Problem building regards weird noise (TDA7240)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I was building amplifier based on chip TDA 7240A, but encounter a weird noise.
It only happen when i connect to my music source (netbook), but doesn't happen that when i directly connect to my speaker unit (without amplifier).

It notice when i'm not using AC (use battery power) on my netbook, that noise doesn't happen. This also same for my previously build discrete amplifier.

But if i use any commercial brought amplifier, no such problem. So i guess need to incorporate something to make this noise disappear ?
 
hm.... i decided to record the noise using my netbook's in-built mic, so its a bit not good. Attached in following link :

Anyway, at the first 4 second, it was the ambient noise (the amplifier is turn off) and was raining (doesn't sounds like raining is it :scratch: my mic sucks).
After the first 4 second, i turn on the mic amplifier, with AC powered netbook as input. (without any music on, just plug in), the noise is obviously heard, but the pitch is different in real world (damn my lousy mic) Is it oscillation ?

About at 12 second, I plug out the AC charger, thus on battery powered only. That makes the noise goes away, and speaker is nice with just a little hum (acceptable and can't heard by the mic).

At 22 second, I plug in the AC charger into netbook again, thus the noise came out again. About at 28 second, I remove the input TRS jack, thus no noise heard. At 30 second, I plug in the input jack to netbook, thus making noise again.

Any advise how to remove such noise from my amplifier (discrete amp noise is different)?
 
a-netz.de - Audio Amplifier with TDA7240
I have build based on this link, although my 470uF in the picture is 2 x 220uF, others mostly the same. (I use point to point solder technique, which uses point by point proto board)
THIS is the manufacturer PDF : http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/DATASHEET/CD00010656.pdf
As for the power supply, i'm using a computer power supply (capable to supply to 14A of 12V)


This is a very common problem when plugging computers into audio systems. The most effective cure is to use an isolating transformer (1:1) - in the audio output feed from the computer.
I think this is the answer i'm searching for, but I don't know much about isolating transformer, mind explain slightly ? What is the pro and con using this method for input ? Does commercial Hi-Fi or good equipment also include this in the design ? (since i have no problem using those)
 
I think I got the right schematic. The isolation transformer couples AC (as in audio) but not DC. This is a way to isolate your equipment. Cost can be considerable, but it dosent have to be. Point to point ---not a problem, Lots of fine equipment is built that way. Mabe someone can suggest a source for the transformer.
 

Attachments

  • audioamp_circuit_small.png
    audioamp_circuit_small.png
    20.3 KB · Views: 170
hm.. then what could my problem ? I'm guessing the AC charger adapter of my netbook connected tend to 'suck' current from my power supply, thus making 'motorboating' sound ? (don't have any other idea why it will cause power supply to go crazy when AC charger of netbook is connected...)

I have a look into purchasing of isolating transformer, it doesn't looks so good... and most of it having frequency limit of 400-4kHz, which different alot with audio (20Hz-20kHz)..... and those can allow those audio frequency and expensive and build for high powered (>50W). A bit too much and over, both in function, quality and COST.
Any technique to isolate the input ground and amplifier ground to prevent this matter, at the same time allow the music into amplifier ?

Edit : I suddenly think of an idea, can i attach a capacitor in series with input ground to amplifier ground ? this kinda isolate DC between them. Its may not be an solution, but will it danger to my system/component ? (wanted to test)
If so, what value should i use to permit at least 20Hz or audio is allow to pass ?
 
Last edited:
tried with input ground disconnected, the 'motorboat' sound still persist if i'm using AC charger.

Also measure using my DMM, but seems nothing much except for stable voltage drop due to large load(happen when i tune up the music a louder)

Tried connect directly from input connection to speaker unit (bypass amplifier), the sound doesn't change even if i connect to AC charger or using battery.

