Simple Surround Sound Decoder

This article is written with an intent to explain in detail behind the making of a simple surround-sound decoder circuit.

By: Dhrubajyoti Biswas

Overview

The concept of the decoder was first introduced by David Hafler in the 70s. His research illustrates the way to use two speakers as rear speakers on a surround system.

The figure below is a diagram based on Hafler’s research:



Figure 1

According to Figure 1, Hafler designed the circuit to enable the rear speakers generate the difference of signal between right and left output.

While every stereo encoded system maintains difference of signal between the right and left channel, it is that difference of signal when received by the rear speakers gets reproduced.

However, it is vital to keep in mind not to earth the negative terminals of the rear speakers, else the rear will behave parallel to the main front speakers.

Line Level Passive Version


Using individual amplifier for rear speakers is not actually possible. However, there is a way-out which we figured out after some research. Referring to Figure 2, it is totally passive, but it needs an ideal transformer – a transformer with impedance of 10K [1:1 ratio], which is quite rare to find, but available.


Figure 2

As an alternative we have tried using a 600ohm unit. But it is for the impedance the output we received was not good as it lacks bass.

However, upon loading the transformer, it increased the bass quality but the preamp doesn’t seem to work at its fullest because of the impedance. It is for this reason that we have used telephonic transformers with 600:600ohms, and it worked well.

The circuit in Figure 2 illustrates the way we followed. Following this design, it worked, but it has very low impedance on all cases barring solid-state preamp.

Using 600ohm unit, the loss generated is around 3dB. The low frequency is -3dB on 100Hz. However, it varies based upon the quality of the transformer.

600ohm telephony transformer is widely available in the market, but many of them are not up to the mark to use it in this experiment.

Most of the hi-powered transformers are sold in bulk and is therefore hard to procure a single copy. So, the alternative would be to use dual opamp to design the system, and its process is mentioned below in detail.

Explaining the New Circuit


The schematic diagram in Figure 3 gives a detailed view behind this development of the simple surround sound decoder circuit.





Figure 3

While the new design [Figure 3] will follow Hafler’s principle, this new circuit has simplified wiring, albeit we needed extra power amps. There is now a center channel signal and the sub-woofer to receive mono signal is also set.

You may have encountered similar type of circuit on other papers, but there are some twists into it. We avoided any active electronics on left/right channels and introduced opamps to zero down the factor that may cause degradation of sound.

The 50K impedance will not pose any barrier for a preamp, as the main signal is parallel to the additional circuit.

Extra volume control has been excluded from the system, because of the presence of volume control in the preamp. Moreover, the power amp of rear channel also has level control to balance the front and rear levels.

Please note, if you are following the circuit as in Figure 3 do ensure to make the rear speakers wired-out phase.

Let one speaker connect to the amp on a normal fashion and the second should be connected keeping the leads of speaker reversed.

Though the difference maybe negligible, but to derive the best quality effect it is always advisable to opt for out-of-phase connection.  This helps in maintaining left-right and right-left signals.

The way surround sound decoder circuit works

A1 opamp should be connected in the form of subtracting amplifier, and if same signal is passed to both speakers, the result will be Zero.

This will result to removal of all information that are common from the stereo signal, and would produce the difference signal, similar to that of Hafler’s. A2 on the other hand is a summing amplifier. Its output has all necessary information from the left and right channels.

Center Channel Control


VR1 pot is set to level the center channel. It can either be a conventional pot or trimpot with the rear mounted.

Adding up the two channels [left / right channel] where signal is not mono, -3dB will be will be the level of center channel.

For instance, if the center channel speech is mono then the level becomes equal on both speakers. The possibility of amp overloading or the speaker is a rare case here, since the speakers and channel amplifier are not as much powerful compared to the left/right channels.

The sound of center channel does not need to be high. It has to be stable and the available level control is quite enough to generate required output.

The use of C1 Capacitor is not mandatory as it provides roll-off frequency of 8kHz. This actually helps to reduce any issues on the signal of the main stereo.

Output – Sub-Woofer


The output of the sub-woofer is taken from central channel mixer and added no-pass filter because it is hard to determine a sub where there is already a filter.
Other factors

100ohms resistors are used to block the oscillation of opamps by preventing the capacitance of the signal lead. Following this would not result to loss in frequency, but if you use 100m long signal leads, it may pose problem.

Referring to Figure 3, the rear speakers have two outputs in parallel.

The reason to do it is to enable easy wiring to facilitate connection of stereo amplifier with the rear speakers.

