Yesterday night I did all the mods to be according to the schematic posted earlier also did check all connections and solder Q and replaced the few track connections with hard wire to be consistent with the rest of the board, everything looks clean.
Still no cigar, mute switch works 0mV at output and when switched ON 11.5V at outputs (BTL config), did this without load and with a 100R load. Also tested the cap at pin 12 the way it’s used on the similar TDA1553Q, no change there too.
One thing that called my attention on the chip is that the signal gnd pin (2) as shown on the block diagram (page 3 spec sheet) looks to be connected with input ref voltage (pin 12) but it shows 10K DCR on the multimeter.
Attached is the TDA1557Q spec sheet and the Philips AUDIO AMPLIFIER IC’s DESIGNER’S GUIDE that describes most all chipamps from them, nice for selecting for specific use.
Antonio
Still no cigar, mute switch works 0mV at output and when switched ON 11.5V at outputs (BTL config), did this without load and with a 100R load. Also tested the cap at pin 12 the way it’s used on the similar TDA1553Q, no change there too.
One thing that called my attention on the chip is that the signal gnd pin (2) as shown on the block diagram (page 3 spec sheet) looks to be connected with input ref voltage (pin 12) but it shows 10K DCR on the multimeter.
Attached is the TDA1557Q spec sheet and the Philips AUDIO AMPLIFIER IC’s DESIGNER’S GUIDE that describes most all chipamps from them, nice for selecting for specific use.
Antonio
Last edited:
The drama continues ... always possible you bought a batch of dud chips - you seem to have all the datasheet material and experience to sort these issues out - how many chips have you got to play with, 
?
Thanks for putting up the Philips selection guide, a whole line of chips I was unfamiliar with - the one that intrigues me, that I would immediately go for, is the TDA1514A. High power, protection on board, and the bit that really interests me, a pin that is ground! Normally opamps don't give you this, and this gives one scope for trying decoupling ideas, if it truly links to the internal ground of the circuit.
We have a local parts supplier which I would be fairly confident stocks genuine, or as close to as possible, parts. The TDA1514A is significantly cheaper than TDA1557Q - interesting ...
?Thanks for putting up the Philips selection guide, a whole line of chips I was unfamiliar with - the one that intrigues me, that I would immediately go for, is the TDA1514A. High power, protection on board, and the bit that really interests me, a pin that is ground! Normally opamps don't give you this, and this gives one scope for trying decoupling ideas, if it truly links to the internal ground of the circuit.
We have a local parts supplier which I would be fairly confident stocks genuine, or as close to as possible, parts. The TDA1514A is significantly cheaper than TDA1557Q - interesting ...
Last edited:
I declare the chip dead for the second time, frustrating but this happens when you deal with fakes.
Next, I got a sister chip TDA1553Q which has markings more on the side of originals besides having a more usable gain (26dB) instead of the 1557 that has 46dB, otherwise specs are the same.
I will try a transplant on the same board and hope this time around will work.
Antonio
Next, I got a sister chip TDA1553Q which has markings more on the side of originals besides having a more usable gain (26dB) instead of the 1557 that has 46dB, otherwise specs are the same.
I will try a transplant on the same board and hope this time around will work.
Antonio
Frank,
Thanks for your offer but I have a supplier (not very reliable) which is not a close friend but a good guy who does its best to serve and is close to home.
Looking to other block diagrams I learn that the Input Reference Voltage has an internal cap and the negative of that cap goes to Signal Gnd so that´s the reason the Input Ref Voltage has no continuity with SG.
I`ll have a look to that 1514 you are eyeing, you should try and assemble one of those, time permitting, just for the fun of it.
Antonio
Thanks for your offer but I have a supplier (not very reliable) which is not a close friend but a good guy who does its best to serve and is close to home.
Looking to other block diagrams I learn that the Input Reference Voltage has an internal cap and the negative of that cap goes to Signal Gnd so that´s the reason the Input Ref Voltage has no continuity with SG.
I`ll have a look to that 1514 you are eyeing, you should try and assemble one of those, time permitting, just for the fun of it.
