The No.1 Website for Pro Audio
Mixer mod
Old 23rd February 2019
  #1
Gear Maniac
Mixer mod

Hi

I have a Soundtracs 32/4/2 mixer, not used at the moment but still a good unit. Just feel it's a bit lifeless... I'm going to mod it so that first 24 channels will get OEP A262A3C input transformers for the mic XLR, after phantom dc block, before the ssm2017 preamps. Noise floor of ssm2017 chips is quite impressive and after the tranny mod get 70dB gain.

I will be also adding -20 pad, phase reverse and sellectable load resistor for the tranny secondary (freq.resp.adj. The last 8 channels will get Lundahl 1:1 line transformers and will act as summing channels from DAW for stems. I have most of the parts, just adfitional 500€ investment and the worktime.

I was just wondering if anyone has done extensive mods to their mixers and was it worth it. I have been working with electronics as a profession now for 22 years. The actual work is not a big deal. I will design additional PCBs and get those printed, 1 per each 8 channel pcb, then it's just soldering connecting cables basically.

Last edited by Transmutant; 24th February 2019 at 09:14 AM.. Reason: Typo
Old 24th February 2019
  #2
Lives for gear
 
ruffrecords's Avatar
I think there is a possibility this will not work. The A262A3C transformer has a 1:6.45 step up ratio. so a 150 ohm source will look like about 5.8K to the ssm2017. Unfortunately, the ssm2017 is most likely arranged to provide a 1.5K or 2K input impedance those this is going to heavily load the transformer secondary. This means you will not get all the extra gain you were hoping for and more importantly you will lose a lot of bass response.

I just looked at the Soundtracs Topaz specs which says the mic input impedance is 1.2K.

Cheers

Ian
Old 24th February 2019
  #3
Gear Maniac
Thanks for the thought. I'm aware of that 1k2 load. I am removing those and changing the ssm2017 input to match transformer secondary, 6K load. Also adding another load resistor behind a flip switch to tweak the frequency response little bit.

Actually the 1K2 resistors are part of the line in voltage divider to lower the input volume.

I'm also removing the Line in, insert and direct out sockets from the pcb and routing those to the back pabel. I will use the exposed holes to mount phase, pad & load resistor switches. Lot of work, but I might do it 8 channels at a time.

Last edited by Transmutant; 24th February 2019 at 10:48 AM.. Reason: Typo
Old 24th February 2019
  #4
The SSM2017, 2019, INA217 and that 1510 all require a low source impedance or their noise specs will degrade. A special mic input transformer must be used, Jensen makes it.
Old 24th February 2019
  #5
Gear Maniac
Quote:
Originally Posted by Jim Williams View Post
The SSM2017, 2019, INA217 and that 1510 all require a low source impedance or their noise specs will degrade. A special mic input transformer must be used, Jensen makes it.
Maybe I'm reading different datasheets. For SSM2017 input impedance is marked 1Mohm, INA217 datasheet says 2Mohm.
Old 24th February 2019
  #6
Gear Maniac
Quote:
Originally Posted by ruffrecords View Post
I think there is a possibility this will not work. The A262A3C transformer has a 1:6.45 step up ratio. so a 150 ohm source will look like about 5.8K to the ssm2017. Unfortunately, the ssm2017 is most likely arranged to provide a 1.5K or 2K input impedance those this is going to heavily load the transformer secondary. This means you will not get all the extra gain you were hoping for and more importantly you will lose a lot of bass response.

I just looked at the Soundtracs Topaz specs which says the mic input impedance is 1.2K.

Cheers

Ian
About impedance, it's actually winding relation squared. The ratio is 1:6.4... so roughly 1:7 lets say. So 200ohm load on primary looks like 200*7= 1400/4 =350ohm.
Old 24th February 2019
  #7
Gear Maniac
Anyways, I believe I'm on top of the analog of things. I was just curious if anyone has added transformers to their mixers.

Btw for the transformer would not see the 1k2 resistors as loads, they are connected to the GND, so acting as DC biasing for the SSM2017 inputs and also part of the voltage dividing 33k+10ohm+1k2 for the line input socked. Any loading for the transformer secondary should be between the coil wires, not from both to the GND.

