Running SSM 2143 from single ended PSU.

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

Rob Flinn

Well-known member
Joined
Jun 3, 2004
Messages
5,240
Location
Between Sussex, UK & Aude, France.
Hi folks

I have a scenario where I am building a 4 channel amp using some class D amps that will be running on a single ended PSU that have unbalanced inputs.  They will be used in a potentially noisy broadcast type environment & the plan is to add some SSM 2143 that we have knocking around  to create balanced inputs.

I have knocked up a small pcb which is using a 5534 to split the power rail & provide a centre tap.  However the centre tap will not be the 0v rail of the single ended PSU.  Assuming I use an output cap on the chip my question is how comfortable will the chip be if the output is referencing to it's -ve power rail instead of the psuedo centre tap that I have created to run the chip ?

AM I overthinking this or have I got my idea wrong ?
 
Rob Flinn said:
Hi folks

I have a scenario where I am building a 4 channel amp using some class D amps that will be running on a single ended PSU that have unbalanced inputs.  They will be used in a potentially noisy broadcast type environment & the plan is to add some SSM 2143 that we have knocking around  to create balanced inputs.

I have knocked up a small pcb which is using a 5534 to split the power rail & provide a centre tap.  However the centre tap will not be the 0v rail of the single ended PSU.  Assuming I use an output cap on the chip my question is how comfortable will the chip be if the output is referencing to it's -ve power rail instead of the psuedo centre tap that I have created to run the chip ?

AM I overthinking this or have I got my idea wrong ?
Your description is not clear to me. Are you using the existing unipolar supply to derive the rails? or is it a separate PSU? A sketch would help understanding...
 
Yes, sorry Abbey.  I was somewhere that I couldn't easily post a diagram.  It is a single unipolar supply to derive the rails.  Please see rather crude attached diagram
 

Attachments

  • 20191114_061240_resized.jpg
    20191114_061240_resized.jpg
    490.1 KB · Views: 56
Rob Flinn said:
Yes, sorry Abbey.  I was somewhere that I couldn't easily post a diagram.  It is a single unipolar supply to derive the rails.  Please see rather crude attached diagram
OK. I think you don't need the 5534. Just connect the reference pin to the junction of the two R's and the C.
You'll need to add DC blocking caps to the inputs.
 
abbey road d enfer said:
You'll need to add DC blocking caps to the inputs.

Hi.  I've been trying to determine what value input cap & what direction they need to be.  There do not seem to be any practical circuits on the net that I've come across.  Any ideas ?  I was think 100u, but not sure of the direction.
 
Rob Flinn said:
Hi.  I've been trying to determine what value input cap & what direction they need to be.  There do not seem to be any practical circuits on the net that I've come across.  Any ideas ?  I was think 100u, but not sure of the direction.
Worst-case input impedance is 10kohm, so with 10uF on each leg, the -3dB point is about 3.2Hz, with about -0.1dB at 20Hz. 100uF would be overkill, but since the cost difference is negligible, There's no objection unless space is a concern. Remember that these caps will need to be rated at 25V min.
 
Thanks for all your help with this.  Having played around with this on the bench we have decided we need a seperate PSU for the ssm2143, because they start clipping at too low a level to be useful at the voltage we are using off the single ended supply.
 
Rob Flinn said:
Thanks for all your help with this.  Having played around with this on the bench we have decided we need a seperate PSU for the ssm2143, because they start clipping at too low a level to be useful at the voltage we are using off the single ended supply.
This is weird. I guess it drives the power amp and that it has significant gain, 20dB like, so the line receiver should not clip before the power amp. Have you done a level diagram?
 
abbey road d enfer said:
This is weird. I guess it drives the power amp and that it has significant gain, 20dB like, so the line receiver should not clip before the power amp. Have you done a level diagram?

We were running it off a split 18v supply & it wouldn't do 1v pk>pk without clipping.
 
> split 18v supply & it wouldn't do 1v pk>pk without clipping.

It should do ten times better.

Something is not right.
 
I used once SSM2143 supplied with a single PSU without problems but there was an OPAMP for DC bias as noted in datasheet.
 

Attachments

  • ssm2143.png
    ssm2143.png
    200.1 KB · Views: 16
moamps said:
I used once SSM2143 supplied with a single PSU without problems but there was an OPAMP for DC bias as noted in datasheet.

That is effectively what I did, although I couldn't trim the V from the op amp it was set midway.

Anyway we went with a  small, cheap +&-15v PSU and it's working fine now.
 

Latest posts

Back
Top