Antonio
Good explanation Andrew nice seeing that in detail and thanks for the input that shows the benefit of this type of supply filters.
I am with Abrax on cored chokes, size matters too besides air core needs a lot of copper for the same inductance and sometimes the ¨r¨ gets too high for practical use. Myself use gapped EI cores as well as ferrite reel type depending on application, in the case of the supply for the project on this thread I used a ferrite reel type but ended with a low inductance since I didn’t have the correct wire gage.
Once I did a an air coil for a woofer xover using heavy wire, something more than 3 mm dia which weighed about 2 kg each, still have them some place in the work shop but the end result was that I replace them with ferrite cored ones with better overall response from the speaker, thru the years I have continue to use those ferrite chokes.
Antonio
I fail to see how can You kill 2 ics in a row
I don't really see allso, why go with the tweaked stuff in the first place.
I'd still recommend to build the basic stripped version point to point, and try it from a suitable battery. Then add the other stuff.
I do not think this will ever work.
So far not a lot of joy You had with these chips,
and i yet fail to see where it goes wrong both times.
while You could, power them on from a battery, without external components.
leave it in mute mode, and check for the outputs. Supposedly in standby it should read zero. Then add the 100 nf cap from V+ to power ground, and short inputs to ground.
Then flick it from standby to play. And check for the outputs, still should be zero, or something verry close to it.
then back to the inputs, solder some 20-47 Kohm frominout t signal ground,
add a foil ca in series and test with a source. It should play, and not oscillate.
The fancy stuff, main's psu, huge filter caps, things soldered between the pins, exotic junk, all that can wait. I'm not sure if it is not You making a mistake with both attempts on this chip.
there is simply no need for a pcb with an amp this un-complicated.
otherwise, if You do wish to use one, i can link a source that sells ected pcb-s, along with the parts needed to build the amp. including the chip.
all you need to drill some holes, and solder the stuff.
while it is quite uncommon, but some people manage to overheat the IC while they solder the stuff. That is one more thing i can think of.
I don't really see allso, why go with the tweaked stuff in the first place.
I'd still recommend to build the basic stripped version point to point, and try it from a suitable battery. Then add the other stuff.
I do not think this will ever work.
So far not a lot of joy You had with these chips,
and i yet fail to see where it goes wrong both times.
while You could, power them on from a battery, without external components.
leave it in mute mode, and check for the outputs. Supposedly in standby it should read zero. Then add the 100 nf cap from V+ to power ground, and short inputs to ground.
Then flick it from standby to play. And check for the outputs, still should be zero, or something verry close to it.
then back to the inputs, solder some 20-47 Kohm frominout t signal ground,
add a foil ca in series and test with a source. It should play, and not oscillate.
The fancy stuff, main's psu, huge filter caps, things soldered between the pins, exotic junk, all that can wait. I'm not sure if it is not You making a mistake with both attempts on this chip.
there is simply no need for a pcb with an amp this un-complicated.
otherwise, if You do wish to use one, i can link a source that sells ected pcb-s, along with the parts needed to build the amp. including the chip.
all you need to drill some holes, and solder the stuff.
while it is quite uncommon, but some people manage to overheat the IC while they solder the stuff. That is one more thing i can think of.
You didn’t know about my abilities, difficult as it may seem never managed this before! So there for my record !!!
Wouldn’t call this a fancy supply at least on my book, the chip was always turn ON with the mute engaged in the other hand supply voltage is quite lowish at 14V with no load and 12.8V with load well within spec and design parameters for the IC. But there, we will never know if these fake 1557 were shot or was a mistake from my part such as overheating some pins evens thou I use a controlled temp station for all the work and follow recommendations on time.
I will think about your battery proposal, seems a cumbersome thing to do on a loose chip and as much time consuming as transplant and believe me don’t care much to shot a 3rd one but a valid proposal nonetheless. As I said before this new chip looks original as markings goes even if we know that forgery this days are quite sophisticate.