Last edited by Transmutant; 24th February 2019 at 11:07 PM.. Reason: Typo
Old 24th February 2019
  #8
Lives for gear
 
ruffrecords's Avatar
Quote:
Originally Posted by Transmutant View Post
About impedance, it's actually winding relation squared. The ratio is 1:6.4... so roughly 1:7 lets say. So 200ohm load on primary looks like 200*7= 1400/4 =350ohm.
Absolutely right but you forgot to square it. So it should be

200*7*7 = 9800

Cheers

Ian
Old 24th February 2019
  #9
Lives for gear
 
ruffrecords's Avatar
Quote:
Originally Posted by Transmutant View Post
Maybe I'm reading different datasheets. For SSM2017 input impedance is marked 1Mohm, INA217 datasheet says 2Mohm.
There is a very significant difference between the actual input impedance of the op amp and its optimum source impedance for minimum noise. See page 6 of the Rev B data sheet for more details.

Cheers

Ian
Old 24th February 2019
  #10
Lives for gear
 
ruffrecords's Avatar
Quote:
Originally Posted by Transmutant View Post
Anyways, I believe I'm on top of the analog of things. I was just curious if anyone has added transformers to their mixers.

Btw for the transformer would not see the 1k2 resistors as loads, they are connected to the GND, so acting as DC biasing for the SSM2017 inputs and also part of the voltage dividing 3k3+10ohm+1k2 for the line input socked. Any loading for the transformer secondary should be between the coil wires, not from both to the GND.
So there is 1K2 + 1K2 between the hot and cold. That is what the transformer will see across its secondary. The fact that the centre point of the two resistors is connected to ground does not alter this.

Cheers

Ian
Old 24th February 2019
  #11
Lives for gear
 
Brian M. Boykin's Avatar
To your original question, I have a Soundcraft 400b. Jim did the master module and I’ve been recapping and rechipping the input modules and groups. It’s circa 1980ish. Recapping the modules made the most noticeable difference sonically. New opamps was more subtle when compared 1:1 with unmodded channels. What I noticed with the opamps is the mix stays cleaner as more channels are brought into the mix. Less EQ is needed to keep the mud out. I aim for 24 channel mixes but can go up to 32. I recently built a new power supply with Power One modules and that blew my mind. I can drive the meters into the red and it doesn’t fall on its face. I’m finishing up some modules today. The mic pre is getting individual dale metal film resistors and the the RIC is coming out. So yes, it’s worth it. Every bit helps. Nothing that has ever been excellent became excellent by being good enough.

Brian
Old 24th February 2019
  #12
Gear Maniac
Quote:
Originally Posted by ruffrecords View Post
So there is 1K2 + 1K2 between the hot and cold. That is what the transformer will see across its secondary. The fact that the centre point of the two resistors is connected to ground does not alter this.

Cheers

Ian
So if I change those resistors to 3k + 3k (nominal load for secondary) the downside is worse noise performance of the opamp? What is the load opamp would see? 6k + 9k parallel = 7k5? Without transformer just the 1k2+1k2 & source impedance? Around 1.5k?
Old 24th February 2019
  #13
Gear Maniac
Just another thought, what if I use model OEP A187A11C which is 1:1:1 mic splitting transformer. Each secondary wants to see 1k load. Frequency responce of that transformer is not "as good", just slight drop below 100Hz (-0.2db) but would be a better match for the SSM2017. Maybe that low freq region could be boosted with cap adjustment before primary.

Last edited by Transmutant; 24th February 2019 at 11:42 PM.. Reason: Typo
Old 24th February 2019
  #14
Gear Maniac
 
analogguru's Avatar
 

Quote:
Originally Posted by Transmutant View Post
Anyways, I believe I'm on top of the analog of things. I was just curious if anyone has added transformers to their mixers.
.....
Quote:
Originally Posted by Transmutant View Post
....
6k + 9k parallel = 7k5 .....
I am confused....., what are you talking about ? That you do not want any advice from audio experts instead you need help from another profession ?
.
.
Old 25th February 2019
  #15
Lives for gear
 
ruffrecords's Avatar
Quote:
Originally Posted by Transmutant View Post
So if I change those resistors to 3k + 3k (nominal load for secondary) the downside is worse noise performance of the opamp? What is the load opamp would see? 6k + 9k parallel = 7k5? Without transformer just the 1k2+1k2 & source impedance? Around 1.5k?
In both cases you need to work out the source impedance seen by the the op amp. This is not set solely by the bias resistors.

