I'd use a 0.1uF COG ceramic, but the cap used should be low inductance and small.
Now you get the idea, grounds separate for each channel, connecting together at the
supplies only, and each ground as compact and short as possible within each channel.
The inputs and output channel connections should also have separate grounds.
I use ground planes, but a careful layout can do well with traces only.
Think of the ground traces as a return signal line, and handle them accordingly.
They should be near the corresponding actual signal line, and have a small loop area with it.
This is what a ground plane does automatically, unless there are breaks in it.
Can you do ground planes in your layout program?
Last edited:
Hi Rayma,
Thank you very much for your valuable inputs. I have already started working on the lay out and by evening when I get time, I will finish the layout share it here.
Now there are there are nine traces running to the common ground point.
1. two traces from left and right i/p
2. two traces from volume control
3. One trace from balance
4. One trace from supply i/p
5. two traces from left and right output.
6. One trace from op amp ground
For decoupling caps I will use 0.1uF ceramic parallelled with 10uF25V electrolytic.
I Can do a ground plane in "Eagle" layout program, but some one earlier had recommended not to use ground planes for analog circuits, probably due to the reason you gave (breaks). I will try to do a layout with ground plane and share it here (after crefully checking for breaks).
Regards
Prasi
Thank you very much for your valuable inputs. I have already started working on the lay out and by evening when I get time, I will finish the layout share it here.
Now there are there are nine traces running to the common ground point.
1. two traces from left and right i/p
2. two traces from volume control
3. One trace from balance
4. One trace from supply i/p
5. two traces from left and right output.
6. One trace from op amp ground
For decoupling caps I will use 0.1uF ceramic parallelled with 10uF25V electrolytic.
I Can do a ground plane in "Eagle" layout program, but some one earlier had recommended not to use ground planes for analog circuits, probably due to the reason you gave (breaks). I will try to do a layout with ground plane and share it here (after crefully checking for breaks).
Regards
Prasi
About the input caps, i think they could be a bit bigger, the input impedance of the inverting stage is quite small, with only 2,2uF the low end of the audio band will suffer lower gain than the rest of the spectrum, at 30hz for example the input cap it's virtualy a 2,4k Ohm resistor, and the resistor at the inverting input is only 10k...
If this is the final schematic:
http://www.diyaudio.com/forums/analog-line-level/275643-tone-control-tl074-2.html#post4360608
Than it has a small but important other issue, the gain stage of the right channel has it's inputs reversed, it is not an non-inverting amplifier so the input should not be at the IN+ pin ( pin 3 ), but at the IN- ( pin 2 ). I have not verified the layout but i hope this mistake it's only in the schematic...
If this is the final schematic:
http://www.diyaudio.com/forums/analog-line-level/275643-tone-control-tl074-2.html#post4360608
Than it has a small but important other issue, the gain stage of the right channel has it's inputs reversed, it is not an non-inverting amplifier so the input should not be at the IN+ pin ( pin 3 ), but at the IN- ( pin 2 ). I have not verified the layout but i hope this mistake it's only in the schematic...
Last edited:
Sorry for the double post but i cannot edit my other one any more...
I have verified now the layout, and not only the mistake on the schematic of the right channel was made on the board as well but it has been passed to the other one as well. Both IC1 and IC2 now have the inverting inputs ( pins 2 ) routed to ground, and the signal input is routed to pins 3, it is wrong, you need to have the signal input to the inverting inputs, and have the non-inverting inputs ( pins 3 ) routed to ground.
I have verified now the layout, and not only the mistake on the schematic of the right channel was made on the board as well but it has been passed to the other one as well. Both IC1 and IC2 now have the inverting inputs ( pins 2 ) routed to ground, and the signal input is routed to pins 3, it is wrong, you need to have the signal input to the inverting inputs, and have the non-inverting inputs ( pins 3 ) routed to ground.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Hi MarianB,
Thanks for your suggestions. I had made the schematic based on the TI application circuit http://www.ti.com/lit/an/sloa042/sloa042.pdf. I will study a bit more and get back to you with a reply.
