mod (7 more pairs to go), so I shot them out at the same time. Hopefully some of you are familiar with those and can use them as a point of reference. All the samples are of the same performance and the signal path goes: AT Pro 37>preamp>MOTU 828>Logic. The files are 96K and 24 bit and levels were generally matched on the way in and subsequently normalized with no other processing or plugins. The Pro 37's might not be ideal for the vocal takes, but they are the only mics that I have 4 of.

The 4 preamps are:
•BLA B12a
•DIY EZ1290 (Pretty faithful Neve 1290/1073 clone, with Carnhill xfrs and Grayhill switches)
•Symetrix SX202 Stock
•Symetrix SX202 w/Monte Mods (new output opamps and different coupling and bypass caps)

The samples are:
•acoustic guitar (cheap fender)
•male vox
•Mono drum Overhead

I may add a few more samples later. The vocals ones might be a little hot and I think I might have some better ones in the session.

from your mod, are there any different replacement part with the monte mod?, like the op amp or resistor/cap value?

and can you post some loopback silent recording, between the stock and the modded one, i just want to hear does any noise reduction improvement between the stock and modded one

thanks

[lancebug]

My mods were the first two of the monte mods, pretty much as written. Changed some cap values around the big opamps and swapped out the little opamps for sockets which I filled with BB OPA3124's. I have some other opamps as well that I will probably put in some of the other 7 units I am working on. I will try to remember to do the noise test next time I am doing sound tests.

[blinkerss]

Originally Posted by lancebug

My mods were the first two of the monte mods, pretty much as written. Changed some cap values around the big opamps and swapped out the little opamps for sockets which I filled with BB OPA3124's. I have some other opamps as well that I will probably put in some of the other 7 units I am working on. I will try to remember to do the noise test next time I am doing sound tests.

cool, looking forward for the 202 loppback silent test

thanks

[Jim Williams]

Originally Posted by GLouie

IIRC, the zener diode used to set the P48 is actually about 42v, less than spec. Perhaps this changed during revisions. I never experimented with changing the diode. Maybe the power supply voltage makes a difference for the P48. It could conceivably affect a mike sensitive to the P48.

The phantom is done poorly in these as well as the 528. 42 volts is too low. They attempt to compensate by lowering the feed resistors from 6.81k to 4.99k ohms. A U-87 or Schoeps will not power properly with that.

To fix it, use a 47 volt Zener or I just added a 8.2v zener in series to kick it up to 48 volts. Then replace the 4.99k resistors with a matched pair of 6.81k ohm resistors. Now it powers Schoeps.

[lancebug]

I took a bit of time to reformat the original Monte mods in a way that was more readable for myself while looking at the pcb and the schematics, iron in hand. I broke them up into discrete sections by task and also added footnotes from other contributors as well, (again for my own use). I am posting my edit of Mr Maguire's original text for a point of reference. If anyone objects to this or if any of the subsequent quoted sources (such as Mr. Williams or Mr. Morse) would rather not be included, let be know and I will delete some or all of this post. Also if someone feels key information has been omitted, or otherwise should be included to make if more concisely printable, I will add it in as a footnote. I hope this is useful and that no one is offended by my re-posting of this information.


Modifications to the Symetrix SX202 Microphone Amplifier
Extracted from the original documents and text by Monte P McGuire

Summary:
The SX202 is a now discontinued but still very popular solid state IC mike amp made by Symetrix. It is based upon the now obsolete and hard to find SSM 2015 mike amp IC, and as such, it is a fundamentally good sounding amplifier of the "clean and uncolored" variety. Unfortunately, the circuit design and component choices made by Symetrix greatly reduce the performance that the 2015 is capable of. This paper outlines several methods for removing these performance road blocks and will allow a skilled modifier to greatly improve the performance of this amplifier. The end result will be an amplifier whose distortion products are low and largely benign and whose bandwidth and output drive will be quite extended.
Of course, these modifications require some electronic rework skill, will completely void your warranty and will change the way the SX202 sounds, so keep that in mind before you act upon what's written here.If you feel you're not equipped well enough to do these mods, take your SX202 and this paper to a competent technician, and they'll be happy to do the mods for you.