It is normal to set the impedance seen by the microphone to be about 1200 ohms (as in the original design). Since the transformer turns ratio is 6.25:1 this means the load on the transformer secondary must be 1200*6.25*6.25 = 47K in order for this to look like 1200 ohms to the mic.

In parallel with this is the (mic resistance plus the transformer primary resistance) multiplied by the turns ratio squared plus the transformer secondary resistance. Work out what this is inn parallel with 47K and then feed it into the formula on page 6 of the data sheet to get the actual noise.

Cheers

Ian
Old 25th February 2019
  #16
Lives for gear
 
ruffrecords's Avatar
Quote:
Originally Posted by Transmutant View Post
Just another thought, what if I use model OEP A187A11C which is 1:1:1 mic splitting transformer. Each secondary wants to see 1k load. Frequency responce of that transformer is not "as good", just slight drop below 100Hz (-0.2db) but would be a better match for the SSM2017. Maybe that low freq region could be boosted with cap adjustment before primary.
A much smaller turns ratio input transformer would most likely be a better match for the op amp from the noise point of view.

Cheers

Ian
Old 25th February 2019
  #17
Gear Maniac
Quote:
Originally Posted by analogguru View Post
I am confused....., what are you talking about ? That you do not want any advice from audio experts instead you need help from another profession ?
.
.
Insults are the best. That is what these forums are for... Yeah, maybe I'm a bit cuckoo. After all we can't all be gurus.
Old 25th February 2019
  #18
Gear Maniac
Quote:
Originally Posted by ruffrecords View Post
In both cases you need to work out the source impedance seen by the the op amp. This is not set solely by the bias resistors.

It is normal to set the impedance seen by the microphone to be about 1200 ohms (as in the original design). Since the transformer turns ratio is 6.25:1 this means the load on the transformer secondary must be 1200*6.25*6.25 = 47K in order for this to look like 1200 ohms to the mic.

In parallel with this is the (mic resistance plus the transformer primary resistance) multiplied by the turns ratio squared plus the transformer secondary resistance. Work out what this is inn parallel with 47K and then feed it into the formula on page 6 of the data sheet to get the actual noise.

Cheers

Ian
The page 6 formula, I had missed that. Also all of the parts working together in analog world is way more complicated than the digital electronics I am maybe more used to. I'm willing to read and learn... And not afraid to ask stupid things. If this is not tolerated I think it would be a shame.

I'm most grateful for all the help.
Old 25th February 2019
  #19
Gear Maniac
Also maybe another stupid question. The mic inputs in my Soundtracs desk are HOT. I have been using -10 or -20db H-pads when needed. Now that I would like to add a permanent -20 pad, would it be better to place it on the primary side and how would that reflect to secondary? Again, if this something you all are familiar with, sorry in advance.
Old 25th February 2019
  #20
Gear Maniac
I gather U-pad, 150ohm across hot and cold on the primary side? At least that I would have done. Maybe there are things I have missed.
Old 25th February 2019
  #21
Lives for gear
 
Radardoug's Avatar
 

I'm confused. You were going to put a stepup transformer on these inputs which would have given you more gain, and now you are telling us you often have too much gain. Huh?

The exercise you have embarked on will be quite expensive, and you will end up with a modified mixer of doubtfull value. Plus you have no idea whether what you propose will return the results you seek. What about just selling this mixer, and using the money to buy what you want, if necessary by obtaining a trial on a different mixer. And it would be much quicker!

