TDA2050 parallel. HELP!

Matthewong · 18th January 2006, 03:49 PM

Hi, I've done the parallel of two TDA2050 rather than LM3886, because LM3886 very hard to find in my place and second TDA2050 is cheaper. But I have a problem that I hope you all can help me. My newly made power amp (TDA2050 parallel) suddenly blown the right channel. I only drive a pair of normal 8 ohms speaker and my input voltage is only around 20vdc. Does anyone know what's happened?

lineup · 18th January 2006, 04:01 PM

Quote:

Originally posted by Matthewong
Hi, I've done the parallel of two TDA2050 rather than LM3886, because LM3886 very hard to find in my place and second TDA2050 is cheaper. But I have a problem that I hope you all can help me. My newly made power amp (TDA2050 parallel) suddenly blown the right channel. I only drive a pair of normal 8 ohms speaker and my input voltage is only around 20vdc. Does anyone know what's happened?

did you use Zobel network at output?
1 cap in series with 1 resistor across output

also TDA2050 is sensitive to layout
I hope you have read TDA2050 datasheet closely.

Otherwise TDA2050 and TDA2052 are good chips for the money.

AUDIO POWER AMPLIFIERS ICs (ST)

Quote:

TDA2040
(by SGS-Thomson)

This IC is almost the same as TDA2030, only it doesn't need the protective clamping diodes on the output. It behaves almost like general op-amps. I had a trouble with oscillations when I built this amp, which was caused by wrong PCB design. In my later designs it didn't occur anymore. This amp is fast, at least it's faster than TDA2030, and it's got one decimal order lower THD.
There is a similar IC on the market, also by SGS-Thomson, called TDA2050, which should be identical to the TDA2040 except for higher power.

Page of Cheap Amps

Bazukaz · 18th January 2006, 05:39 PM

Why do you parralel them , if driving a 8 ohm load ?
Maybe case to heat-sink insulation broke down ?

Matthewong · 19th January 2006, 03:10 PM

I've done everything. I just follow the layout given by ST Electronics. And the TDA 2050 has built-in thermal shut down, so it won't be the heatsink problem. I've using it for several hours, but in low volume, when I pushed it to around 11 o'clock, it only lasted for several minutes. What should I do to prevent this problem occur again? I only drive 8 ohms speakers not 4 ohms, if i drive 4 ohms speakers, i think it will blow in just 1 to 2 minutes time. Please help me to solve this problem.

Bazukaz · 19th January 2006, 04:00 PM

Chips usually blow if conneted not properly.Inspect everything , and make sure that you used a mica insulator/thermal paste to heat sink , if it is grounded.
Some chips may not like parallel configurations.If doing so , you must
use 0.1% tolerance feedback components and ~ 0.3 ohm(2w) output resistors after each output.

Did you bypass power supply filtering with 0.1 mkf ceramic caps ?
Are all ground connections soldered to one point ? It is very important to ensure stability.

sakis · 20th January 2006, 10:17 AM

the small caps in parallel with big ones used to decouple or bypass an other armonic for example whats the use puting next to a10.000 mfd one 100nf and one 1nf???? the use is that each cap is responsible for a another armonic .... and also makes the big cap a bit faster ...
try to use a tantalium next to the output of any regulator and you will see very good results no mater if the rectifiers cap is 2200mfd .

Matthewong · 20th January 2006, 12:19 PM

Ok, I have checked all the connection, and I found out the I use zobel network before the output resistor. that's mean there got 2 zobel network in one channel. Will it be the problem? And one more thing, I used 0.1 ohm 5w for output resistor.

Leolabs · 22nd January 2006, 08:03 AM

Hi,I thought LM3886 is easily available in KL.

davidsrsb · 22nd January 2006, 12:41 PM

How are you paralleling them? There is nothing in the data sheet. The opamp approach of wiring unity gain bufers with current sharing resistors on the output does not work with audio amps as they are not unity gain stable

consort_ee_um · 22nd January 2006, 03:22 PM

There is an application note on that concept here:
http://eportal.apexmicrotech.com/mai...em_id=&f_page=
as david said the TDA2050 would have to be unity gain stable which it is not.
I suspect it is more like the National semiconductor application note AN1192 which shows 0.1% resistors. That concept should work with the TDA2050 but if the 2 channels are not well matched destructive currents can flow...