The five main issues that can be addressed in the SX202 are the output amplifier chip selection, the 2015 compensation capacitors, the coupling between the SSM 2015 and the output stage, power supply bypassing and the grounding scheme. I also discuss things you can do to improve the gain pot, coupling capacitors, a mod to remove the peak detector and finally a list of things you should not attempt to fix...!

One issue that has to be addressed is that there are essentially two different flavors of the SX202, each using different PC board layouts and component choices. So, the instructions and component designations may vary depending on which unit you have. The older units with the rev. C circuit board have 6 output jacks on the rear of the unit. The newer units with the rev. G circuit board have only 3 output jacks.

1. Output amplifier changes:
The output circuit is a very simple one. It uses a follower and a unity gain inverter to provide buffered in phase and out of phase outputs from the SSM 2015 amplifier stage. While better balanced output topologies could be used, I have decided to stick with the stock arrangement because of its relative simplicity and good performance into balanced inputs. Although the noise gain of the stage is low, it contributes the bulk of the preamp's distortion.

As stock, the SX202 output stage uses some variant of the 4560 dual bipolar op amp, usually either the Exar XR4560 on older units or the Rohm BA4560 on later units, the Rohm version being especially repugnant. At best, these amplifiers are somewhat slow, and this adds excess distortion above 3KHz or thereabouts. The Rohm devices seem to be marginally unstable in addition to being too slow to pass audio cleanly, and this causes the HD spectrum to be littered with many high order harmonics. The end result is an overly bright and hard sound, solely due to the 4560.

As of this writing, the best possible replacement chip is the new LT1469CN8 from Linear Technology. It is a dual, low noise bipolar input op amp that is remarkable in that it has extremely low distortion, excellent output drive and a very high gain-bandwidth product. It is unity gain stable and it can operate from high supply voltages, so it is a simple drop in replacement. The only drawbacks are cost and availability.

Some varieties of NE5532 will also perform very well, but not quite as well as the LT1469. In my tests, the JRC 5532D from New Japan Radio seems to perform consistently better than most 5532 flavors, and the Fairchild units were fairly close. The Philips 5532 varieties did not do so well distortion wise. The main advantage to the 5532 is price, so I would recommend that chip only if funds are tight and it's hard to find the LT1469. However, it will not produce the best possible peformance.

The LT1364CN8 is also a reasonable amplifier, and if ultrasonic distortion is of prime importance, this might be the best choice. However, due to its high idling dissipation, it should be heatsinked, especially when driving low impedance loads at high levels. Its midband performance is not quite as good as the LT1469, so it is not my first recommendation.
I have recommended some JFET amplifiers from Burr Brown in the past, such as the OPA2132 and OPA2604. While these amplifiers do sound very good, they are not quite as clean at very high frequencies as the 1469 or even the 5532, and I cannot recommend them anymore if cleanliness is your goal. My opinion about these amplifiers changed after finding ways to clean up the SSM 2015 stage, which I will discuss in the next section. After that modification, it became clear that the dominant source of HF distortion was the Burr Brown chip, and the differences between the 1469 and 5532 became more noticeable too.

There are three 4560 chips per SX202 to replace. Install high quality machined pin low profile sockets after you remove the old chips, as soldering the chips directly could damage them. Sockets also make it easier to modify later on if a better chip or hybrid becomes available.
When removing the old chips, clip the pins off near the epoxy chip body and discard the old op amp. The PC board is far more valuable than the old op amp - sacrifice the op amp to help minimize stress on the PC board. Once the bodies of the 4560s are clipped out, remove each pin of the old IC with a soldering iron and needle nose pliers and the excess solder with a desoldering tool, solder sucker, solder wick, braid or what have you. Again, if you don't know what you're doing or you don't have the right equipment, have a technician do these modifications! The PC board is fairly simple to rework since it is only a two sided board, but it can be damaged by carelessness.

i. From Jim Williams: Also, there are better choices for the 4560's, [Instead of the LT1469] I use National LM6172's. I replace the .01 uf psu bypass caps near these parts to .1 uf mono ceramics. That seems to tame some stability issues.

ii. From Justin Morse: Rather than paying lots of money for the hard-to-find LT1469, there are plenty of other chips out there that are much more readily available and affordable (but still very high performance) you could try instead. The single most worthwhile mod is to replace the 4560s with good machined-pin sockets so you can drop whatever you want in there without soldering the IC directly. I would probably start with an OPA2134 if I were doing it today. I would also replace all the electrolytic capacitors with Panasonic FM or Nichicon HE parts of the same voltage rating and package size (which will result in a decent increase in capacitance).