Normally, mono amplifier would do fine as long as it drives parallel to the two rear speakers. But this may not be feasible if you are using 4ohm speakers and if you do use it then ensure to connect them in the form of series. In order to enable out-of-phase connection, the red terminals need to get joined, and further connect the terminal of the speakers to the output of the amplifier.

Building the System


You can place the entire system on a metal case. Using metal case blocks the hum or other noise coming from mains etc.

While there is no factor of heat generation, you can use small case. However, ensure to maintain space to fit the RCA connectors and rest of the components.

Also, be sure to not to set the components loose as this may lead to short circuit.

You can wire the components and the dual opamp on a Veroboard. Also do ensure to apply 1% metal film all over it to lower the noise.

You may keep the RCA connectors hard-wired. Do ensure to check the earthing.

The power supply center lap and the RCA connectors should maintain secure connection to avoid noise pickup. You can also use 100uF polyester caps to connect with 100uF supply bypass capacitors in parallel, but this is not mandatory.

Delay Line

If you are planning to enrich the sound you can also apply delay line in order to delay the sound going into the rear speakers. But that is again not mandatory.

Overall, performance of your system is fully dependent on the way you have arranged the circuit. If the proposed simple surround sound decoder circuit is not well built you may face constant issues compared to a well-built one.

Need Help? Please send your queries through Comments for quick replies! And please Bookmark my site :)




Comments

Sir i have 3 doubts in last circuit using TL072 IC
1). will the IC capable of driving 5 nos of 10 W speakers with maximum power output at extreme level of music ( maximum dB ).. i need distortion free working of these 5 speakers with maximum capable sound level without adding any noise into them and operate the subwoofer separately

2). can i add a separate amplifier section for a subwoofer of 100 W power rating at pin no. 6 of the IC TL072 for good bass ? for that can i use the circuit posted by you in this blog ?
https://homemade-circuits.com/2012/01/how-to-make-outstanding-home-theater.html?m=1

3). What is the maximum power output of the above mentioned home theatre system with subwoofer and one tweeter ? i can make decision about the selection of an appropriate subwoofer after hearing your reply


expecting your reply soon
Sir one more doubt..
How to control the volume of the entire setup ?
Swagatam said…
HI RT, here are the answers:

1) the circuit is designed for 4 speakers only, not 5...and it needs to be attached with an external amplifier for powering the speakers

2) yes all the speakers will require amplifiers for the reproduction, the TL072 is only for processing, it cannot amplify the sound for the speakers directly...you can use any of the mosfet based amplifiers from this site for the sub

3)this will again depend on the amplifier wattage that you are supposed to use externally...

Swagatam said…
you can feed the sound through a pot to the specified inputs and these pot can be used as volume controls
Sir actually what happens in a less expensive surround sound system which doesnt utilize any types of dolby or dts decoding method ?
(1). Are the processes which these systems doing only addition and subtraction of stereo signals ?
(2). Are those hometheatre systems available in market at cheap rate ( below 5k or even 10k ) based on this principle or using correct decoding according to dolby algorithm ?
still persists.... and when power returns back the LED becomes brighter instead of turning off automatically.... so i am expecting a suitable solution from you for this...



sorry for a very long description about thia .. i have mailed the circuit your inbox... hitman2008... plz check ot out and respond to me as earlier as possible
sir i have emailed the circuit to your hitman2008 inbox.. plz check it out and respond
Swagatam said…
RT, sorry I do not have much idea regarding the systems which are available in the market, not sure how these are specified with their features
Swagatam said…
which circuit are you referring to?
sir please reply to my email... what is the problem in my circuit sir ?
sir actually what i need is when darkness is deteted during power failure ( say power failure during night ) an led lamp should turn on immediately and when the power returns it should turn off automatically....

This setup have to have an additional feature of a touch switch to activate or deactivate the lamp at any instant...

if i am using a OR logic for the DARK SENSOR output and the TOUCH ACTION i can never switch off the LAMP if i wish to do that during power failure at night ( say when the DARK SENSOR is functioning).. so please suggest a suitable logic for this sir.