18th January 2006, 03:49 PM	#1
Matthewong diyAudio Member Join Date: Jun 2005 Location: KL	TDA2050 parallel. HELP! Hi, I've done the parallel of two TDA2050 rather than LM3886, because LM3886 very hard to find in my place and second TDA2050 is cheaper. But I have a problem that I hope you all can help me. My newly made power amp (TDA2050 parallel) suddenly blown the right channel. I only drive a pair of normal 8 ohms speaker and my input voltage is only around 20vdc. Does anyone know what's happened?

20th January 2006, 10:17 AM	#6
sakis diyAudio Member Join Date: Jan 2006 Location: nea makri athens greece	small capacitors the small caps in parallel with big ones used to decouple or bypass an other armonic for example whats the use puting next to a10.000 mfd one 100nf and one 1nf???? the use is that each cap is responsible for a another armonic .... and also makes the big cap a bit faster ... try to use a tantalium next to the output of any regulator and you will see very good results no mater if the rectifiers cap is 2200mfd .

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Tda2050	Jpoole908	Vendor's Bazaar	0	1st June 2008 06:34 PM
Replacing TDA2030a with TDA2050 in parallel	hernanstafe	Chip Amps	4	16th March 2007 01:52 AM
TDA2050 parallel. HELP!	Matthewong	Chip Amps	5	9th March 2006 03:47 AM
Tda2050	Mr. Triatic	Solid State	8	26th April 2004 06:00 PM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

18th January 2006, 05:39 PM	#3
Bazukaz diyAudio Member Join Date: Nov 2005 Location: Vilnius	Why do you parralel them , if driving a 8 ohm load ? Maybe case to heat-sink insulation broke down ?

19th January 2006, 03:10 PM	#4
Matthewong diyAudio Member Join Date: Jun 2005 Location: KL	I've done everything. I just follow the layout given by ST Electronics. And the TDA 2050 has built-in thermal shut down, so it won't be the heatsink problem. I've using it for several hours, but in low volume, when I pushed it to around 11 o'clock, it only lasted for several minutes. What should I do to prevent this problem occur again? I only drive 8 ohms speakers not 4 ohms, if i drive 4 ohms speakers, i think it will blow in just 1 to 2 minutes time. Please help me to solve this problem.

19th January 2006, 04:00 PM	#5
Bazukaz diyAudio Member Join Date: Nov 2005 Location: Vilnius	Chips usually blow if conneted not properly.Inspect everything , and make sure that you used a mica insulator/thermal paste to heat sink , if it is grounded. Some chips may not like parallel configurations.If doing so , you must use 0.1% tolerance feedback components and ~ 0.3 ohm(2w) output resistors after each output. Did you bypass power supply filtering with 0.1 mkf ceramic caps ? Are all ground connections soldered to one point ? It is very important to ensure stability.

20th January 2006, 12:19 PM	#7
Matthewong diyAudio Member Join Date: Jun 2005 Location: KL	Ok, I have checked all the connection, and I found out the I use zobel network before the output resistor. that's mean there got 2 zobel network in one channel. Will it be the problem? And one more thing, I used 0.1 ohm 5w for output resistor.

22nd January 2006, 08:03 AM	#8
Leolabs diyAudio Member Join Date: Dec 2004 Location: Bukit Mertajam	Hi,I thought LM3886 is easily available in KL.

22nd January 2006, 12:41 PM	#9
davidsrsb diyAudio Member Join Date: Dec 2005 Location: Kuala Lumpur	How are you paralleling them? There is nothing in the data sheet. The opamp approach of wiring unity gain bufers with current sharing resistors on the output does not work with audio amps as they are not unity gain stable

22nd January 2006, 03:22 PM	#10
consort_ee_um diyAudio Member Join Date: May 2004 Location: Hertfordshire	There is an application note on that concept here: http://eportal.apexmicrotech.com/mai...em_id=&f_page= as david said the TDA2050 would have to be unity gain stable which it is not. I suspect it is more like the National semiconductor application note AN1192 which shows 0.1% resistors. That concept should work with the TDA2050 but if the 2 channels are not well matched destructive currents can flow...

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)