2. SSM 2015 Compensation:

Caveat: …the size of the input coupling caps and the cap that couples the 2015 stage to the output stage. Instead of the 22uF phantom blocking caps I recommended, you could go with higher values like 47uF or even 100uF. Also, the interstage cap, which mostly sets the highpass frequency, could be made larger if you have a suitable cap. I used .22uF with 100 to 150K ohm load, and this could be easily raised to 1uF or even more if you can find a suitable, high quality film cap. Some sort of metallized polypropylene would work well.

The next area for improvement is to change the way the SSM 2015 mike amp stage is compensated. The SSM 2015 is the heart of the mike amp and it provides all but 6dB of the gain of the completed circuit. TheSSM 2015 requires three small valued capacitors to set the internal feedback structure at high frequencies. By changing these values, not only is the stability of the circuit affected, but the high frequency distortion and slewing behavior are also greatly affected.

The rev. C boards have less than optimal values, but the rev. G boards have values that drastically increase HF distortion with no other benefits. The capacitor betwen pin 1 and pin 13 of the 2015 is the most critical capacitor, and the rev. G value of 47pF is simply too large. Changing this to 7.5pF will move the frequency at which distortion sharply rises up an entire octave from 3-4KHz to over 40KHz. The common mode amplifier capacitor should be increased to 56pF. I do not know why this improves HF distortion, but it does.

In both the rev. C circuit and the rev. G circuit, these components are C16, C17, C18 for channel 1 and C7, C8 and C9 for channel 2. On both boards, they are located at both ends of the 14 pin SSM 2015 chip and look like little orange-ish tan discs with two leads.

For rev. C boards, change C9 and C18 from 47pF to 7.5pF, change C8 and C17 from 10pF to 7.5pF and change C7 and C16 from 47pF to 56pF. For rev. G boards, change C12 and C29 to from 47pF to 7.5pF, change C13 and C30 from 47pF to 7.5pF and change C16 and C32 from 47pF to 56pF.
In case I got any component legends wrong, make sure these are the capacitors connected to the SSM 2015 chip after you do the mod: 7.5 pF between pins 1 and 13, 7.5 pF from pin 1 and ground and 56pF between pin 6 and pin 7. Select a quality monolithic ceramic capacitor with .2" radial leadspacings that uses the NP0 / COG dielectric. You could use a polystyrene capacitor instead, but it must be noninductively wound and be connected with short leads to the PC board or it will not function properly. Note that the NP0 dielectric is a very linear dielectric, far different than conventional ceramics used in bypass caps. Since the capacitance is so small, degradation is very unlikely even if the dielectric weren't so clean, so, in my opinion, there is little benefit from using anything but a high quality NP0 ceramic capacitor.

In my units, I made the 7.5pF capacitor up out of two series connected 15pF NP0 surface mount (0805 size) chip caps. It takes a little dexterity and a tiny bit of hookup wire to arrange this, but it can work out well. Alternatively, through hole parts can easily be used here too.
To illustrate the benefit of this mod, a stock rev. G with 5532 output amplifiers but the stock compensation caps will produce almost 1-2% distortion at 100KHz, whereas the same amp with the recommended compensation caps will produce less than .01% distortion at 100KHz. This is measured with the circuit set for 40dB of midband gain, 150 ohm source, 600 ohm load and a -20dBV input.

From Justin Morse: The voltage rating isn't critical. Anything over 50V will be fine. The important thing is that you get the right physical size and shape, and NPO or COG dielectric. Search for Capacitors at Digikey.com and then click on the "ceramic" selection. You'll get a parametric table that lets you narrow the selection based on parameters you care about. I would look for the package style you need, and choose the "COG, NPO" dielectric. You'll find that 7.5pF caps in NPO/COG radial package are not stocked, so you'll have to settle for 7.0 or 8.0pF, or you could put two 15pF caps in series (in which case you might choose axial-lead caps). 7.0pF is probably close enough. I've used caps from BC, Vishay, Murata, Panasonic, Epcos, and others, but I generally buy Kemet or NIC ceramic caps for production, depending on who I'm buying them from.