1). the DARK SENSOR and the TOUCH circuit should be completely independent
2). The Touch circuit have to function at any time
Swagatam said…
sorry I still did not check it... but will do it soon...
Swagatam said…
Hi RT, I'll try to figure it out soon and let you know...
Swag Majumdar said…
Hi RT,

you can do this, configure the LED with the transistor associated with the LDR circuit, and configure the touch sensor transistor with the LDR transistor such that it switches it ON/OFF as per the touch operations.
sir it doesnt fulfill my need. if i am doing so the touch operation will be available only when the dark sensor is active... i want the touch operation to be available at any instant the user wants to do

my requirements for this fully automatic intelligent emergency light is

(1). There will be a touch switch for this lamp which will basically switches it on whenever the touch button is pressed( say touched ) and switches off when another touch comes ( touch output 0 )

(2). The LED lamp should turn on automatically when power failure occurs during night . that means it shouldn't turn on during night with mains power available. ( so in my proposed diagram i am putting a 10k resistance to the base of the transistor associated with dark sensor and the other end of the 10k resistor is given to the +5 V from a mobile charger.. hence the dark sensor output will be high only when power failure occurs during night )

(3). THIS IS A VERY IMPORTANT AND CRUCIAL CONDITION.
Suppose the power failure occurs during night... the LED lamp turns on automatically.... Now if i want to turn off the led I press the touch button.. so a logic should be needed which will immediately turns off the lamp completely when a manual touch is detected during night power failure .. and at any instant, during the power failure stage, if the user wants to activate the lamp again, another touch will have to do the purpose.
Now suppose i have pressed the touch switch during night power failure to turn off the LED .. now the LED turns off.. Touch button output is high now ( since the SGL8022W latches its output state untill the next touch input arrives ) .. suddenly power returns.. now the dark sensor output becomes in low state but the touch output still remains in high state... because of this LED turns on again when the Power returns.. but i don't want this to happen ... whatever may be the condition of lamp during night power failure, it should automatically be turned off when power returns.

(4). The opposite of case (3). whatever be the condition of Touch switch ( 0 or 1 ) the LED lamp should turn on immediately when power failure occurs during night...
the normal problem caused is if we are enabling Touch operation during power available and when suddenly power failure occurs, the lamp doesnt turns on automatically since the touch enable is detected similar to case (3) which will cause the lamp to immediately turns to off state.
Swagatam said…
RT, In short you want the following things in your system:

Touch operation should be the primary switching option which should be able to override the other options anytime.

During power failure when the LED is switched OFF through the touch switch, the situation should reset to the initial condition as soon as power is restored.

The circuit should not light up during day unless the touch switch is activated.

Please confirm!
yes sir... you said it... exactly the same i want.. The Touch switch will be the master switch... and the condition of Touch switch has to reset to 0 when a variation of mains power detected... that means when Power failure occus touch stage output resets to 0 . same is the case for return of power .. touch output state becomes zero. so for both these case Touch action will be treated separately and completely independent
Swagatam said…
OK, but does the IC SGL8022W have a reseting facility in it or can it be reset if by switching it OFF and then ON?
i was also searching for that in the website.. but didnt succeed to find any information regarding that.. i got aome pdf about this ic.. that is all.. but no such information is broadly given in these pdfs.. i got this ic from a 2 year old chinese touch emergency lamp... this ic is very feasible to operate with touch input. other than its touch function we can control its output voltage by continously pressing on the touch plate which causes a reduction or increment of output voltage from the ic which will immediately affects the brightness of the lamp.... also since this ic not available in local markets i am not ready to experiment with this ic for any reset input anymore.....
Swagatam said…
you can try by switching the supply off and then ON to check whether or not the IC resets back to zero...if this does not happen then it would be difficult to accomplish the required outcomes.
k sir thank you very much


can you please suggest a simple circuit to measure the flow of current through a path and when the current drops below a certain value ( say 10 mA ) the output suddenly switches off.... i appreciate a simple circuit with either transistors or opamp
Swagatam said…
You can refer to the following design and modify it for your specific application:

https://homemade-circuits.com/2015/04/battery-current-indicator-circuit.html
Sir in the surround sound system circuit, why the rear right speaker's input is taken from the same input pin of reat left ? why same input for rear right and rear left ? can you please explain the functions of these pins and connections ?
Swagatam said…
RT, the respective signals are processed by the two opamps in such a manner that only the difference of content between the two channel are reproduced across the rear channels...all the frequencies which are similar are rejected.

