TDA2003 hum ¿PSSR problem?

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi to everyone in this forum.

I am building a TDA2003 power amp to drive a small 8 ohms speaker. I had designed and built several amps in the past with that chip and with others, but the problem I am now facing is some rather new for me.

The amp is for a railroad train model and is powered by a typical linear power supply (127/12 VAC transformer, full wave bridge, filter cap, and 7805 and 7812 regulators). The output from the 7812 feeds the supply voltage of the TDA2003 and the amp works as expected. However I tried to attach to that same +12 volt output from the 7812 a pair of 12-volt incandescent 5-watt lamps (the kind used for directional signaling ligths in the cars). The lamps turn on and bright ok, but a horrible noise and hum is output to the speaker. I had tried any kind of filtering heavily the power supply, increasing the transformer current capacity (of course the offending noise and hum is proportional to the overall current of the load that is placed in the same +12 volts line that powers the TDA2003), adding chokes to filtering the +12 volts line to the lamps and the +12 volts line to the TDA2003, but with no success.

I know that the most simple way should be to use smaller current lamps or better use super-bright LEDs and my problem will be concluded.

However, I am not in that direction and I am wondering how could I get the audio amp working ok with the lamps described above (I think there should be a way). My main suspect is about the PSSR of the TDA2003. I am working now for testing the LM386, and LM1875 that have better PSSR to see if the offending noise is lower. Are there any other parameters that could affect the TDA2003 in such way? Any ideas?

Thank you very much for any help.
Hi sync40

No one answered your question!
probably a dead thread but here goes anyway.

I have found with this amp (and others) that you need to keep the power and the signal grounds seperate or you get hum. make sure that the earth connection to the signal parts of the circuit is connected via a long thin track/wire from the main ground which is carrying current. Check the data sheet pcb layouts and you'll see some weird gnd tracks used to fix this problem.
My TDA2002 (previous version if the chip) is now quiet after I fixed this.

Regards Philip
Thank you very much for your reply.

Since I redesigned the PCB with separate traces the noise almost goes out, but it is still audible. I almost killed all of the noise by using a very large capacitor for filtering the secondary. However, I have some problems now with the logic not resetting (because of the low discharge of that capacitor). Another partial solution is to increment the AC voltage of the secondary, but now the regulators run very hot.

So, I am now planning to use maybe a SM2211, LM4871, or TPA0211, running from the +5 volts supply. I hope that any of these amplifiers could provide enough power like the TDA2003 at 8 ohms.
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.