HELP! Differential Input Circuit - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 23rd September 2004, 05:18 AM   #1
diyAudio Member
 
Join Date: Sep 2004
Location: Surrey, BC.
Send a message via MSN to mtnickel
Default HELP! Differential Input Circuit

Hi there. I'm fairly new to the DIY scene especially with the electronics side.

My problem here is car audio related, but i figured the viewers of this thread would know the most.

I'm getting engine noise in my car due to poor ground isolation in my amplifier. I'd like to add differential inputs to the amp, or at least just outside of it.

What is happening is I have a potential from the rear of the car to the front and current flows through the RCA shield. This dirty, ripple filled current makes noise in the signal and is amplified as whine.
Transformers filter this fairly good, but result in very bad frequency response.

My problem is most articles i've found are for balanced inputs. Understanding the CMRR in those is fairly straight forward and all the noise rejection that comes with it.

I however don't know how to make one for an unbalanced source.
The output signal ground of the differencial circuit must have a very high impedance to stop any current flow.
Any help, links, direction on this subject would be sooo very much appreciated. Thanks to all who contribute.

Sincerely,
new Diy'er Mark Nickel
  Reply With Quote
Old 23rd September 2004, 06:47 AM   #2
diyAudio Member
 
Join Date: Sep 2004
Location: Surrey, BC.
Send a message via MSN to mtnickel
Default Possible Solution

Okay...after looking around the posts here for a while, i've come up with a possible solution...

A Burr DRV134 unbalanced to balanced line driver, straight into a INA134 which is a balanced to unbalanced converter. The INA134 states it has ground loop isolation. So i think this sort of thing may work.

I'm like a true beginner with this, so bare with me.

The Vcc and V-, can this be my 12-15 VDC source in the car? Or do i need actual -5 v and +5 v minimum. I'm worried about the cleanliness of my DC source too. perhaps a cheap DC-DC converter with 12v+/- outputs would be in order.

If anyone could give a beginner a hand here, it'd be greatly appreciated.

Also, what would be a good online dealer to supply small quantities of these IC's.

Thanks alot guys.

Mark
  Reply With Quote
Old 23rd September 2004, 06:57 AM   #3
Eva is offline Eva  Spain
diyAudio Member
 
Eva's Avatar
 
Join Date: Oct 2003
Location: Near the sea
Send a message via MSN to Eva
The common approach to prevent groung loops in these circumstances is to use a floating SMPS on the amplifier [or at least partially floating], this forces the speaker/signal ground and the chasis ground to be no longer coupled

Currently almost all amplifiers use this cheap approach

In the other hand, almost all amplifiers not having a SMPS are powered directly from +12V and thus rated at 15Wrms@4 ohms per channel so they are useless nowadays. Special class H ICs [50W@4 ohms] designed to be powered from +12V like TDA1562Q also include differential inputs internally
  Reply With Quote
Old 23rd September 2004, 07:05 AM   #4
Konrad is offline Konrad  Norway
diyAudio Member
 
Join Date: Jun 2003
Location: way up north
Quote:
I'm getting engine noise in my car due to poor ground isolation in my amplifier. I'd like to add differential inputs to the amp, or at least just outside of it.
Naaaa I belive it's lack of gnd. Short wire too gnd/chassi is the common way, why use a lot of cable when not nessy? A long wire or two to pick up more noise, like an antenna?
  Reply With Quote
Old 23rd September 2004, 07:20 AM   #5
Banned
 
Join Date: Feb 2002
Location: As far from the NOSsers as possible
Default Don't knock balanced.......

The cable isn't going to be any longer if it is balanced, and even if it was.......it would sound better. Period.

OK.....another way........if you have a source of +/- 12 volts.....a lot of this car audio stuuf does, with all those SMPS inside.......

Analog Devices makes some parts called.......SSM-2142/3, etc. for this. Don't sound too bad, a little rolled off in the top, but I don't think that will be a hinderance in this application.

Yeah....you could buy some cheap DC-DC converters......cheaper still if you can find them surplus..........

I know.....another overkill solution form an engineer. Hey, that is what we do best. Sometimes.