Please go through Hafler's principle which might help you to understand the concept with greater depth.
sir in a simple charging circuit, positive of the battery ( say 3.9 V ) is connected to the Emitter of a NPN transistor and the +ve charging voltage from an adaptor ( say 5 V , 500 mA ) is connected to the Collector terminal.. so according to the transistors working principle the charging can be done only if a Vbe is applied.. so i have a control voltage of 5 V to control the charging.. when i am applying 5 V in the base through a 10K resistor the battery is getting charged... but the problem is the amount of charging current drawn by the battery from the source is very less ( say 60 mA ) eventhough the adaptor is capabl of delivering 500 mA charging current ... i have tried using PNP, power transistors, BD139 but no considerable change has been seen... can you please suggest a better solution for this ?.. A circuit which can charge the battery with its maximum deliverable current but under a controlled signal ?
Swagatam said…
Hi RT, the NPN should be preferably a Darlington type, for example try a TIP122 and see how it performs.

You cannot force a battery to consume current above its rated capacity, if you do so then it can get damaged.

I would suggest that you connect your 5V supply directly with your battery through a 1N4007 diode and check the current, if still your 3.7V Li-Ion cell consumes 60mA then probably it could be faulty...please check this first to confirm the exact reason of the issue.
(1).Sir even connecting the battery after diode yields a current flow of 250 mA... so please suggest a way in which i can increase the charging current to a minimum of 150 mA...
The TIP122 is giving charging current of 80 mA or less only.. Power Transistors like 2N5294 and MOSFETs are also not helping...


(2). Sir suppose an led lamp connected in between the collector terminal of the NPN transistor and the +ve DC voltage.. when a base voltage is applied the Lamp turns on... Similiar to the above case i want the brightness of the lamp similiar to the case when the lamp connecting directly between supply rails.. actually the brightness of the lamp decreases while controling with a transistor... Same thing happens even for Darlington Transistors, Mosfets etc.. Is there anyway to get the full brightness level of the lamp ? This is an another case of my present problem
Swagatam said…
Sorry, do you mean you want to decrease the current to 150mA..??

with TIP122 try 6V, that will boost the current a bit.

2) what exactly are you trying to make, please explain it correctly, is it a DC lamp dimmer circuit? By the way a mosfet or BJT will produce 99% full brightness on the lamp...what you are saying is incorrect unless your transistors are faulty.
(1). i want a minimum charging current of 150 mA sir not to decrease to 150 mA... 150 or above... but in my case even a TIP122 is not giving that much current .. in its emitter terminal it gives only 80 mA current drawing from the source which is connected to its collector terminal... i need a setup to charge the battery with minimum current of 150 mA and the charging to be controlled by a control voltage ( say VBE of the BJT ). . i repeat not to decrease the current to 150 mA....


(2)and for the lamp case.. I just need a emergency lamp with some features like dark sensor, touch switch etc... I am taking the output between the collector of the output transistor and the +ve battery terminal....

First i have checked the lumens produced by the lamp by connecting directly between battery terminals... i have used an app in my mobile to find the lumens.. it has shown a value of 2250 at 30 cm distance from the light source.. when i have checked the value after connecting the lamp in between the output transistors collector terminal and the positive battery rail , it showed only

800 lumens for BC548, BC547
1500 lumens for D965 NPN transistor
1200 Lumens for MOSFETS ( IRF540, 3205, Z44N )
1845 lumens for TIP122.

since i got highest value with TIP122. I thought a pair of TIP122s connected in darlington mode will give the maximum possible lumens ( brightness ).. but it resulted only a slight difference.. merely 1900 lumens... so i am asking you how to get minimum 95 percentage of lumens ( brightness ) ?


i have sent you the schematics to your hitman inbox. please check those out
Swagatam said…
the first sentence in your earlier comment confused me that's why I wanted to get it clarified.

As I mentioned you can try a slightly higher voltage with TIP122 for boosting the current.
By the way if you are getting 250mA by direct connection then why not use it in that way? Direct connection will give you the highest possible charging current, although that would mean a constant monitoring making sure that the cell does not heat up.

Your meter readings are incorrect. There cannot be so much difference between direct connection and connection through a mosfet or BJT, provided these are correctly configured.

The correct method is to keep the LED at the end of a pipe and check the lumens from the other end of the pipe.

or may be your transistors are not original, that's why these are not giving optimal results

BC547 cannot handle more than 100mA so testing with BC547 is out of question.