3. Coupling between the 2015 and output stage:
The original rev. C boards simply connected the 2015 stage's output to the output op amp circuit, and this was found to cause stability problems. Most SX202s will have a 511 ohm resistor retrofitted in series between the 2015 and the output amplifier to tame some stability issues caused presumably by output to input coupling. In the rev G board, this part was designed in and called R78 and R79. If your unit does not have this retrofit, it should, or else it will probably suffer from marginal stability and the resulting increase in distortion. In all units I have seen, this resistor is installed, but there may be a few early rev. C units that do not have the resistor. If that's the case, you will have to cut some traces and install the resistor yourself. I decided to use 432 ohms because it was handy.

As part of the coupling cap modifications, place a small film coupling cap in series with the resistor to block any DC developed in the SSM 2015 gain stage. If DC is removed here, the coupling caps after the buffer stage, which have to be much larger to handle low impedance loads, can be removed. This coupling cap is loaded by the 10K input resistors of the summed ch 1+2 output stage, (R37, R38 on rev G boards, R51, R52 on rev C boards), so it must be scaled accordingly. I decided to use a .22uF film-foil polypropylene cap, so the stock 10K load would not be appropriate. On some units, I have removed the summing stage entirely (by removing R37 and R38 or R51 or R52) and placed a 150K resistor across pins 3 and 5 of the two remaining output amplifier chips. This yields a single order 4.8Hz highpass filter, which is only .24dB down at 20Hz.

On other units, I changed the summing stage to use 100K summing and feedback resistors, and while this increases the noise in the summed output, it's not an issue for my purposes. Note that if this approach is taken, the offset voltage of the summed ch1 + ch2 output stage amplifier determines the output DC sit point of the ch1 and ch2 outputs. A quality, low offset JFET amplifier like the OPA2132P will do nicely there.

In addition to the series resistor and coupling cap, a small capacitor to ground can be added across the input of the output amplifier to further reduce transmission gain at extremely high frequencies. A 120pF capacitor added across pins 3 and 5 of the output amplifier chips U3 and U5 will accomplish this. This capacitor should be a small leaded NP0 ceramic capacitor added to the non component side of the PC board, right at the pins of the IC socket. With this component installed, the SX202 is only a fraction of a dB down at 100KHz, so no useful bandwidth has been lost, only the potential for distortion causing instability and slewing.

4. Power supply bypassing:
This is a somewhat complex topic, but I can summarize my recommendations quite easily. Don't add extra bypass caps or increase the size of any exiting ones, except as I outline below!! The problem with adding bypassing is that you're not really bypassing anything, you're merely coupling transient currents on one supply rail to a part of the ground network. Given that the circuit uses bipolar power supplies and dual supply amplifiers, coupling one rail via a bypass to ground can easily couple rectified audio current into ground, with a variety of possible results, depending on the circuit and the ground node chosen. Given that the ground layout of the SX202 boards is less than ideal (especially on the newer 3 jack revision), this can be a recipe for disaster.

The SX202 relies upon a pair of three terminal voltage regulators to provide low impedance power to the SSM 2015 and output stages, and this scheme actually works pretty well. The PC board is small so the impedance at the amplifier chips is still quite low.

The caps across the outputs of the three terminal regulators should be upgraded from the stock 10uF/50V to a 470uF/25V low impedance electrolytic cap. The stock .01uF ceramics can be upgraded to .1uF multilayer ceramics in parallel with a small 5-10uF surface mount tantalum cap. The Sprague 195D106X0025Y2 SMD tantalum cap fits nicely on the foil side across the pads for the stock .01uF ceramic and seems to be pretty reliable. Use silver solder when installing this cap (and any other SMD cap) to prevent damage to the terminations. Supposedly, metals from the component's end terminations can leach away from the SMD chip if conventional solder is used and weaken or harm it. Why gamble...
This composite network will provide very low impedance across a wide frequency range with few drawbacks. Note that the composite must have less than 1000uF or else the voltage regulators may fail when the circuit is powered down. This is because the capacitance before the regulator is 1000uF and if the capacitance after the regulator is equal to or greater than that, the input voltage to the regulator will go below the output voltage of the regulator when the circuit powers down; this reverse current will destroy the regulator. Protection diodes could be installed to prevent this problem, but I haven't seen the need. Increasing the size of the pre-regulator capacitors to counteract this is also not recommended, as this increases the peak switching currents in the rectifiers and the ground system and therefore increases the noise of the power supply.