Also experiment using either input's hot or cold only, the 'motorboating' sound still there, but different pitch.
What could possibly induce this noise ?
 
hm... now i tried to connect a high capacitor as reservoir (2.2mF) is quite big for this amp i think, but overkill won't be problem.
The problem is still persist, but its ONLY a little bit help it, ALMOST no different from the 1st record. (but slightly better, just SLIGHTLY)



This time, i'm not making it standby, but playing music.
About after 8 second, I connect the AC charger. Thus the heavy ''''music'''' came out !
about at 22 second, I disconnect it, thus nice music again. After few seconds, I unplug the input, there is a little buzz (hard to listen from the recording), it was worst when i didn't use the reservoir capacitor. Any idea to solve these problem ?
 
hm... just found something..... this chip don't have ways to adjust gain, and itself is very high(40dB)
Is there anyway to reduce the gain, thus i can make the input more louder, thus minimize pick up noise(hiss) ?
Does any different if i put attenuation VR at the board before input ? (compare to controlling gain for this purpose)
 
An attenuator pot at the input will help if there is still room to turn the level on your netbook higher, because the noise isn't coming from the amp, it comes from the netbook itself. It sounds like your netbook has a noisy power adapter that makes its way into the audio circuits. Mine as well... The cure to that is usually buying a high quality external soundcard.
 
haha, i have a nice external soundcard, but doesn't dare to try X.x

But i'm making the amplifier+speaker to give as a gift for my friend, so i hope to thoroughly solve this problem.
It seems that the AC adapter doesn't DIRECTLY input the noise, because when I connect direct to speaker unit, it doesn't cause problem. It probably because of something that in amplifier induce into it by affect of AC charger.... still thinking, probably oscillation, which won't happen alone with speaker only. (only happen with presence of amplifier)
 
It's an ordinary ground loop, I bet. You don't happen to have amplifier ground connected to chassis ground, which in turn is connected to security earth? Even with a nominally floating amplifier, coupling over the mains transformer may be big enough to cause audible problems.

With as few pins as this chip has, you probably have no chance of setting gain externally.
 
Test with a battery powered MP3 player.
If no noise persists during test then follow this:
Purchase a 6 amp 90w capacity 15vdc laptop power pack and use it for your TDA7240. They're inexpensive. Avoid all models with a 3 prong cord (cannot work for car chip amplifiers plus computer source)!!! Use only a model with a 2 prong cord. That will give your laptop the opportunity to be the ground, without fighting over it.

Here is one possible example.
AC Adapter PA2521U-3ACA 15V 6A 90W for Toshiba Laptop | eBay
Please notice the 2 prong mains cord!!
There is also an authentic Toshiba part with 15vdc 6a available on the used market, and there's the possibility that some ordinary broken/retired laptops may have serviceable power packs.


Your input load may be missing. That can make a terrible noise when source device isn't connected. I couldn't find the circuit equivalent schematic, so I'm not sure if the input load is inbuilt or not. However, it seems to need a 10k resistor between pins 1 and 3. It also seems that the cap at pin 1 is far too small--I would attempt 22uF//220uf//4.7nF or something similar. You'd like your roll-off about an octave or more below speaker capacity, to avoid any bass blocking within the frequency response capacity of your speaker. The datasheet example supports 3 inch speakers. :) So, double-check me on that--I haven't used this particular car chip.


But, I do use a car chip with my computer. It has been running on the laptop power pack (SMPS) for 3 years. Of the many car chips that I tested, only LA4628 (one of the best performers) is able to overload a 6A laptop power pack.

The old model high gain philips when not bridged is able to have supplementary feedback circuitry added to 2 of its 4 channels and can achieve the highest quality performance for music playback fidelity on those two channels at up to 6 watts output power. The other two channels (incompatible with external circuitry additions) may be driven by a single drv134 (or a good buffer chip with some resistors added for the job) and used to power an additional woofer to 20 watts.
That same chip is awfully dull and noisy when auto-bridged (simple bridge) per the datasheet. It is true of most car amplifiers. Auto-bridge just isn't that good. Balanced line drive feeding quality parts is a better performance, albeit a bit more soldering.

