modding a behringer autocom mdx1200

Somewhere in 1995 (the printboard says it's build then  ;D ) i bought a mdx1200 autocom compressor which i used a lot (and actually liked) and later on i bought another one that i sold some time ago. The mdx1200 was just sitting in the rack for years, not even connected and this week i was thinking of going to sell it but found out the leds for bypass and auto didn't work and so i opened it up and was... surprised... in quite a possitive way that is..
No smd, well layed out board and nice poly caps and decent elco's and i even found 2 that 2159 vca's inthere! 
The leds is no problem, they are powered by a bc556 transistor which is blown but now that i've seen the interior i don't want to sell it, i can get maybe 50 euro's but think of all the fun modding it!  ;D

So basically, does anyone have a schematic? i'm curious how things are actually working in there.
It has a jrc4580 opamp in both channels, i think i will replace it with a OPA. There are 2 TL074's per channel, what will be a better option? or maybe it's just like the gssl not shocking because it's in the sidechain.
There is also a CA3046 which is a transistor array http://www.datasheetcatalog.org/datasheet/intersil/fn341.pdf and it also has a BE037 behringer quad opamp (at least, that is what they say) which is brought up before: http://www.groupdiy.com/index.php?topic=1328.0

So... anyone?

I've always been surprised when I open up beh*inger equipment. First being that how they can afford to produce stuff and sell it so cheaply and second how they could manage to use so many Electro CAps! They do have some surprising elaborate  circuits though! I've taken apart and modded an ultrafex and a parametric EQ. Up grading the caps really helps. Unfortunately I can't help you with your current model. All I can do is wish you good luck! Replacing those JRC opamps (if that is what your piece uses, all the beri ger stuff i've opened uses them) really cleans up the sound. Then upgrading the electros of course...
Good luck!

I council you to change all these IC op amps using OP275.

so yours has THAT VCAs ???

mine doesn´t, the labeling has been taken away from the vca´s. the only thing which is readable is "singapore".

I was always wondering if the THAT work as drop-in replacement. I had plans to mod the unit but then there were (and are) so much other projects...

anyway, what i wanted to do was :

- Replacing the carbon resistors in the audiopath with metal film resistors
- Get rid of / swap capacitors
- Overhaul the power supply. Put in a toroid, in my unit the chassis has already a screwhole for that. Bigger filtering caps.
- new opamps, did not pick any specific model yet

P.S.

after thinking about half an hour about your thread I decided to start to try some mods. Maybe we can work it our together without the schematics.

I still have a few MDX 2100 and MDX2000 Composers that I modified many years ago. I exchanged the make-up gain opamp, exchanged ceramic caps for Wima MKS2 and bypassed lots of electrolytics with said Wimas. In hindsight, I don't think it was worth the trouble. I haven't used any of my four Composers for at least 5 years. They were quite useful, when I still used a tape recorder. Not least, because I could gate all eight tracks. But in a DAW setup they are pretty useless. There are lots of plug-ins that are so much better than a Behringer.

Practically speaking, you won't notice much difference with better opamps if your unit already has 4580s. They used 4558 and other lame opamps for a while, and upgrading those does make quite some difference. I know people hate 4580s, but they are quite okay, actually. Fast enough for decent audio. You'll find them in some quite respectable boxes and soundcards.

Behringer often uses pretty small PSU decoupling caps; sometimes none at all. They get away with it because they use low current opamps, but if you upgrade to audiophile opamps, which typically eat quite a few mA, you may run into problems.

everybody has his own point of view....

anyway I started some mods today and did some measurements.

I pulled out the 4580 replaced it with a 5532

As you can see, the 4580 produces 10dB more k3 than the 5532. The difference is 10dB.

This measurements were done with 6dB compression, I nulled the k2 hamonics with the trimmer individually before.

judge yourself

here with 4580

nice! you allready started! Mine is dismanteld on the bench.  ;D
First thing will be the 4580 and the audio caps since i have all of these already.

Maybe off topic or for later consideration, but as for the BEOs, the LT1358/9 might be possible replacements (same pin layout, better specs, but not cheap). I remember some very respectable audio upgrader mentioning them particularly as replacements for the BEOs. This might seem strange, because in S. Groner's opamp distortion study these LTs do not look exciting at all. Anyway, I ended up using the LT1359s not in a Behringer (where I had intented to use them) but in a different piece of gear for testing in the summing stage instead of the stock TL074s. I didn't like Burr Browns there for their signature sound and the LTs are very fast after all... and that's what we want in a summing stage(!). Also, at least on paper, the LT1359 looks better than the TL074 in terms of CMRR, bandwidth and slew rate. It might need more power though, IIRC, so that needs checking.

The quad BEO should be a JFET as is the TL074. Whereas the MC3307x is BJT... When swapping for the MC, I remember someone noticed, depending on the circuit, you might end up with scratchy pots. I can't really confirm that, but I do have a cheapo piece of gear stock-equipped with MCs and pots are scratchy -- however, this might be for a different reason.