Jocko
  Reply With Quote
Old 23rd September 2004, 08:09 AM   #6
diyAudio Member
 
Join Date: Sep 2004
Location: Surrey, BC.
Send a message via MSN to mtnickel
Default Re: Don't knock balanced.......

Quote:
Originally posted by Eva
The common approach to prevent groung loops in these circumstances is to use a floating SMPS on the amplifier [or at least partially floating], this forces the speaker/signal ground and the chasis ground to be no longer coupled

Currently almost all amplifiers use this cheap approach

In the other hand, almost all amplifiers not having a SMPS are powered directly from +12V and thus rated at 15Wrms@4 ohms per channel so they are useless nowadays. Special class H ICs [50W@4 ohms] designed to be powered from +12V like TDA1562Q also include differential inputs internally
I'm sure the amp has SMPS. I'm quite a beginner here, but what I can tell you is there is about 10Kohms between the RCA ground and chassis ground. This is not big enough and DC current flows...

Quote:
Originally posted by Konrad


Naaaa I belive it's lack of gnd. Short wire too gnd/chassi is the common way, why use a lot of cable when not nessy? A long wire or two to pick up more noise, like an antenna?
The amp ground is 4 gauge and very short. To assure best connection, i welded a bolt right to the frame and secured the ground ring with a nut. There are no long wires to pick up EMF. The noise is not induced. It comes from the very bad ripple in the DC supply. The RCA sees this ripple because of the slight potential difference in the car. If i power the head unit with an isolated power source, the noise is completely eliminated. A DC-DC 12v-12v isoloted converter is not cheap and not very common for the size of one i'd need to run the Head unit.


Quote:
Originally posted by Jocko Homo
The cable isn't going to be any longer if it is balanced, and even if it was.......it would sound better. Period.

OK.....another way........if you have a source of +/- 12 volts.....a lot of this car audio stuuf does, with all those SMPS inside.......

Analog Devices makes some parts called.......SSM-2142/3, etc. for this. Don't sound too bad, a little rolled off in the top, but I don't think that will be a hinderance in this application.

Yeah....you could buy some cheap DC-DC converters......cheaper still if you can find them surplus..........

I know.....another overkill solution form an engineer. Hey, that is what we do best. Sometimes.

Jocko
I'm not worried about powering long cable runs. I've got 4V rms output into a 600ohm load from my alpine deck.
How is your solution different than mine aside from the IC's? Isn't that the same as what i thought i could do. You're right about the +/- 12v source though...i'm sure i could open up a component and find a good regulated supply. If i can, i will attempt to put this entire circuit inside my amp. That'd be pretty cool i think.


The only requirement I have for solution is...
Take an unbalanced source, and remove it's ground. harder than it sounds beacuse the ground carries the signal. This would happen if i used the Burr chips, wouldn't it?

Thanks for the quick responses guys.
Perhaps with my added facts, you guys will know more of what i'm talking about (not likely, cuz i don't even know what i'm talking about! :P )


Mark
  Reply With Quote
Old 23rd September 2004, 09:24 AM   #7
Eva is offline Eva  Spain
diyAudio Member
 
Eva's Avatar
 
Join Date: Oct 2003
Location: Near the sea
Send a message via MSN to Eva
10K isolation between RCA/speaker ground and +12V ground should be more than enough, even values as low as 220 ohms work quite right

Now I'm almost sure that your noise pickup problem does not come from a ground loop

I would check that the RCA ground is the same as the speaker ground of the amplifier to discard some quite common problems like fused ground tracks on the amplifier [RCA connectors touching 12V by accident]

I would also check that the RCA ground of the head unit is the same as the chasis ground to discard fused tracks inside the head unit [again RCA connectors touching 12V by accident]

Then, I would look for magnetically induced noise. This happens usually with 4 channel amplifiers when there are two stereo RCA wires between the head unit and the amplifier and each follows a different path.