I will continue until we have a working amp…
Antonio
Wouldn’t call this a fancy supply at least on my book, the chip was always turn ON with the mute engaged in the other hand supply voltage is quite lowish at 14V with no load and 12.8V with load well within spec and design parameters for the IC. But there, we will never know if these fake 1557 were shot or was a mistake from my part such as overheating some pins evens thou I use a controlled temp station for all the work and follow recommendations on time.
I will think about your battery proposal, seems a cumbersome thing to do on a loose chip and as much time consuming as transplant and believe me don’t care much to shot a 3rd one but a valid proposal nonetheless. As I said before this new chip looks original as markings goes even if we know that forgery this days are quite sophisticate.
I will continue until we have a working amp…
Antonio
fine by me then
to be honest the "cruel" way some kids do it here is just apply power from pc psu, connect speakers and source from soundcard.
read: no external components at all, and they work.
yeah, sound is horrid, but gotta hand it to them, it is cheap, better then anything that pocketmoney can buy, and basicly it works
just gotta first wait till psu manages to stabilise rails, then the fun can start.
ghetto amp.
i did allso mention i think i never come by any real chip, just fakes.
dunno how on earth it did not work properly for You, mostly the oscillation part is interesting.
sometimes they can "motorboat" , catch DC fan inducted noise, radio station,
but high freq. oscillation, i did never have.
and i have done some clumsy builds with it to be honest.
to be honest the "cruel" way some kids do it here is just apply power from pc psu, connect speakers and source from soundcard.
read: no external components at all, and they work.
yeah, sound is horrid, but gotta hand it to them, it is cheap, better then anything that pocketmoney can buy, and basicly it works
just gotta first wait till psu manages to stabilise rails, then the fun can start.
ghetto amp.
i did allso mention i think i never come by any real chip, just fakes.
dunno how on earth it did not work properly for You, mostly the oscillation part is interesting.
sometimes they can "motorboat" , catch DC fan inducted noise, radio station,
but high freq. oscillation, i did never have.
and i have done some clumsy builds with it to be honest.
An air core with 3mm diameter wire will work fine for very low frequencies but become very lossy indeed due to proximity effect at higher freqs (above a few 100s of Hz typically), My B&W CM1s have a similar air core inside them - ferrite's a much better (lower loss, lower noise pick up) solution provided its kept from saturating.
+1 - great reference work that for top-notch sounding amp chips!
Many of the Philips chips have multiple grounds which is a bonus because they identify whether its signal or power ground they're talking about. Unlike on the NS chips where there's no apparent idea that signal and power grounds might want to be separate pins. Philips/NXP also support asymmetrical supplies on various chips and in their application circuits which makes PSU design (already hard enough) twice as easy.
Over here the TDA1514 is about the most expensive of all in the range, on Taobao starting at $1. It could have something to do with Philips branding it 'Hi-fi'. I recently bought a box of TDA1521s for about $0.11 a piece
What are you wanting to do on Taobao? You'll not be able to order things on there because you won't have AliPay. If you want to buy stuff there is AliExpress which is in English. If you want to use Taobao just to check availability and prices then type in the part number like 'TDA8566'.(without the quotes) and hit 'search'.
Have done numerous sanity checks thru the days…did some changes to the supply and decoupling caps…For one thing took away the inductor to wright out the possibility of stray inductance (magnetic fields) and replaced it with small value resistor (0.33 Ω). Since my main concern at this point is the possibility of oscillation also replaced the 270u Oscon’s decoupling caps with ERO’s 100nf as closed to the chip leads as possible and stay as close as the spec sheet recommends.
Even thou this small project seems as easy as changing a light valve it has not been for me so far, cant discover what’s going on except for possible oscillations, as I said before I’m out of a scope to look into the output.
The chip itself doesn’t seem fried as it shows a voltage at the output same as before (11.5V on a 13V supply).
I know Arty has worked before over this chip TDA1557/53 care to splain your setup and decoupling? Are you using decoupling caps at each power takeup?
Antonio
Even thou this small project seems as easy as changing a light valve it has not been for me so far, cant discover what’s going on except for possible oscillations, as I said before I’m out of a scope to look into the output.