The output amplifier chips can have additional local bypasses only if a new grounding scheme is added to the board with stout copper foil. This will be discussed below. However, if you do not do the copper mod, don't add or change any other bypasses. The SSM 2015 basically drives only a high impedance load, so it really won't benefit from such a reservoir of current.
Since the circuit relies on the three terminal regulators for a low impedance supply, it would seem wise to try to upgrade the regulators themselves. The problem is that I know of no simple drop in part that will provide lower output impedance, so the only solution would be to use a more complex circuit on a piggyback board. If anyone does any work regarding this, please let me know! I know Walt Jung has done some excellent work on ultra low impedance regulators for audio circuits, but as far as I'm aware, the designs are too large for this chassis. Perhaps with a good PC layout and SMD parts, they could be adapted to fit the SX202.

5. Grounding:
Caveat: The last paragraph of the Grounding section where I talk about removing R80, the 10 ohm resistor from chassis to ground, should probably be ignored. The whole issue of chassis to ground is not dealt with very well in the 202 anyway, and especially WRT RF sensitivity, something much better should be done. That will happen soon, as I figure it out.
There are a few things you can do to improve the internal grounding of the SX202. A strip of copper foil added to connect each 'pin 1' of the input and output jacks at the rear of the PC board will help to make the 202 behave better, and also ends up creating the ideal ground for output stage supply bypasses, if you choose to add them.

I used rather thick (.25 mm) and wide (1 cm) copper strip stock, cut it into a piece 15 cm long and soldered it over the sleeve connection of each 1/4" jack so that the foil was nearly flush with the edge of the PC board. I then cut a pair of 2 cm strips to reach to each of the XLR pin 1 connections and soldered those to the main strip and to pin 1. This is actually sort of difficult to solder, since the copper strap will expand from the heat more than the PC board. If you get the whole thing to solder reflow temperature, the PC board can warp when it cools as the copper contracts. So, you want to make a connection on one end of the board, let it cool, make a connection at the other end and then work your way into the middle.

Once this strap is in place, you can remove R80, a 10 ohm resistor that connects chassis to signal ground. This component is near the input XLR connectors, and is present only in rev G. units. You can also remove C49 and R65 (on rev C units), two components located right next to the input power jack.

6. Output Bypass Caps:
If you wish to add bypass capacitors to the output buffers, then you need to add another pair of ground straps. These 6 cm. straps are connected between the new 'pin 1' strap at the edge of the PC board and pin 5 of each output amplifier. It's easier to add this strap after you've installed sockets for new output ICs and added the tackdown resistor and 120pF shunt capacitor between pins 3 and 5 of the output amplifier chip. The foil can rest on the leads of these components at the pin 5 node and thus be lifted away from the other traces and cut component leads on the PCB. You may also want to make a pair of bends near where this strap connects to the pin 1 strap to further elevate it away from other traces near the rear of the PC board.

Once the straps are in place, output bypasses can be added. I used a pair of 5uF/35V Kemet axial solid tantalum capacitors for this purpose since they were thin enough to fit between the PC board and the chassis, they have very good electrical performance at high frequencies, they are quite reliable and they shouldn't need to be changed on a regular basis.
A cap goes between pin 4 and the output strap as well as from pin 8 and the output strap. Remember that pin 8 is positive and pin 4 is negative, so install the caps with positive towards pin 8 and positive away from pin 4. Make sure to keep the lead lengths very short. I found it easiest to rest the caps along either side of a ground strap and solder that connection in place to hold the part. Then, the connection to pin 4 or pin 8 is made once the cap is soldered to the strap.