I suggest checking if you get noise with none, 1, 2, 3, etc... RCAs connected

In case of fused tracks on the amplifier you may get noise even with no RCAs connected

In case of ground loop or fused tracks on the head unit, you should get no noise with no RCAs and noise with one or more RCAs

In case of magnetic couplig you should get no noise with one or no RCA connected and noise with two or more

Dont panic, I've faced that problem many times and there is allways a simple clever way to solve it
  Reply With Quote
Old 23rd September 2004, 10:01 AM   #8
diyAudio Member
 
Join Date: Sep 2004
Location: Surrey, BC.
Send a message via MSN to mtnickel
>I would check that the RCA ground is the same as the speaker >ground of the amplifier to discard some quite common problems >ike fused ground tracks on the amplifier [RCA connectors >touching 12V by accident]

The RCA ground to spaeker ground of the amp is 500ohms.
The RCA Left to right is 1000ohms.
The RCA Ground to chassis ground is 10Kohms.


>I would also check that the RCA ground of the head unit is the >same as the chasis ground to discard fused tracks inside the >head unit [again RCA connectors touching 12V by accident]

0 ohms, a direct short.


>Then, I would look for magnetically induced noise. This happens >usually with 4 channel amplifiers when there are two stereo >RCA wires between the head unit and the amplifier and each >follows a different path.
>I suggest checking if you get noise with none, 1, 2, 3, etc... >RCAs connected

With only 1 RCA connected, Noise results.

>In case of fused tracks on the amplifier you may get noise even >with no RCAs connected

Absolutely NO noise.

>In case of ground loop or fused tracks on the head unit, you >should get no noise with no RCAs and noise with one or more >RCAs

This is where I have noise.

>In case of magnetic couplig you should get no noise with one or >no RCA connected and noise with two or more

No different with 2.

I've read and talked to Jason Cuadra (maker of elyptic crossover, electrical engineer)...
He diagnosed some other guys and posted all the steps. Seems to be the most knowledgable guy in noise in electrical systems i know. Posts i read from... http://members.fortunecity.com/jason...noisehunt.html

However...he did not mention any fused tracks...so maybe you are the one to talk to

Anyways, but the results i've found thus far, somehow a ground loop is resulting.

I have a fealing it is a very dirty alternator. With my dads car turned on and jumped to my car, there is no noise. This led me to believe it was completely ground loop/alternator sourced.

The only thing i do to help minimize the noise is run an 8 gauge from head unit ground to amp ground. This effectively reduces the potential. However, there is still noise. Using a transformer removed it, but there were super cheap and not shielded. I cannot afford the Jensen ones

So this is the solution i've come to...differential inputs.

I initially thought it would work as well, becuase i used to have a rockford fosgate with differential inputs and there was no noise in my system. However, i think the potential was too great and in 2 weeks, the circuit blew and there was a direct short from chasis to amp rca ground...extreme noise.

Anyways, any further info would be great. Thanks for the trouble shoot Eva.

mark
  Reply With Quote
Old 23rd September 2004, 12:37 PM   #9
Eva is offline Eva  Spain
diyAudio Member
 
Eva's Avatar
 
Join Date: Oct 2003
Location: Near the sea
Send a message via MSN to Eva
Quote:
The RCA ground to spaeker ground of the amp is 500ohms.
The RCA Left to right is 1000ohms.
The RCA Ground to chassis ground is 10Kohms.
These measurements suggest that your amplifier already has diferential inputs with 500 ohms input impedance on the RCA ground [inverting input], but those 500 ohms may be also due to a fused ground track

To test the input of the amplifier :

Try shorting the tip and the ring of a test RCA connector and plug it to the amplifier. Does it produce noise? [It shouldn't]

Then try connecting both the tip and the ring of the test RCA connector to chasis ground or head unit ground and plug it to the amplifier. Does it produce noise now? [It shouldn't]


Quote:
I initially thought it would work as well, becuase i used to have a rockford fosgate with differential inputs and there was no noise in my system. However, i think the potential was too great and in 2 weeks, the circuit blew and there was a direct short from chasis to amp rca ground...extreme noise.
Some amplifiers use two diodes in paralell, placed back-to-back, to force speaker/RCA ground potential to be within +-0.7V of 12V ground potential. This usually works as expected but in some cars potentials higher than +-0.7V are reached due to bad grounding on the head unit or the amplfier [or excessive chasis/ground wiring impedance] and the diodes blow. Then a dead short between RCA ground and 12V ground appears and it causes lots of noise. It's very easy to repair though [I've seen also PCB traces blown in extreme cases]

