Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Suitable replacement for TRX6221?

Status
Not open for further replies.
Awesome, thanks.
 

I have another question. Since an equalizer is used to adjust music to one's own liking, does it actually matter if high accuracy resistors and capacitors are used?
 

The original purpose of an equalizer was to flatten the frequency response of the system to accomodate variations in room acoustics, speakers and the amplifier itself so that the sound was reproduced as much like the original source as possible. Liking or not should be a separate issue.:wink:

The precision of resistors and capacitors used in your circuit are not critical except for the ones in the filters, especially C1 - C20.
 
Last edited:
  • Like
Reactions: nabiul

    N

    nabiul

    Points: 2
    Helpful Answer Positive Rating
Well people will find new ways to use equipment. To me recorded music sounds nothing like the real thing, so instead I'd rather just adjust it to my liking.

C1-C20, wait that's pretty much all of them. :O

Any idea how expensive high accuracy components are? All I've used so far are 5% resistors and electrolytic and ceramic caps.
 

C1 - C20 can all be MLCC X5R types rated at 50V and 10% and range in price from $0.15 to $0.85 when purchased individually at Mouser Mouser Electronics- Distributeur de composants électroniques. Shop around at Digikey Electronic Components Distributor | DigiKey Corp. | US Home Page and Jameco Jameco Electronics - Electronic Components Distributor. I don't think you'll benefit from tighter tolerance components and they would cost several times as much.

By the way, what you have in that schematic is a series of Sallen-Key active band-pass filters. For more on that, look at section 16.5 in this paper: **broken link removed**.
 
Last edited:
  • Like
Reactions: nabiul

    N

    nabiul

    Points: 2
    Helpful Answer Positive Rating
Thanks, I'll read up on it after my exams are done.
 

This shows the effect of a 50% variation in one of the capacitor values in the 1kHz T section with the pot at 25% and 75% positions. An error in the other capacitor would affect the high frequency corner. You can decide how much tolerance you are prepared to accept.

Keith.
 

Attachments

  • Audio equaliser.pdf
    9.1 KB · Views: 107
  • Like
Reactions: nabiul

    N

    nabiul

    Points: 2
    Helpful Answer Positive Rating
That doesn't seem so bad, only appears to be a 150Hz overlap with the low frequencies. And regular ceramic capacitors are rated at what, 10% usually?
 

Yes, I didn't think it looked too bad - it depends on your point of view! Given that a graphic equalizer is probably just adjusted by ear until you like the sound of it then I don't think using 20% capacitors would really be a problem. That gets you some cheap ceramics and allows the use of electrolytics where necessary.

Keith.
 
  • Like
Reactions: nabiul

    N

    nabiul

    Points: 2
    Helpful Answer Positive Rating
I got some surface mount capacitors for the small values and regular electrolytics for the large ones. Unfortunately the surface mount types come in packs of 10 so I bought way more than I needed. Total cost including copper clad board, etchant and 10 pots (I'm making seperate equalizers for each channel) came to $50. Once the first one is assembled and I can verify that it works, I'll get the parts that I didn't buy enough of to made the second one.

The only problem right now is a power supply, they didn't have bare transformers where I bought the stuff from. I'm thinking about just breaking open one of the numerous 12v adapters I have lying around to scavenge it's transformer.

What would you guys suggest for constructing the circuit, make the distance between discrete components and traces as large as possible, or does it not matter? I'm thinking I'm going to make everything surface mounted and have the other side of the pcb connected to ground to reduce interference... or is that really not necessary/bad?
 

Using the back of the PCB as a ground plane is good. So is leaving all the unused copper on the top as ground and it saves on etchant. I had assumed that you were going to use the existing artwork for through hole construction, but I like SMT. Let us know how it works out.
 
  • Like
Reactions: nabiul

    N

    nabiul

    Points: 2
    Helpful Answer Positive Rating
I will, unfortunately I can't begin construction until dec 15th. I'm supposed to be studying right now for my exams... but my head is somewhere else.
 

Another question. When it comes to avoiding overvolting things like LEDs, what matters is the voltage dropped across it right and not the absolute voltage that it sees?

For example, say I wanted to make a VU meter by taking the output of those filters, amplifying them to have an output of like 7 volts and then applying that to a bunch of LEDs connected in series, with a resistor to ground between each led connection. As long as the voltage dropped across each LED is less than like 3 volts, they won't blow right?
 

C1 - C20 can all be MLCC X5R types rated at 50V and 10%
Click to expand...
Sorry for placing my comment a few days late. I completely agree about the 10% tolerance in case of an octave equalizer. But I would add a resevation for high dielectric constant capacitors in the filters (and also for electrolytic capacitors). They are probably O.K. for usual home applications. But they will cause detectable distortions due to their voltage dependant capacitance. So you won't use them for filter applications with professional or "high end" audio equipment, only high quality film capacitors (or possibly linear low er ceramics) would be choosen for filters in this field.
 
  • Like
Reactions: nabiul

    N

    nabiul

    Points: 2
    Helpful Answer Positive Rating
Erm, so for the higher values I should get tantalum capacitors instead of electrolytics?

I am going to use this for hooking up two studio monitors that I'm using as computer speakers. They sound fantastic as they are, but I'm annoyed that there isn't a global equalizer in windows.

I'm not so sure now though after a few days of listening to them on 'flat' settings. I may just adapt the circuit for a large VU meter.
 

As long as the voltage dropped across each LED is less than like 3 volts, they won't blow right?
Click to expand...
Wrong. LEDs are usually damaged or destroyed by overcurrent. Blue and white LEDs typically won't go to their full rated current until they have almost 4V while a red or yellow led will be destroyed by 4V. If you build a VU meter in the way you describe, you should endeavor to limit the current to a safe maximum regardless of voltage and that limit will be different for each LED color/chemistry.

A VU meter as it's name implies is used to measure Volume Units without regard to frequency. What you're getting with your 10 bands is an audio spectrum analyzer. You can use an LM3915 **broken link removed** to make 10 levels for each frequency band.
 
Last edited:
  • Like
Reactions: nabiul

    N

    nabiul

    Points: 2
    Helpful Answer Positive Rating
so for the higher values I should get tantalum capacitors instead of electrolytics
Click to expand...
I'm not sure, if they are much better regarding non-linearity. "Detectable" distortions doesn't necessarily mean, that they are perceptible with usual sound signals.
there isn't a global equalizer in windows
Click to expand...
Depends on the involved sound chip driver. Realtek e.g. has a 10-band equalizer implemented with it's HD PC codecs. But it's a software equalizer consuming processor time.
 
  • Like
Reactions: nabiul

    N

    nabiul

    Points: 2
    Helpful Answer Positive Rating
I was considering those and their more accurate cousin the LM3916 until I found out that my source sells them for $6 each. Even ordering online from digikey would cost something around $2.50 each. A resistor and diode network would be a lot cheaper I think. Still haven't priced the LEDs. I was also considering using a single UV led for each frequency under a tube of fluorescent liquid, then varying the intensity to increase the distance in the liquid the light travels before completely diffusing. But I need to test this out to see how it looks.



As for my audio chip, it's some realtek variant, but the available driver for my motherboard has compatability isses with windows 7. Some service or other constantly crashes while using it and the EQ stops working.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top