EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: Adhith on July 24, 2018, 04:44:21 pm
-
hello everyone...
I'm building a speaker setup using the 25w+25w amplifier board based on the class D amplifier IC TDA7492P. The amp board also has an inbuilt Bluetooth receiver module and also I'm using a 4ohm driver at the output. The description about the amp board says that it is ideal for 8 ohm drivers even though it could work with 6 and 4 ohms without problem. Since 4 ohm was on the borders of limit and since Im powering the amp board with a 4S1P battery pack I'm thinking of using a higher ohm driver to reduce the total power consumption.
So what my question is, is there any way to increase the impedence with out changing the 4ohm driver that i'm currently using. does a 4 ohm resistor in series with the driver work?? if so what should be the wattage rating of it??
-
you can add a resistor, but you can also just keep the output level low enough to not exceed the rating of the amp
-
If you add a resistor you will just drop on efficency and peak power capability. Also damping factor will drop dramatically.
With 4S you have about 15V nominal, into the 4Ω it's baerly 25W rms if the tpa goes close to rails, as it should, with music the average power consumption will be much lower, like under 10W with a reasonable THD. But. You could get 800W PMPO if you want to sell it.
Adding the resistor makes no sense, higher impedance coils makes better response and efficency for higher voltage situations, but adding a resistor in series only makes those numbers worse.
JS
-
you can add a resistor, but you can also just keep the output level low enough to not exceed the rating of the amp
The reason that I'm planning to higher the impedence is because, during startup the amp instantly consumes a peak current of 1.8A for 2 seconds and then lowers the current draw to a very low level. This peak current is making the problem of the bms to shut down(maybe it was kind of confusing as a short due to the high current draw of 1.8A). So the board is not at all even getting started up at the first place since the bms turns off before it.
But if the drivers are disconnected during startup and connected after the first two seconds of startup, then there is no problem and everything works smoothly.
This board does have this thing of drawing too much current at the start up since i have found some articles about the same on internet.
-
If you add a resistor you will just drop on efficency and peak power capability. Also damping factor will drop dramatically.
With 4S you have about 15V nominal, into the 4Ω it's baerly 25W rms if the tpa goes close to rails, as it should, with music the average power consumption will be much lower, like under 10W with a reasonable THD. But. You could get 800W PMPO if you want to sell it.
Adding the resistor makes no sense, higher impedance coils makes better response and efficency for higher voltage situations, but adding a resistor in series only makes those numbers worse.
JS
Thank you very much for your reply. So I understood that adding resistor is a bad choice. So are you saying that the better option is finding another driver of some higher impedance like 8 ohm??
-
you can add a resistor, but you can also just keep the output level low enough to not exceed the rating of the amp
The reason that I'm planning to higher the impedence is because, during startup the amp instantly consumes a peak current of 1.8A for 2 seconds and then lowers the current draw to a very low level. This peak current is making the problem of the bms to shut down(maybe it was kind of confusing as a short due to the high current draw of 1.8A). So the board is not at all even getting started up at the first place since the bms turns off before it.
But if the drivers are disconnected during startup and connected after the first two seconds of startup, then there is no problem and everything works smoothly.
This board does have this thing of drawing too much current at the start up since i have found some articles about the same on internet.
A common solution for these problems is a relay that will connect the speakers only after a few seconds delay - once all the capacitors are charged and the amp has stabilized. It also prevents the nasty "THUMP" from the speakers on power up which could damage them.
Something like a 555 timer in a monostable mode + a transistor to drive a small relay would do the job just fine.
-
A common solution for these problems is a relay that will connect the speakers only after a few seconds delay - once all the capacitors are charged and the amp has stabilized. It also prevents the nasty "THUMP" from the speakers on power up which could damage them.
Something like a 555 timer in a monostable mode + a transistor to drive a small relay would do the job just fine.
Thank you very much for you help Sir. Since I'm a beginner could you help me with a suitable diagram?? maybe a circuit that connects the speaker after 4 or 5 seconds??
-
So are you saying that the better option is finding another driver of some higher impedance like 8 ohm??
8 ohm is very common, 16 is also not hard to get from OEMs.
-
See Paul McGowan's explanation (https://www.youtube.com/watch?v=j8liG1nLfcY) on "impedance matching" and some of the common preconceptions. As mentioned earlier, adding a resistor will at least dissipate half of the power in the resistor, among other unwanted effects. Note: Paul McGowan is the CEO of Psaudio.