TIP122 is fine...and it should give you 99% of brightness, unless you are doing something wrong.
Swagatam said…
I'll check my inbox as soon as I am free...
sir thank you .. but i still didnt get the answer for my first question.. i need to control the charging of the battery as it is a part of an automatic battery charger circuit. the battery is connected to the output transistor and its charging is controlled by the base signal applied to it.. when a higher cut off is detected by the automatic battery charger circuit before that transistor ( i am not giving the details of the circuit ) base to emitter voltage becomes zero and charging stops.. unless otherwise Vbe will be high always resulting in the charging of battery... every circuit portions other than this charging portion are perfectly working... but i am stucked in this stage.. not geting sufficient charging current for the battery even with TIP122...dont know what to do..
and also the charging voltage is seen 6.9 V .. so i am charging the battery with an adaptor giving approx. 7 V and 500 mA current..

in every automatic battery charger circuits relays are utilized to make the battery terminal to contact directly with the +ve dc supply rail which results in the maximum possible charging current... but here since my battery voltage is less ( say 4.2 V ) and using a 7V charger i dont want any relays anymore.. that is why i have used transistor switching technique according to a control voltage to charge the battery .. but it is resulting these less charging current problems
Swagatam said…
RT, I did not know the source of voltage to the NPN that's why I could not understand your problem clearly...the NPN has to have sufficient base current if it's used in the "emitter follower" mode to drive an equivalent load...may be the IC is unable to provide enough current to the transistor

For an optimal response either you can try the transistorized circuit from this article:

https://homemade-circuits.com/2012/07/make-6v-4ah-automatic-battery-charger.html

or use a TIP127 and connect its base with the collector of the existing NPN (should be a BC547), this junction must be connected to the positive rail through a 10k resistor, emitter of BC547 now must go directly to the negative rail.
finally you can connect the battery across the collector of TIP127 and the negative rail
Swagatam said…
Please follow the instructions as mentioned in the previous comment and it will definitely work
Sir in one of your previous comments u have said to test the current drawn by the battery while charging from the source when a diode is used in between the source and the battery also test the circuit without the diode.. when i am charging the battery by connecting directly to the dc source maximum charging current reading is shown in multimeter.. when a diode is used the charging current reduces ..

(1)what is the cause for this sir ?
(2)how the diode acts as a current limiter ?
(3)is there anyway to increase the current after the diode ?
Swagatam said…
RT, a diode can never act like a current limiter, if it does then it could itself become hot and get destroyed...in your case the reduction in the current could be because of the 0.6V drop created by the diode, and this may be a significant value since the cell voltage is not large, just 3.7V.

if you are connecting the battery through a transistor, then the transistor would itself work like a zero drop diode, no need of adding a series diode.
sir as per the haflers principle as shown in this figure
https://4.bp.blogspot.com/-QM998gRS9lA/V3T7P6BjgqI/AAAAAAAAOLM/O1iTG2j80NYKS07JeRaVnc93R_6W-UJtQCLcB/s1600/surround.png

the rear speakers are connected in such a way that the -ve of one speaker goes to the -ve of the other one and the other terminals are fed with the respective left and right signals.. but as per the circuit diagram put forwarded by you, the rear left and rear right speaker's -ve terminals are grounded and the other terminals are fed with a single signal (from Pin#1 of the IC ).. what logic is there in this sir ? since both the speakers are driven in a same way ( both the -ve terminals grounded )they will take same signals through the 100 ohm resistors and will operate simultaneously...

i have tested the circuit and came to know that there is no difference in sound production happening in these speakers.. so please confirm this and suggest the correct configuration.. yhe explanation given in this article about the circuit is not at all satisfactory and a normal person cant understand anything from this.. so please help sir
I have two other issues to say which i have found operating this circuit.. please give suitable remedies for these ;
(1). When the power is withdrawn from the circuit , there is a high frequency peak sound heard from the rear speakers.. i have sent the recorded sound file to your email..

(2). The center speaker produces slight humm and the volune is less compared to the Front Left and Front Right speakers even it is expected to be more than them since it is the sum of the signals produced by them

(3). I have one doubt also.. What is the real function of pot VR1 ?
Swagatam said…
RT, the first diagram shows a general configuration whereas the circuit diagram is designed with a complex configuration, and the indicated connections must be correct.

By the way this post was written by another author, so I do not have much idea regarding the subject and the diagrams.

I am sorry I am not quite well versed with surround sound concepts so I won't be able to provide any expert advise regarding this....
Swagatam said…
RT, I'll try to investigate by researching the concept and if possible try to upadte the info....
chihacheol loudspeaker said…
How can I get it to work on a SINGLE SUPPLY?
HURRY!
Swag said…
Single supply may not be possible. You can easily build one using the following circuit:

https://homemade-circuits.com/wp-content/uploads/2012/01/powe-supply-1.png

Contact me for Customized Circuits

Name

Email *

Message *


 Follow on G+  Follow on Facebook   Follow on Tweeter  Follow on G+  Follow on G+

Follow Homemade Circuits