The chip itself doesn’t seem fried as it shows a voltage at the output same as before (11.5V on a 13V supply).
I know Arty has worked before over this chip TDA1557/53 care to splain your setup and decoupling? Are you using decoupling caps at each power takeup?
Antonio
Thisone was built in rughly 15 minutes or so.
sits in a cardboard box.
I did not even bothered to solder every connection.
the only cap that is soldered directly to the chip is the 100 nf cap between supply and power ground.
as you can see all the input caps and powersupply de-coupling is quite far from the chip, no twisted wires or what so ever.
Yet, it is DEA silent, up to around half power i can not hear a hiss from the speakers, with the amp connected to a computer. The spekaers have efficiency past 90 dBL and are 4 ohm nominal rated.
In short, it is not noisy, does not oscillate.
The schematic is the most basic one can use,
expect i added a pot that can attenuate stuff over 40 hz.
sortha like an adjustable BSC compensation.
The basic schematic is realy simple,
it ha snothing else than the input caps and the 100 nf cap for stability,
and a swich wired to mute/standby function.
other than that all it needs is star grounding,
22-47 Kohm ressitors from input to signal ground,
a 5 ish ohm ressitor between star and power ground, and its good to go .
the amp i show you uses 2200 uF /25 volt decoupling,
and the power source is a DC adaptor, unregulated, 15 volt DC unloaded, 45 VA.
it is realy that simple.
no oscon caps, no pcb, no clever stuff to make it work.
Recently i added 330 pF caps from input to signal ground,
at high levels some radio station was audible.
that solved the issue.
the reason this amp was created is to have a "beater" amp to be abused on outdoor party times. as it runs from a car battery too.
sits in a cardboard box.
I did not even bothered to solder every connection.
the only cap that is soldered directly to the chip is the 100 nf cap between supply and power ground.
as you can see all the input caps and powersupply de-coupling is quite far from the chip, no twisted wires or what so ever.
Yet, it is DEA silent, up to around half power i can not hear a hiss from the speakers, with the amp connected to a computer. The spekaers have efficiency past 90 dBL and are 4 ohm nominal rated.
In short, it is not noisy, does not oscillate.
The schematic is the most basic one can use,
expect i added a pot that can attenuate stuff over 40 hz.
sortha like an adjustable BSC compensation.
The basic schematic is realy simple,
it ha snothing else than the input caps and the 100 nf cap for stability,
and a swich wired to mute/standby function.
other than that all it needs is star grounding,
22-47 Kohm ressitors from input to signal ground,
a 5 ish ohm ressitor between star and power ground, and its good to go .
the amp i show you uses 2200 uF /25 volt decoupling,
and the power source is a DC adaptor, unregulated, 15 volt DC unloaded, 45 VA.
it is realy that simple.
no oscon caps, no pcb, no clever stuff to make it work.
Recently i added 330 pF caps from input to signal ground,
at high levels some radio station was audible.
that solved the issue.
the reason this amp was created is to have a "beater" amp to be abused on outdoor party times. as it runs from a car battery too.
An externally hosted image should be here but it was not working when we last tested it.
Thats right, I have only checked between outputs (BTL) but not to gnd which of course will not sit a 1/2 rail in this condition.
Arty,
Thanks for your explanation. This is my fourth chip, the last 3 has shown same anomaly and have been treated with care during installation and soldering. Sometimes suppliers send faulty parts to 3rd world countries which only care for price and not quality, retail for this parts here are 3.70$US and are marked to be produced in Taiwan, who knows if they are faulty even if I resist to accept that.
So you use 2 Decoupling caps one at each power takeup?
Thanks for your explanation. This is my fourth chip, the last 3 has shown same anomaly and have been treated with care during installation and soldering. Sometimes suppliers send faulty parts to 3rd world countries which only care for price and not quality, retail for this parts here are 3.70$US and are marked to be produced in Taiwan, who knows if they are faulty even if I resist to accept that.
So you use 2 Decoupling caps one at each power takeup?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.