7. Coupling caps:
All transformerless mike amps require input blocking capacitors to keep P48 out of the front end. I had previously replaced the input phantom blocking capacitors with 10uF/100V metallized polypropylene capacitors, but there are problems that can result from this modification. The physical size of the replacement capacitors forced me to mount the caps for channel 2 above above the output stage, and this caused some undesireable ultrasonic coupling and resultant instability.

While film caps do sound cleaner than electrolytics, their size can cause input - output coupling problems, especially once you have extended the bandwidth of the amplifier, so I feel that replacing the stock caps with new, high quality replacement electrolytic caps is a better way to go with the SX202. I chose 22uF 63V Panasonic HFQ radial electrolytics. If you choose a higher capacitance value, leakage will increase, and this can upset the balance of the first stage, so it's probably wise to stick to a cap in the 22-47uF range. Low leakage is a very important parameter to optimize here, as well as any other dielectric losses. In practice, very little signal voltage will develop across these caps, so the nonlinearities of aluminum electrolytics should not be a problem if you use a fresh, quality capacitor. Since the SX202 is an older unit, a few of your input caps are probably starting to leak and need to be replaced anyway.

There is a set of coupling capacitors after the output buffer amplifiers that can be jumpered if the output amplifier and intserstage coupling mods are done. They are 100uF 35V polarized electrolytics located near the output jacks. The output offset of the modified preamp should be very low, less than a millivolt, so unless you have special requirements, these caps can be removed and jumpered.

There are also some 4.7uF nonpolar coupling caps after the polarity switch and pad, and these too can be jumpered if you don't mind hearing ticks or pops when the pad or polarity switches are used. I decided to remove them just so I would not have another electrolytic cap that would need to be replaced after some years.

8. Gain Pot:
The gain pot on the SX202 can be a source of excess noise and possibly microphonics. Ideally, it should be replaced with a switched resistor network to eliminate these problems. I used a 12 position C & K miniature rotary switch and some conventional RN55 style resistors to assemble a switched gain network and I am very pleased with the results. While conducting distortion tests listening to the analyzer residual, I noticed that moving the gain pot caused a lot of scratchy hash, and in some cases, some pot positions caused somewhat high level low frequency noise to be heard. Both of these problems are caused by poor contact from the wiper to the irregular carbon track of the gain pot, and this can sound nasty. A simple, high quality switch and quality metal film gain setting resistors will have none of these problems.

The only drawback to this mod, aside from the complexity of assembling the network, is dealing with a stepped gain control. Some engineers prefer them for interchannel accuracy and resettability, but note that the gain of a switched gain preamp is almost impossible to change gracefully while recording.

If you're interested in a switched gain knob, set up a spreadsheet to calculate the gains given a set of resistors. While it's useful to know the gain of each step, the difference in gain between steps is also useful to see while choosing resistor values. I opted for steps of 2.5 to 2.75dB, and with 12 positions, this gives me around 30db of gain range. The maximum gain of around 50-55dB might be too low for some uses, but you can take whatever sort of tradeoff desired if you design your own gain network.

9. The Input Pad:
I often record using high output mikes placed close to loud sources, so the SX202's stock 15dB pad and 26dB minimum gain was a problem. Reducing the 2015 minimum gain is possible up to a point, but this can reduce the stability margin and honestly, I don't know if you can lower it more than a handful of dB.

A simpler thing is to change the pad from 15dB to 20dB. This can be done in a way that lowers the effective input noise of the preamp, so it's not such a bad tradeoff for most situations. Finally, the SSM 2015 has less distortion if it is presented with a smaller input signal, so using a pad and higher gain is often not such a bad idea for medium - low gain situations.
The SX202 pad is actually a pair of unbalanced pads, one for each input leg. This is actually a good design since it will be more effective than a pi network for mikes that have single ended outputs. Some modern transformerless condensers like the high output Neumann TLM 103 have single ended outputs and if an unbalanced pad is not used, massive, unattenuated common mode signal is presented to the mike amp, and this can cause problems.

The trouble with this design is that the two pairs of resistors used in the unbalanced pad must be matched carefully to preserve the input stage's balance. I used 3K09 and 301 ohm resistors instead of the stock 3K01 and 681 ohm resistors, yielding approximately 21dB of padding when the next stage's 20K load is considered.

