Easy way to test if TDA 7294 chip is dead
Hello, being bit noobish and all, i need your guys help !
I have recently bought Logitec Z 5500 sub and im trying to fix it ! I saw that on heatsink one amp is missing, and i wanted to check if others work, but i dont know how to see if they are dead or problem is somewhere else.
Any easy tactics i could test it ?
Hello, being bit noobish and all, i need your guys help !
I have recently bought Logitec Z 5500 sub and im trying to fix it ! I saw that on heatsink one amp is missing, and i wanted to check if others work, but i dont know how to see if they are dead or problem is somewhere else.
Any easy tactics i could test it ?
Last edited:
Test it like you'd test any other questionable amplifier. Connect a 100 watt incandescent light bulb in series with the mains and power it up. If the light stays bright, it's drawing too much current. If it flashes and then goes dim, good sign. If it flashes bright and dim, it's oscillating.
Once that test is passed, check for output voltage offset. You're looking for 25-30 mV or less - anything over 1 volt is a major fault.
If it passes both tests, then hook it up to a speaker and source and listen.
This procedure is considered basic of basics and is discussed repeatedly in this forum. A simple browse of the posts would have revealed that.
Once that test is passed, check for output voltage offset. You're looking for 25-30 mV or less - anything over 1 volt is a major fault.
If it passes both tests, then hook it up to a speaker and source and listen.
This procedure is considered basic of basics and is discussed repeatedly in this forum. A simple browse of the posts would have revealed that.
Figure #1 in the data sheet is called 'test circuit'. Its a basic configuration to check the device is operational.
Only other first line methods are a quick visual to make sure the device hasn't split apart (it happens) and that there are no shorts between supply pins and supply pins and output pin.
Beyond that and you have to build/breadboard a functional circuit to check it out.
Only other first line methods are a quick visual to make sure the device hasn't split apart (it happens) and that there are no shorts between supply pins and supply pins and output pin.
Beyond that and you have to build/breadboard a functional circuit to check it out.
We all know about it, problem is your question was poorly written.
"Chips" containing, say, 20-30 to more than 100 internal components can not be tested "as a component" which is what you asked, but "as a system", what they really are.
You simply don´t have access to internal components for measurement, testing or voltage readings, only to chip external pins, so you test them as part of larger system: either in the original PCB or soldered/breadboarded into a datasheet suggested "test circuit" , (which by the way is not designed for troubleshooting but to check prformance), apply needed voltages and basically "make it work".
Meaning:
* it should not blow fuses,overheat power supply or itself, nor crack/smoke/explode. The series bulb limiter is a good idea to at least reduce the mini Nuke effects.
* it should show accepted, very low DC offset voltage at idle
* if fed audio, it should drive a speaker.
* it should not buzz/hiss/hum/motorboat with input shorted, although some of this depends more on layout and supply quality than chip itself.
In a nutshell: if it does not match the minimum functions described above, does not pass Audio or it´s very dirty/distorted/chopped but a new chip behaves as expected, *then* you know theoldone was faulty.
Now if you expect to "measure something" with the multimeter, with chip in your hands, to decide whether to order a new one or not, sorry, it does not work that way.
Now if you expect to "measure something" with the multimeter, with chip in your hands, to decide whether to order a new one or not, sorry, it does not work that way.
This is what i was kinda hoping for
Ah well, soo, basically i can take measurement inside a system, aka on main PCB with all the other stuff ?
And sry for probably asking a very noob question, but im kinda doomed to fix this system all by myself and, im going to be honest, i have no idea wtf i am doing and where to begin with.
There's always the shotgun approach. If by amp missing you mean one 7294 is ripped out, just buy enough to replace them all and go from there. I never understood the 7294 as the 93 seems to be the same part with slightly bigger output transistors. Anyway in your case the difference is probably nil except they are a bit more rugged, so may as well get a some 93's.