Reg
prasi
Thanks for your suggestions. I had made the schematic based on the TI application circuit http://www.ti.com/lit/an/sloa042/sloa042.pdf. I will study a bit more and get back to you with a reply.
Reg
prasi
Greetings Prasi.
Let's put it graphically:
This is what i intended to point out, the slope at the left ( the lower end of the audio band ) is quite steep, and starting to fall quite early. This is because of the low input impedance of this type of amplifier ( inverting one ), wich requires a higher coupling capacitance.
Do not miss out on the other problem i have pointed out, it is very important to correct it.
All the best.
MarianB.
Let's put it graphically:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
This is what i intended to point out, the slope at the left ( the lower end of the audio band ) is quite steep, and starting to fall quite early. This is because of the low input impedance of this type of amplifier ( inverting one ), wich requires a higher coupling capacitance.
Do not miss out on the other problem i have pointed out, it is very important to correct it.
All the best.
MarianB.
For simpler layouts like this one, it is quite possible to use only one side for most (or all) of the traces,
leaving a contiguous ground plane on the other side. This is the best layout method for audio, as well
as other types of circuits.
Hi MarianB,
Thanks for the illustration. I get the point. What do you think is a suitable value of cap? would a 6.8uF 250V Non-polarized be suitable? even this cap is too big (33mm L x 18 mm D). or is there any other simple solution that I am missing?
regarding the other point, it was a silly mistake on my part (I must be sleepy when I did the schematic.)
regards
prasi
Thanks for the illustration. I get the point. What do you think is a suitable value of cap? would a 6.8uF 250V Non-polarized be suitable? even this cap is too big (33mm L x 18 mm D). or is there any other simple solution that I am missing?
regarding the other point, it was a silly mistake on my part (I must be sleepy when I did the schematic.)
regards
prasi
You are welcome Prasi. I don't see the point in using big coupling caps like that, you are aware i think that this project is not an high performance one, so why bother with that when bipolar electrolytic caps should do just fain. I know what people say but i do know what i have listened as well, and i assure you, you'll be more than satisfied. As for their value, if you would decide to go with bipolar caps, probably more like 10uF would be the best way to go.
Be sure to correct your mistake on the layout as i have pointed out, as well as in the schematic, and do not worry about making mistakes, everyone makes them, or has made them at some point, More so, i think the most valuable lessons are learned from mistakes.
Cheers,
Marian.
Be sure to correct your mistake on the layout as i have pointed out, as well as in the schematic, and do not worry about making mistakes, everyone makes them, or has made them at some point, More so, i think the most valuable lessons are learned from mistakes.
Cheers,
Marian.
Hi Marian,
Here are the modified layout and schematic as per your suggestions. Hopefully I have done it right this time. I have kept the option of using either electrolytic or film caps for i/p coupling. Will try both to see which gives better performance.
Yes, indeed, I learnt a lot while doing this small project and without guidance from yourself, Mooly and Rayma, I would have been stuck.
It is somewhat disheartening to know that it is not high performance circuit. I have been working on this for quite a while now!!
. Would you give me a link to any of the tone control schematics/PCB that can b considered as audiophile quality?Hi Rayma,
Ok. I got it. Ground plane on one side and all signal and power traces on the other. It will reduce quite a bit of my headache while doing the layout, trying to lay traces through tiny spaces.
One again thank you all for your help.
Reagrds
Prasi
Attachments
I wouldn't describe it as not being high performance, its a fairly typical standard type circuit. Yes, there are always things you can alter and change but as usual, its playing one aspect of performance off against another. The only thing I might look at would be the 100k pot treble area, 100k is a bit high, not just for noise but for being more prone to stray pickup of noise. The 1k output resistors are to high for my liking, somewhere closer to 100 ohm would be my choice. Output caps... well with FET opamps there won't be any offset worth worrying over and so they could perhaps be omitted.
Doug Selfs tone control from his 1996 precision preamp is well worth a look as it has separate bass and treble turnover points that are continuously adjustable.