And what you should bear in mind is:
If your source is good, your recording is good. If your source is crap, your recording is crap. The sound of the source is THE MOST important part of this equation. Gear is secondary.
Old 25th February 2019
  #22
Gear Maniac
Quote:
Originally Posted by Radardoug View Post
I'm confused. You were going to put a stepup transformer on these inputs which would have given you more gain, and now you are telling us you often have too much gain. Huh?

The exercise you have embarked on will be quite expensive, and you will end up with a modified mixer of doubtfull value. Plus you have no idea whether what you propose will return the results you seek. What about just selling this mixer, and using the money to buy what you want, if necessary by obtaining a trial on a different mixer. And it would be much quicker!

And what you should bear in mind is:
If your source is good, your recording is good. If your source is crap, your recording is crap. The sound of the source is THE MOST important part of this equation. Gear is secondary.
Hmm... I believe I said mixer is ok as is, only a bit lifeless. Transformer are non linear so they add 'colour' (whatever that is to anyone) aka distort low frequency and add phase shift across the whole frequency band. If I wanted transparent preamps I would most likely not do anything to it. But it is good platform, rather that bying 300€/channel DIY kits with transformers. Most of the parts are already there... And in a great user interface. I have let it be 'as is' now for 5 years and bought some other gear to record with. I believe my quest is not accidental, of course I will mod only one or two channel first and evaluate the results with tone generator etc.

Preamps being hot is not the same as maximum gain. With the desk preamps all the way down they are still clipping with some combinations, usually closed micked drums. And now rather than slashing out with something again, figure out why most pro gear come with pads in all channels. So they can be turned on when needed.

The initial thought was to increase that gain window for quiet mics (sm7 or ribbons) with the step up transformers and use the pad for the hot ones but I didn't take account to the opamp noise specs. With the "unity gain" 1:1 ratio transformer this should be less of an effect. So I can just change the gain pot series resistor from 10ohm to 3ohm, going from 60db max gain to 68db max gain If I need it and I think I do for ribbon mics. SSM2017 has 70db gain for 0ohm resistance. Of course this will also boost the now slightly elevated noice floor, that is why I though getting that +10db boost before the opamp would have been better, but in life you rarely get all you want.

The re-sale price for the desk is 250€, at least where I live. I've had it for sale multiple times without luck. No-one wants a huge analog desk in their home studio it seems. Theres plenty for sale for little money.

I'm also a bit confused... Why are you on a forum where people modify existing gear if the gear should be used as they are. Like noted on this thread every change for the better helps, I's just a matter of taste. I think spending the money to serve your vision is always a good thing, the 500-700€ is nothing in my mind. I can't even get another 1073 style stereo channel module with that.

Last edited by Transmutant; 25th February 2019 at 07:58 AM.. Reason: Typo
Old 25th February 2019
  #23
Lives for gear
 
ruffrecords's Avatar
Quote:
Originally Posted by Transmutant View Post
The page 6 formula, I had missed that. Also all of the parts working together in analog world is way more complicated than the digital electronics I am maybe more used to. I'm willing to read and learn... And not afraid to ask stupid things. If this is not tolerated I think it would be a shame.

I'm most grateful for all the help.
No problem; we are here to help.

In another post you mention the mixer inputs are hot and you suggest adding a pad at the input. Looking at the Topaz schematics, the ssm2017 is configured so its gain can be turned right down to 10dB. So signal levels as high as -10dBu should be handled with ease and there is enough headroom to accommodate peaks much higher then this. If your (mic) input signals are hotter than this then you will need a pad. Which particular signal sources are you finding too hot?

Cheers

Ian
Old 25th February 2019
  #24
Gear Maniac
Quote:
Originally Posted by ruffrecords View Post
No problem; we are here to help.

In another post you mention the mixer inputs are hot and you suggest adding a pad at the input. Looking at the Topaz schematics, the ssm2017 is configured so its gain can be turned right down to 10dB. So signal levels as high as -10dBu should be handled with ease and there is enough headroom to accommodate peaks much higher then this. If your (mic) input signals are hotter than this then you will need a pad. Which particular signal sources are you finding too hot?