Your car is probably suffering from such excessive potentials and the amplifier may be internally protected and simply clamping them [and causing noise too]

I would test the system with the head unit powered directly from the +12V and GND binding posts of the amplifier through long enough wires. This makes still a ground loop but the potential difference should be low enough to be handled by any amplifier [except in case it's broken]
  Reply With Quote
Old 23rd September 2004, 04:21 PM   #10
diyAudio Member
 
Join Date: Sep 2004
Location: Surrey, BC.
Send a message via MSN to mtnickel
Default Testing 1, 2, 3

>These measurements suggest that your amplifier already has >diferential inputs with 500 ohms input impedance on the RCA >ground [inverting input], but those 500 ohms may be also due >to a fused ground track

This impendance from input to output ground could be the "Controlled Electronic Noise Suppression "Cens" " circuitry that is supposed to eliminate noise. It in fact does change the noise from a moderate level to a softer one, but never eliminates it.
I'm not sure if they are differential...perhaps we can find out.


>Try shorting the tip and the ring of a test RCA connector and >plug it to the amplifier. Does it produce noise? [It shouldn't]

With this done, there is no noise.

>Then try connecting both the tip and the ring of the test RCA >connector to chasis ground or head unit ground and plug it to >the amplifier. Does it produce noise now? [It shouldn't]

I have not tried this and am at school now without my car, but seems like an interesting test. I'll definately try this when i get home. My thoughts are that there is noise with this setup simply because the chassis ground now has a link to amplifier RCA ground. This is just speculation and I'll let you know in about 12-13hours.


>Some amplifiers use two diodes in paralell, placed back-to-back, >to force speaker/RCA ground potential to be within +-0.7V of >12V ground potential. This usually works as expected but in >some cars potentials higher than +-0.7V are reached due to >bad grounding on the head unit or the amplfier [or excessive >chasis/ground wiring impedance] and the diodes blow. Then a >dead short between RCA ground and 12V ground appears and >it causes lots of noise. It's very easy to repair though [I've seen >also PCB traces blown in extreme cases]

I think you are dead on with this one....as i remember, i was moving the deck or playing with it or something and probably had a bad if not completely disconnected ground while everything was on. I wish I had known it was so easy to repair. I sold the amp as a dedicated sub amp with the LPF on so that no High freq noise would be heard. Is such a solution possible with my amp?


>I would test the system with the head unit powered directly >from the +12V and GND binding posts of the amplifier through >long enough wires. This makes still a ground loop but the >potential difference should be low enough to be handled by any >amplifier [except in case it's broken] [/B][/QUOTE]

This is where i was a little mistified. I tried this recently with the deck nearly on top of the amp. I connected power and ground right to the amp with 3ft test leads (14 gauge). Signal through a spare RCA (not through car). I still had Noise!
This can only lead me to believe that the it's when the RCA input ground of the amp sees a chassis ground, it somehow amplifies noise. So I'm still a little lost as to the actual cause.

All in all, would these differential converted inputs I would make be isolated from ground completely? Would it work as a solution. I would like to find the right solution however. But it is taking some time. I will probably take my car in to have the alternator tested to see if it is excessively dirty. I'm sure it's original, car has 110k miles. And i have ran a heavy duty stereo for a year and a half now. I usually just lowered the gains as far as completely possible. I now have speakers that will take alot of power and need to have the gains higher as to get very loud .

Anyways, your help Eva, has been the best so far. Thanks

Mark
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
JLH with a differential input stage mohan72 Solid State 46 10th September 2008 07:59 PM
Adding a third input to a differential input stage? maudio Solid State 15 11th October 2006 02:16 AM
Circuit for Pseudo Differential Input? thespeakerguy Analog Line Level 0 6th February 2006 04:43 AM


New To Site? Need Help?

All times are GMT. The time now is 12:34 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2