Doug Selfs tone control from his 1996 precision preamp is well worth a look as it has separate bass and treble turnover points that are continuously adjustable.
I did not mean to disappoint you, so if i did that then i do apologise for that. Maybe i underestimated your expectations...
Most of high performance mixers have loads of small IC's ( tens, if not hundreds ), most of them are SMD, great circuit and pcb complexity, that is why they are expensive, that is what i meant when i said this project is not a high performance one, it should not need to be, because true high performance is only measurable with complex and special lab devices, not by ear wich is a very subjective organ. So i am quite sure that you will love the end result of your project, and i doubt very much that you will be able to sense a true difference between bipolar coupling caps and the other ones, bipolar caps are those caps that have no polarity:
It looks like a normal electrolytic capacitor, just that it does not have any polarity, they say this type of capacitor is better than normal polarised ones, for the signal path, and they are cheap, so you should at least try them, i am sure you will enjoy the end result.
As for the layout, you did exactly what i suggested, it seams to be ok now. Don't overdue trying to get a perfect board layout, i do not think there is such a thing, as it has been said, this is more of an art, and very good layouts only come with great experience, you should first concentrate on getting all the electrical connections correct, than on choosing the best path for each trace, wich usually is the shortest, try to understand correct ground tracing wich is very important, than you could concentrate on the small stuff, you first need to obtain a correct functioning layout, than you can think about finishing touches, like let say pads and vias, wich i think should be a bit bigger, you will se the benefit of bigger pads and vias when soldering the components ( trust me ).
Regards,
MarianB.
Most of high performance mixers have loads of small IC's ( tens, if not hundreds ), most of them are SMD, great circuit and pcb complexity, that is why they are expensive, that is what i meant when i said this project is not a high performance one, it should not need to be, because true high performance is only measurable with complex and special lab devices, not by ear wich is a very subjective organ. So i am quite sure that you will love the end result of your project, and i doubt very much that you will be able to sense a true difference between bipolar coupling caps and the other ones, bipolar caps are those caps that have no polarity:
It looks like a normal electrolytic capacitor, just that it does not have any polarity, they say this type of capacitor is better than normal polarised ones, for the signal path, and they are cheap, so you should at least try them, i am sure you will enjoy the end result.
As for the layout, you did exactly what i suggested, it seams to be ok now. Don't overdue trying to get a perfect board layout, i do not think there is such a thing, as it has been said, this is more of an art, and very good layouts only come with great experience, you should first concentrate on getting all the electrical connections correct, than on choosing the best path for each trace, wich usually is the shortest, try to understand correct ground tracing wich is very important, than you could concentrate on the small stuff, you first need to obtain a correct functioning layout, than you can think about finishing touches, like let say pads and vias, wich i think should be a bit bigger, you will se the benefit of bigger pads and vias when soldering the components ( trust me ).
Regards,
MarianB.
Hi,
Should I change the pot to more like 10k or 25 k for the treble? what change do I need to make to accomodate this? Also I will change the o/p resistor to 100 Ohm. If amp is A/C coupled, then I will replace the o/p caps with jumpers.
regards
prasi
To reduce the pot by a factor of ten would need any resistors associated with that network to also be scaled by the same factor. Caps would need increasing by a factor of ten. There are infinite permutations of these networks and values. The opamp has to drive the feedback network though and so you should ensure you don't go to low such that the opamp is unable to deliver enough current to the network.
This might help explain how it works.
Op-Amps as amplifiers and Tone Controls
This might help explain how it works.
Op-Amps as amplifiers and Tone Controls
I did not mean to disappoint you, so if i did that then i do apologise for that. Maybe i underestimated your expectations...
Most of high performance mixers have loads of small IC's ( tens, if not hundreds ), most of them are SMD, great circuit and pcb complexity, that is why they are expensive, that is what i meant when i said this project is not a high performance one, it should not need to be, because true high performance is only measurable with complex and special lab devices, not by ear wich is a very subjective organ. So i am quite sure that you will love the end result of your project, and i doubt very much that you will be able to sense a true difference between bipolar coupling caps and the other ones, bipolar caps are those caps that have no polarity:
It looks like a normal electrolytic capacitor, just that it does not have any polarity, they say this type of capacitor is better than normal polarised ones, for the signal path, and they are cheap, so you should at least try them, i am sure you will enjoy the end result.
