diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Chip Amps (http://www.diyaudio.com/forums/chip-amps/)
-   -   TDA7386 Car Amp (http://www.diyaudio.com/forums/chip-amps/128225-tda7386-car-amp.html)

Redshift187 16th August 2008 07:51 PM

TDA7386 Car Amp
 
2 Attachment(s)
Hi all. I've been reading about chip amps and drooling. :) I love music, and making fairly simple electronics. For my first try, I'm making an amp for my fiancee, since we only have a head unit with pre outs available for her car. I decided on the TDA7386 because it seems to be the highest output 4 channel 12V single supply chip amp available. I have a few questions about the attached circuit diagram (the only one I could find besides the one in the datasheet):

1. Am I correct to assume ceramic caps are used for C7, C10, C11 and C12?

2. Is a connection to power ground from signal ground (pin 10) really necessary? I'm worried about ground loops, although I understand that R9 would hopefully take care of that.

3. I would like to change the DC blocking caps on 2 of the inputs (C1 and C4) to raise the lower frequency cutoff. It states in the datasheet that 0.1uF has a lower frequency cutoff around 16Hz. I would like to raise that to somewhere between 50 and 80Hz, but I'm not sure what value the caps would then be.

4. Is the input impedance (5700 ohm?) suitable for my head unit? It has 8V 55 ohm outputs.

pacificblue 16th August 2008 09:11 PM

Re: TDA7386 Car Amp
 
Quote:

Originally posted by Redshift187
I decided on the TDA7386 because it seems to be the highest output 4 channel 12V single supply chip amp available.
How about the TDA7560? It is pin compatible, so you can use the same schematic and PCB and get 4x45W. Not that much more power, but significantly better distortion levels.

Quote:

Originally posted by Redshift187
1. Am I correct to assume ceramic caps are used for C7, C10, C11 and C12?
Should do the job.

Quote:

Originally posted by Redshift187
2. Is a connection to power ground from signal ground (pin 10) really necessary? I'm worried about ground loops, although I understand that R9 would hopefully take care of that.
Yes to both.

Quote:

Originally posted by Redshift187
3. I would like to change the DC blocking caps on 2 of the inputs (C1 and C4) to raise the lower frequency cutoff. It states in the datasheet that 0.1uF has a lower frequency cutoff around 16Hz. I would like to raise that to somewhere between 50 and 80Hz, but I'm not sure what value the caps would then be.
The formula is f=1/(2*PI*R*C). With f=16Hz and C=0,0000001F that gives R~100kOhms. That means you will need 33nF for ~50Hz and 22nF for ~80Hz.

Quote:

Originally posted by Redshift187
4. Is the input impedance (5700 ohm?) suitable for my head unit? It has 8V 55 ohm outputs.
Will work somehow. R1 through R8 however seem to be there for a connection to hi level outputs. They reduce the input level by a factor of ~5,75. Try to replace them with potentiometers 10k to 25k log. That brings the input impedance to a more practical level and gives you the possibility to adjust the sensitivity per channel like in any commercial amp.

Eva 16th August 2008 09:22 PM

If simple 6" car speakers (or smaller) are going to be used, a highpass filter at 50Hz or higher is recommended. You may find a good compromise by trial and error.

Redshift187 17th August 2008 12:11 AM

Thanks so much guys. I will definitely use the TDA7560.

I'm using 4.5" coaxial front speakers,and 6.5" woofers in the rear. I'm not sure yet what frequency I'll high pass them at, maybe try 40Hz for the rear and 120Hz for the front. Which should be ~40nF and ~13nF respectively. I'll have to check the speaker datasheets to see what they can handle.

Redshift187 30th August 2008 11:52 PM

2 Attachment(s)
Well, the first step is done, I've soldered all components p2p. I used a TDA7560. Eleven in total. From the schematic above, I removed R1-R8, R11. I used an 8R2 for R9 because that's what I could get. For PS caps I followed the example circuit in the chip documentation and used a 2200uF and a 0.1uF. Input caps are 0.01uF for the front speakers, giving a cutoff around 150hz and 0.033uF for the rear speakers giving around 50hz cutoff. Now I think I have about 19 wires to solder for power and input/output.

Redshift187 30th August 2008 11:53 PM

2 Attachment(s)
Another shot of my ugly soldering :)

Redshift187 31st August 2008 07:21 PM

Tested and it works, although it's pretty quiet with the lowly 11V my test power supply puts out. DC offset is < 50mV. Can't wait to try it out in the Jeep.

MartyM 3rd September 2008 03:08 PM

Hi there-can I make a suggestion or two?

1. You will definitely have to be prepared for bad ground loop noise on any DIY car audio design. You have to be prepared for to noise canceling circuits or ground isolation (ground loop isolators, for example).

2. It would be really beneficial to build it on a good prototyping board, and put it inside some type of plastic or metal enclosure so it's reliable and you don't have to pull over on the side of the road to troubleshoot it!

3. Add an input or gain control, as applicable, for each channel or the pair.

Sounds like a nice starter project-have fun. :snoopy:

Redshift187 3rd September 2008 06:02 PM

Hi MartyM

I have an 8.2 ohm 2W resistor between the signal grounds and the power ground.

Too late to build it on a protoboard. But it works :)

It's going in a small Hammond aluminum enclosure with lots of copper (it was cheap) heatsinking on the outside. I'll be putting a strain relief on the cables so they don't cause pins to move and short.

Redshift187 9th September 2008 04:59 PM

So I failed in my first attempt. First, I noticed that the schematic I posted has a mistake, the caps for SVR and AC-GND are switched. Not a big deal, but I had pop on turn on until I switched them. Also, after I swtiched them my DC offset dropped from ~40mV to unmeasureable.

That doesn't really matter though, as I'm pretty sure I toasted the IC. I accidentally hooked V+ up and ground to the standby pin. No big deal, except I forgot to hook ground up to ground. Got a nasty squeal for a second before I turned off the power. Now my output sounds like mute is enabled (sound, but barely audible when source is on max), and I've tried everything I can think of. I followed the reference schematic with the following exceptions: Mute pin is tied directly to V+ and the input caps are smaller values for some high-pass filtering. I checked voltages, I'm getting 11.5V at the chip. I tried putting a 47k resistor between the Mute pin and V+. I tried completely bypassing the input caps. I tried completely bypassing the input RCAs.

At this point I'm ready to scrap the project and start over, using MartyM's suggestion of using protoboard.


All times are GMT. The time now is 12:19 PM.


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