These resistors should be closely matched from a supply of quality metal film resistors. Buy 50 or 100 of a particular value and measure them with a quality ohm meter. These days, the HP 3490A multimeter can be found as surplus for $25-100 and it's hard to beat for this work. It has a 5.5 digit display, provision for 4 wire Kelvin measurement leads and is sensitive enough to measure the effects of your body heat and the temperature coefficient of a typical resistor. Sorting resistors is tedious, but it can help to preserve the input stage's balance and it's probably cheaper than specifying ultra precision parts.

10. Peak detector:
The peak detector stage is a transistor follower driven by a diode attached to the output of the SSM 2015 stage. While the input impedance of the follower is large and the output impedance of the SSM 2015 is low, the load on the SSM 2015 is nonlinear and it is remotely possible that distortion could be introduced into the amplifier's output by this arrangement. I personally do not require a peak detector, since I almost always have an analog to digital converter on the output of the SX202 that clips before the SX202 will. If you wish to disable the peak detector, remove diodes D20 and D21. This removes the nonlinear load and also disables the peak detector function. I have no opinion as to whether you or I could hear any benefit from this mod, but it seemed like a prudent thing to do, so the peak detector is gone. If you disable the peak detector in this way, it is very simple to reconnect should you decide you need it again.

11. Things that don't work:
The SSM 2015 chip uses a resistor to program the first stage bias current, and in the SX202, this resistor is set to the maximum bias current that the 2015 can safely handle. Do not reduce the value of this resistor, as you will probably damage the 2015 chip. These chips are very hard to find these days, so you will have to live with the extra distortion from a damaged chip for some time. What's more annoying is that increasing the bias beyond the stock arrangement does not improve performance. I found this out the hard way, so leave the 27K bias resistor alone! There is some evidence to suggest that distortion might be improved by increasing the value of the bias resistor, but this increases noise, and the preamp is already on the noisy side. Leave the bias alone, as it already seems to be optimal!

Good luck with these modifications and enjoy your new preamp!!
Monte McGuire
mcguire@theworld.com

[mecano]

Originally Posted by Jim Williams

You have an older model that does not have the 511 ohm series resistors into the 4560 opamp stages. You can add them by cutting the trace and soldering the resistor across the cut. Also, there are better choices for the 4560's, I use National LM6172's. I replace the .01 uf psu bypass caps near these parts to .1 uf mono ceramics. That seems to tame some stability issues.

Jim Williams
Audio Upgrades

Last year I acquired a sx202 coming from a guy in NL, second channel was weak and got distortion, it is not until recently (and because I had need for the second channel) that I decided to open it and see what can be done. It's a rev C, the resistor mod you are describing was already done! I put dil8 sockets and waiting for OPA2134 from my electronic suppliers I plug 5532's in it ; both channels are now on pair, no more distortion or weak gain on channel two! I said waiting for OPA2134 because LM6172 are available by min. bunch of 5 here (Paris, France) and even it's not so much money I hate stocking electronic parts. So my post has two goals :
- On national website LM6172 are said to be outperformed by LMH6622 ("much better performance" quoting National prose) and LMH6655 ("Lower Noise, Improved Bandwidth, Lower Cost"), anyone got feedback on these?
- If someone in the Paris, France area wants to perform the Operational Amplificators mod with LM* (6172, 6622 or 6655) we can make a group buying.

[edit] almost forgot! Jim, could you please name the .01 uf to replace to tame stability issues? How many are there?

[ulysses]

You might start to hear op amps when they're doing a lot of work (providing a lot of gain) or when there are a whole lot of them in a circuit. Replacing one really good op amp with another really good one in a single unity-gain circuit will be inaudible and, in my opinion, a waste of time and money. There are plenty of other mods left to beef up this preamp, I wouldn't worry about these particular chips.

[mecano]

Thanks Ulysses, words of wisdom are always welcome.
I got the 2134 didn't swap them in yet as I had to service a black and red "made in west germany" behringer parametric eq, which is full of 4560, and now I must restrain myself to change opamps in each gear I inspect.
One more thing I forgot to mention in my first post, a big thank to the people of this forum for the valuable infos and help !