Well, main PCB houses chips, 2xTDA in bridge for sub (which in my case, one is missing, prob prev owner removed it or something) and what i think is 5xTDA for other channels + one small for controlling ? I can post pictures later.
Then there is a smaller, control PCB which has opamps (if that they are called ?) for 3x channels and some regulators. Power supply is torus trans which reads 26.4 - 0 - 26.4 and 14.5 - 0
In that case then I would do as JMFahey suggests and use a bulb tester. Do not have speakers connected at this stage but power it up and check the supply voltages and the DC voltage on the output pins of all the TDA's. I'm assuming the outputs would all be zero volts although the missing chip (if part of a bridge) could cause its other half to give a misleading reading.
First thing is to see if it powers up with the bulb tester and take it from there.
First thing is to see if it powers up with the bulb tester and take it from there.
I got a exactly same main PCB board from a friend's sub, with all TDA intact, but i dont know if they are busted or not, i was thinking to take one from spare board and put in mine, because friend of mine tested it and said power tru main unit is ok. Unfortunately, he left the next day, and will be while till he returns, so thats why im trying this on my own
I will do the bulb test tomorrow and post results
Thank you guys 
Ok, so by mains you mean Vs + and Vs - (pin 13 and 15) on the chip ? I have attached a picture of chip pinouts Imgur: The most awesome images on the Internet
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Now you know the light bulb test. It is a very simple procedure that tells you a lot, fast. I have a rig permanently assembled and a small assortment of 100 to 300 incandescent watt bulbs to use. I always use it to power up unknown equipment, or equipment that hasn't been powered up for a long time.
You mentioned resistors in series with the rails. This is also a good idea and will tell you a lot, fast. Value depends on expected current draw. You want the resistor to drop negligible voltage with the expected current. That way, when you power it up, you measure the voltage after the resistors. If it has too much voltage drop, you have a clue. If one side or the other or both drop too much voltage, you have clues. If you use a cheap underpowered series resistor and it burns up (essentially a fuse), then you not only have a clue but you might have also saved some components.
Finally, if you're going to be testing a TDA chip out of the circuit, be aware that they have a tendency to burn up if some of the connections are loose. They are definitely best tested in-circuit.
You mentioned resistors in series with the rails. This is also a good idea and will tell you a lot, fast. Value depends on expected current draw. You want the resistor to drop negligible voltage with the expected current. That way, when you power it up, you measure the voltage after the resistors. If it has too much voltage drop, you have a clue. If one side or the other or both drop too much voltage, you have clues. If you use a cheap underpowered series resistor and it burns up (essentially a fuse), then you not only have a clue but you might have also saved some components.
Finally, if you're going to be testing a TDA chip out of the circuit, be aware that they have a tendency to burn up if some of the connections are loose. They are definitely best tested in-circuit.
Power up without speaker connected and look for DC offset on output.
If no significant DC offset then try music and look on scope.
If it looks ok connect a speaker.
I would be a little careful with putting the IC into a test circuit with no attention to layout. The chip could oscillate and give a false impression in a bad layout.
The datasheet should show a test circuit. This can be configured in a point to point wiring configuration keeping all wires as short as possible and observing all normal amplifier precautions.
The datasheet should show a test circuit. This can be configured in a point to point wiring configuration keeping all wires as short as possible and observing all normal amplifier precautions.
Hey guys, i finally made the tester, works great and all.
Now, when i connected sub to the tester and turned it on, light bulb started bright and dimmed down to nothing over course of, hmm, half a second. Is that ok ? There were no flashing or oscillations what so ever.
P.S. Whoever told me that desoldering these chips would be a nightmare, he was totally right !
Now, when i connected sub to the tester and turned it on, light bulb started bright and dimmed down to nothing over course of, hmm, half a second. Is that ok ? There were no flashing or oscillations what so ever.
P.S. Whoever told me that desoldering these chips would be a nightmare, he was totally right !
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.