Cheers

Ian
Thanks! Usually modern dynamics such as Beta series (neodymium magnets) close mics on drums, snare, kick or toms. After retiring my Soundtracs I have pulled out the transformers from my Beta56 mics which helped the frequence response pretty much the same way the legendary SM57/58 MOD. Because of the neodyniums the output is still ok without the step-up.

Last edited by Transmutant; 25th February 2019 at 11:53 AM.. Reason: Typos
Old 25th February 2019
  #25
Gear Maniac
Maybe the treshold for the 072 for the clip indicator led is too low, have not checked if the preamp is actually clipping.
Old 25th February 2019
  #26
Lives for gear
 
ruffrecords's Avatar
Quote:
Originally Posted by Transmutant View Post
Thanks! Usually modern dynamics such as Beta series (neodymium magnets) close miced drums, snare, kick or toms. After retiring my Soundtracs I have pulled out the transformers from my Beta56 mics which helped the frequence response pretty much the same way the legendary SM57/58 MOD. Because of the neodymiums the output is still ok witout the steup.
Close miked drums can output extraordinarily high levels so perhaps for those cases a 20dB pad is necessary. A U pad would be the usual way to go. You can make on from a pair of 680 ohm resistors and a 150 ohm. THis will maintain the input impedance at around 1.5K and provide a 150 ohm source to the mic pre.

Cheers

Ian
Old 25th February 2019
  #27
Gear Maniac
Quote:
Originally Posted by ruffrecords View Post
Close miked drums can output extraordinarily high levels so perhaps for those cases a 20dB pad is necessary. A U pad would be the usual way to go. You can make on from a pair of 680 ohm resistors and a 150 ohm. THis will maintain the input impedance at around 1.5K and provide a 150 ohm source to the mic pre.

Cheers

Ian
Thanks Ian for your input. I understad a little more about the interaction. After yesterday I wanted to read more about these low level transformers and accidentally stubled upon this great page:

Transformers for Audio

There's a good part of Info about the cap + primary coil resonant circuit, I'm wondering if there will be sa kind of phenomena with the 47uf DC blocking caps acting as bipolar 23.5uf cap resulting in similar 5-10db boost as described. Should I change the to 470uf so the combination is 235uf or is it even an issue with there lower mH coils?

What think I I will be doing is I will buy 2 of those 1:1 transformers and experience with the combination.
Old 25th February 2019
  #28
Gear Maniac
Quote:
Originally Posted by Transmutant View Post
Thanks Ian for your input. I understad a little more about the interaction. After yesterday I wanted to read more about these low level transformers and accidentally stubled upon this great page:

Transformers for Audio

There's a good part of Info about the cap + primary coil resonant circuit, I'm wondering if there will be sa kind of phenomena with the 47uf DC blocking caps acting as bipolar 23.5uf cap resulting in similar 5-10db boost as described. Should I change the to 470uf so the combination is 235uf or is it even an issue with there lower mH coils?

What think I I will be doing is I will buy 2 of those 1:1 transformers and experience with the combination.
As it happens to be the 1:1 transformer each coil inductance is reporter to be 2.82H, with the 47uf caps back to back makes 23.5uf + 2.82H = 19.551Hz. 10x the caps freq goes down to 6-7Hz.
Old 25th February 2019
  #29
Gear Maniac
Ok, I ordered 2x A187A11C transformers. They were 87€ including shipping. Not the cheapest but the price drops 5€/per transformer when buying more than 10pcs + free shipping over 100€ pucrhases. I did a quick summary (schematic) based on Ians great help. This what I'm starting with and will see how it goes. Maybe it's a complete waste of time. Maybe it's not.
Attached Thumbnails
Mixer mod-soundtracs_changes.jpg  
Old 25th February 2019
  #30
Gear Maniac
Ian should I use just the other secondary and change the 1k2 resistors to 600ohm? Using both would make it 1:2 transformer and again make the impedance matching difficult. Right? It would not be way off but a bit of out of place. Also should the unused secondary be connected to a load or just having the leads not connected. Just thinking is there a possibility that something is picked up by it that effects the transmission. Might be silly questions again...
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
Forum Jump
Forum Jump