As for the layout, you did exactly what i suggested, it seams to be ok now. Don't overdue trying to get a perfect board layout, i do not think there is such a thing, as it has been said, this is more of an art, and very good layouts only come with great experience, you should first concentrate on getting all the electrical connections correct, than on choosing the best path for each trace, wich usually is the shortest, try to understand correct ground tracing wich is very important, than you could concentrate on the small stuff, you first need to obtain a correct functioning layout, than you can think about finishing touches, like let say pads and vias, wich i think should be a bit bigger, you will se the benefit of bigger pads and vias when soldering the components ( trust me ).
Regards,
MarianB.
Hi Marian,
Thanks for the elaboration. I too hope that I will enjoy my creation. I will find about bipolar caps (farnell does not seem to have them) and if available will definitely use them.
I have checked and rechecked the wiring, trying to reduce signal path length and jumpers. I was a bit successful in that I reduced the jumpers from 6 to 4 and also reduced trace lengths for a number of traces. Also increased the pad sizes.
I will work out pot needed for treble and corresponding values of the passives as Mooly suggested and will try it out on proto board itself.
I am again reattaching the final schematic and layout for anyone interested to comment.
wish me luck!!
Prashant
Attachments
Hi all,
Today I built the tone control board and tested it on my LM3886 gainclone amp. However the is an audible buzz heard when either the treble pot / bass pot metal shaft is touched by hands. really strange. Also volume on left channel is quite low compared to right channel.
Also there is something wrong with bass adjustment circuit , when I rotate the bass pot counterclockwise, the bass is increased. where as with treble, its working correctly.
I wonder anyone has faced similar problems.
regards
prasi
Today I built the tone control board and tested it on my LM3886 gainclone amp. However the is an audible buzz heard when either the treble pot / bass pot metal shaft is touched by hands. really strange. Also volume on left channel is quite low compared to right channel.
Also there is something wrong with bass adjustment circuit , when I rotate the bass pot counterclockwise, the bass is increased. where as with treble, its working correctly.
I wonder anyone has faced similar problems.
regards
prasi
Hi prasi,
I just stumbled onto the thread this morning and thought I'd toss in my ideas even though I'm a late entry into the game, and you have already built a fairly standard tone control. I'd like you to take a look at a design I posted a while ago that addresses a lot of issues that I believe are inherent weaknesses of the usual Baxandall designs. It's posted here: http://www.diyaudio.com/forums/chip-amps/199130-usable-tone-control.html. I recommend you just read through the whole thread for the ideas that were incorporated into it. The circuit also went through some improvements as it evolved some over time. Post #18 has a major improvement that I believe really sets my design apart for all standard types. Oh, and one more thing. I recommend sticking with polypropylene caps in the filter sections , but Mylar is acceptable for input/output coupling.
Mike
I just stumbled onto the thread this morning and thought I'd toss in my ideas even though I'm a late entry into the game, and you have already built a fairly standard tone control. I'd like you to take a look at a design I posted a while ago that addresses a lot of issues that I believe are inherent weaknesses of the usual Baxandall designs. It's posted here: http://www.diyaudio.com/forums/chip-amps/199130-usable-tone-control.html. I recommend you just read through the whole thread for the ideas that were incorporated into it. The circuit also went through some improvements as it evolved some over time. Post #18 has a major improvement that I believe really sets my design apart for all standard types. Oh, and one more thing. I recommend sticking with polypropylene caps in the filter sections , but Mylar is acceptable for input/output coupling.
Mike
Last edited:
I think that can only be down to the pots used. Are they metal bodied ? If so then perhaps try grounding the body.
That sounds like a simple construction error/wrong part value somewhere.
That sounds like the board layout has the two end pins of the pot reversed so that its working back to front.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.