P.S.
Car alternators run the voltage up to 14.85 volts DC when the engine is on. A 15v dc power pack is suitable for any reasonably designed car amplifier chip; unfortunately, some automotive products are horribly dangerous, like LED refit bulbs, that cause fire if operated beyond 12v, or Tripath car chip that burns up if operated beyond 13.9v. Clearly, not all automotive products were designed by car owners. This type of thing usually uses power diode drop to bring down the voltage to safe levels.
However, your chip can tolerate the Real full automotive voltage range, according to its datasheet.
Its always good to double-check both datasheets and real product sample too.
 
It's an ordinary ground loop, I bet. . . .
Almost certainly.
Its a car chip with no car. :)

With as few pins as this chip has, you probably have no chance of setting gain externally.
That's right.
Fortunately, a computer can push that amplifier almost all the way to potential even without increased gain. Unfortunately, the majority of computers will clip themselves at beyond half blast at the sound card. It may be worth a try to make a small preamp for the computer.

A fun little Jfet op-amp or the MooseFet project may be fun to try out. I like my MooseFet, but will admit that its a lossy triode sound processor specializing in straightening out the harmonics of MP4, Itune, HD Radio, AAC file types, and unnecessary harmonic distortion for MP3 or CD. But, not getting a headache when listening to Mp4 and HD radio is really just priceless. The gain is 6.

See the Class A NE5534 project at Decibel Dungeon: Building a buffered Gainclone chip amp.
This is NOT as complex as it looks--what you're seeing is both the preamp and its power supply built onto the same board. Have a look at those two sections individually and then it begins to look simple.

And heck, I even like JRC4580's, two of them run with their outputs paralleled (stereo parallel preamp), because it sounds expensive and costs only a few cents. :) That is same as the high current output jack of M-audio Revolution 5.1, a sound card that doesn't clip itself. I don't see any reason why the same chips can't be added to any sound card as an external preamp. JRC4560's work identically. JRC4558's do not.
 
hm... then we can mostly conclude that because of it design for CAR AUDIO, it is very susceptible to groundloop (because there are hardly groundloop in car ?) I was going to try the method called "loop breaker circuit", see if it fits ?
something like that, which sounds good ? (isolation transformer is better, but very costly if sufficient for audio bandwidth )

I'm using 22uF for pin 1 capacitor (feedback), so shouldn't have problem. Cannot adjust gain means that the internal hiss will be loud all the time ? I don't like that....

hm..... Thinking of using attenuate VR instead of gain (we can't adjust gain...), thus minimize the hiss and buzz from input only... but better than none.

Putting 10k load between pin 1 and 3 (feedback & input pin) will get rid of load missing noise ? (its annoying) I don't have exact 10k, could I use 12k, 13k, etc ?
I personally don't like batteries, cause it will degrade over small period of time, compare to power supply. Using for house speaker.
 
Your input load may be missing. That can make a terrible noise when source device isn't connected. I couldn't find the circuit equivalent schematic, so I'm not sure if the input load is inbuilt or not. However, it seems to need a 10k resistor between pins 1 and 3. It also seems that the cap at pin 1 is far too small--I would attempt 22uF//220uf//4.7nF or something similar. You'd like your roll-off about an octave or more below speaker capacity, to avoid any bass blocking within the frequency response capacity of your speaker.
tried using 11k resistor across pin 1 and 3, the 'input load missing' noise really disappear, thanks alot ! but still alot of other noise. Could explain to me how this is work ?
About the bass blocking, maybe it will kill my speaker, so i prefer less bass, haha
 
yay, I have achieve improvement of the noise, recorded and attached link.
The configuration is using a attenuation VR (volume control) between computer output and amplifier input.
Before using this, my computer must configured volume to 5% for normal audio, 15%-20% for booming. above those will deaf my ear.
But now I'm using 100% of computer volume, and control using VR, the noise is significantly reduce ! (difference between 5% and 100% noise)
The record sound is connected to AC charger all the time, just changing VR knob.
With music, its not obvious but audible, obvious when quiet transition. (heard as microwave oven sound) at high volume, the noise doesn't hears like motorboating with music on, just like distortion.

Is the control of knob with no music playing, but connected to computer. with low volume and high volume, the noise is different pattern and sound.

Although it has been better, hope to eliminate it (from root cause).

Extra : using safety ground direct connect to input ground doesn't change anything.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.