[Macky]

Originally Posted by mecano

Last year I acquired a sx202 coming from a guy in NL, second channel was weak and got distortion, it is not until recently (and because I had need for the second channel) that I decided to open it and see what can be done. It's a rev C, the resistor mod you are describing was already done! I put dil8 sockets and waiting for OPA2134 from my electronic suppliers I plug 5532's in it ; both channels are now on pair, no more distortion or weak gain on channel two! I said waiting for OPA2134 because LM6172 are available by min. bunch of 5 here (Paris, France) and even it's not so much money I hate stocking electronic parts. So my post has two goals :
- On national website LM6172 are said to be outperformed by LMH6622 ("much better performance" quoting National prose) and LMH6655 ("Lower Noise, Improved Bandwidth, Lower Cost"), anyone got feedback on these?
- If someone in the Paris, France area wants to perform the Operational Amplificators mod with LM* (6172, 6622 or 6655) we can make a group buying.

[edit] almost forgot! Jim, could you please name the .01 uf to replace to tame stability issues? How many are there?

Oldish thread i know but i am trying to find out the difference between the 202 and the 202 with the extra jacks at the back. It has 6 instead of 3.

Is this what is called the Rev C ?

Or is there a Rev C in the 3 jack 202 model ?

Does the 6 jack model have better electronics than the 3 jack model ?

Sorry i only know them as Symetrix 202.

I read this somewhere.
The 'Rev C' version is based on now obsolete & hard-to-find SSM 2015 mic amp IC

If this is correct, then i assume that the earlier 202 ..3 jack model does not have the SSM 2015 mic amp IC

Any help is appreciated

Cheers
Macky

[brainditch]

As a former Symetrix employee (1987-1994), I can assert that all revisions of the SX-202 that were produced during my tenure there had the SSM 2015 preamp chip in them.

Early revisions were socketed, but as the sockets they were using were the cheaper "bent pin clamps", rather than machined socket style, there were too many RMA's issued due to the chips doing a "thermal & vibration dance" out of the sockets. They therefore began soldering them (and most of the chips in the Symetrix product lines) directly to the PCB's - possibly as early as 1989.

I'll have to review my records for the answer to the Rev. letter of the motherboard vs. the output jack question, as I was a line tech for the 500/500E/400/600 series products pretty much exclusively.

[mecano]

I can confirm my rev C has 6 outputs.

[Macky]

Originally Posted by brainditch

As a former Symetrix employee (1987-1994), I can assert that all revisions of the SX-202 that were produced during my tenure there had the SSM 2015 preamp chip in them.

Early revisions were socketed, but as the sockets they were using were the cheaper "bent pin clamps", rather than machined socket style, there were too many RMA's issued due to the chips doing a "thermal & vibration dance" out of the sockets. They therefore began soldering them (and most of the chips in the Symetrix product lines) directly to the PCB's - possibly as early as 1989.

I'll have to review my records for the answer to the Rev. letter of the motherboard vs. the output jack question, as I was a line tech for the 500/500E/400/600 series products pretty much exclusively.

Thanks brainditch. Last thing i expected was a reply from an ex-employee.
Great insight and thanks for the confirmation of the SSM 2015 preamp chip.

Originally Posted by mecano

I can confirm my rev C has 6 outputs.

nice one, good to have the facts for any future modding

Thanks Guys
Macky


Edit....The info is in the Monte Mods posted earlier
----
One issue that has to be addressed is that there are essentially two different flavors of the SX202, each using different PC board layouts and component choices. So, the instructions and component designations may vary depending on which unit you have. The older units with the rev. C circuit board have 6 output jacks on the rear of the unit. The newer units with the rev. G circuit board have only 3 output jacks.

[s.d.finley]

Hey guys, I have an issue with my SX202 mod. I am attempting to change out D22 from 44v to 47v and I can find the diode on the schematic but I cannot find it on the sx202 pcb. All the diodes and resistors numbers are underneath the actual part. I don't want to lift every freakin diode just to find this one. I thought I found the correct diode as it was located close to Q5, C49 and C50 and it was all by itself but I was able to move the diode and its D14 instead! Ugh! I already replaced the ICs with burr browns in sockets, replaced all the filter caps and other PS caps, upgraded the coupling caps to the 2015 and lifted the peak indicator diodes. Next I am going to upgrade the pad from -15db to -20db.

But I really want the phantom to be correct too, please help.

Thanks!