Anyway, keep us posted about the mods you do. I have some unused old Behringer gear on the shelves too, waiting to be either sold, butchered or modded just for the fun of it And who knows maybe it becomes usable...

ADDED: And yes, the OPAx227 sound very nice

I think that schemas were in gmail..

AK79 said:

everybody has his own point of view....

anyway I started some mods today and did some measurements.

I pulled out the 4580 replaced it with a 5534

As you can see, the 4580 produces 10dB more k3 than the 5534. The difference is 10dB.

This measurements were done with 6dB compression, I nulled the k2 hamonics with the trimmer individually before.

judge yourself

Click to expand...

Just wondering, shouldn't the 5534 not actually be a 5532? 4580 is dual, 5534 is single and 5532 is dual.

kooma said:

I think that schemas were in gmail..

Click to expand...

Great!! Thanks!  :-*

[/quote]
Just wondering, shouldn't the 5534 not actually be a 5532? 4580 is dual, 5534 is single and 5532 is dual.
[/quote]

you´re right. actually i always mix them up when talking or writing about them  ;D

some more specs

EIN 96 dB correction : not EIN but dynamic range sorry. Better measuring equipment shows that it is 93 dB actually.
THD + N : 0,0322 % correction: found out that 0,03 is actually the THD of the oscillator used. Better osc reveals 0.01 here

these did not change with the 5532 and anyway aren´t too bad i think.

Another point I found out is that the 4580 has oscillation problems, there was something around 30kHz which disapeared with the 5532

I had some spare time this morning and so i heated up the iron.

I replaced all elco's with panasonic caps, i think that was a good idea not just because of sound but also because of the fact that they where about 15 years old. The 0,47uf elco in the sidechain path is 1uf now, not sure if that will make much difference but it looks like as if it only uses that for external sidechain.
The audio caps (22uf) i paralleled with a 100nf cap, some people say that is a good thing to do, other say it's not.. not sure yet myself.
The elco's after the 78/7915 are now 22uf (standard 10uf) and i replaced the 100nf's for wima's.
All other 100nf's where also replaced with wima's and i added bypass caps on + and - on the 4580's which i replaced with a 5532 and the other channel with a opa2134 (i had 1 of both left). To me the channel with the opa2134 sounds just a bit more open/natural though i'm not 100% sure since i prefer the opa's so it could be my imagination.

The thing i always noticed with the autocom is that it 'eats' some low-end, it still does but it's just a bit better now to my feeling but i want to get rid of that completely. So any ideas where that could be coming from? I was thinking of maybe getting rid of some caps in the audio path.

Hi dagoose.

I´m not a fan of this capacitor bypassing thing but if it sounds right to you it is right...

I wil try the opa2134 also as soon as I get one & bypassing the IC´s supply rails is a good point.

Some more  thoughts, don´t know if it all will work out :

C24 / R26 :

since this highpass is made up with a electrolytic I would suggest to put in a higher value to get rid of distortion. Try 220uf. I think this will help with the loss of bass you described. This  lowers the frequency of the filter but I think with a 5532 or better this shouldn´t be any problem ? Maybe it is better to check DC-levels after the outputamp before.

IC 5 :
I have no idea if this chip is good or not. Seems like a TL074 replacement, so a LT1359 might be  better.

Enhancer:
This is not a real enhancer is it ? Looks more like a high shelf which can be mixed in. Two options: Get rid of it (remove R55 & R56) or check
if altering the frequency (which is 10k it think) sounds better.

Resistors at the input stage:
Maybe these 20k (R11-14) resistors can be lowered to 10k to get rid of some noise. Also, as I wrote before, substituting all carbon resistors in the audio path with metalfilm will be worth the effort i think. Can´t figure out which ones right now, because I don´t have the unit in front of me right now.

Power supply:
I got some other psus lying around. May be increasing the supply voltage from +/-15V to +/-17V will be an improvement also. Don´t know if all the chips will be happy with this... better to check that first.

Maybe some of the above is nonesense - these are just some sunday-afternoon thoughts while looking at the schematics  ;D

will try these things anyway during the next week.

cheers
Arne

did the following today:

swapped R11-R14 with 10k metal film
C24 is now 220uF

THD + N is now 0.005 %

dynamic range now 93.5 dB (could be measurment error though)

next thing to come is a substitute for IC 5....

C3 wants to be changed too, I guess. Or even better: make it a real hard bypass.

I was thinking, don't you think it's also a good idea to change c4 to 220uf or 100uf as well?
I'm also thinking of making the output balanced, to me it looks quite simple to make some mods to the output opamp to get things balanced. The opamp can handle it, the only thing is that the output pot needs to be moved to the input at pin6.
On second thought it might not be that easy and needs a lot of extra wiring. Maybe something like a THAT1646 driver might be more simple.

I had to order some stuff and so i also ordered mc33079's to replace the tl074, i'm also not sure what to do with the BE037, i'm afraid of destroying it when i desolder it so i need to crush my brains on that first. Might be a good thing to replace with mc33079 as well if it's the same as a tl074 which i think it is. But then.. why did they use a be037 in the audio path and a tl074 in the sidechain path if they are the same?  ???