-
The datasheet of the European TDA7492P says that it can drive a 6 ohm or 8 ohm speaker but not a 4 ohm speaker unless the supply voltage is reduced because its maximum allowed peak output current and its heating must not be exceeded. The Chinese module you have says different things, maybe its TDA7492P is a defective Chinese copy.
Your 6 cells battery is 16.8V when fully charged and the output from the amplifier into a 6 ohms speaker is 16W so its peak voltage will be about 12V into 4 ohms which needs a peak current of only 3A. The Chinese heatsink might be too small for the output to be 16W into 6 ohms and much too small for more than 16W into 4 ohms.
Your battery and its BMS are too small for only one channel of this two channels amplifier.
Adding a resistor in series with a speaker ruins the excellent damping from the amplifier. Then audio sounds ring like a bell and low frequencies sound boomy like a bongo drum and higher frequencies shriek.
-
If you add a resistor you will just drop on efficency and peak power capability. Also damping factor will drop dramatically.
With 4S you have about 15V nominal, into the 4Ω it's baerly 25W rms if the tpa goes close to rails, as it should, with music the average power consumption will be much lower, like under 10W with a reasonable THD. But. You could get 800W PMPO if you want to sell it.
Adding the resistor makes no sense, higher impedance coils makes better response and efficency for higher voltage situations, but adding a resistor in series only makes those numbers worse.
JS
Thank you very much for your reply. So I understood that adding resistor is a bad choice. So are you saying that the better option is finding another driver of some higher impedance like 8 ohm??
No, I actually think 4Ω is the best option, as it usially is in low voltage situations. With 8Ω you will only get about 7W RMS instead of the intended 25.
JS
-
The reason that I'm planning to higher the impedence is because, during startup the amp instantly consumes a peak current of 1.8A for 2 seconds and then lowers the current draw to a very low level.
This isn't just an inrush current? That won't be affected by speaker impedance at all. On the other hand, if you have large DC spikes on the output during power-on those need to be addressed independently or you will eventually destroy the speaker coils.
-
A common solution for these problems is a relay that will connect the speakers only after a few seconds delay - once all the capacitors are charged and the amp has stabilized. It also prevents the nasty "THUMP" from the speakers on power up which could damage them.
Something like a 555 timer in a monostable mode + a transistor to drive a small relay would do the job just fine.
Thank you very much for you help Sir. Since I'm a beginner could you help me with a suitable diagram?? maybe a circuit that connects the speaker after 4 or 5 seconds??
There are plenty of these designs online, search for "speaker delay circuit", "speaker protection circuit", "anti-pop circuit", etc.
E.g. something like this one (didn't test it myself):
http://www.redcircuits.com/Page171.htm (http://www.redcircuits.com/Page171.htm)
Just a transistor/darlington and a few passives. The large capacitor will slowly charge up making the transistor eventually turn on and energize the relay connecting the speakers.
If your amplifier has a microcontroller for something, you can use that for a delay too. There are many ways how to build this.
-
Thank you janoc. I'll refer the links that you have suggested
-
Why not just use a 555 timer?
This should be unnecessary for the TDA7492P, which has mute and standby inputs, that can be used to delay the turn on and avoid any audible clicks, during power on.
In this case, I think this has nothing to do with the speakers. You'll probably find the board still draws a huge inrush current, even with the speakers disconnected. Refer to the circuit on page 5 of the data sheet. It has a 1000µF capacitor (C32, next to J2) across the power rail, which is probably causing the problem.
https://www.st.com/resource/en/datasheet/tda7492p.pdf (https://www.st.com/resource/en/datasheet/tda7492p.pdf)
Try adding a 22R resistor in series with the amplifier board, for the first second, after applying the power, then bypass it with a wire link. It will limit the current to a safe level and stop the BMS shutting down. A more permanent solution is to use a timer relay (either a 555 & relay or you might be able to find a module) to bypass the resistor, after 0.1s to 1s, of the power being applied.
The above solution will only work at low power levels. You need a more powerful BMS which is capable of providing the amplifier, with the current it needs, otherwise this will happen again, when the volume is turned up too high and there's lots of bass in the music. Your amplifier needs a power supply capable of providing 6A peak, to stand any chance of being able to work at full power. Even changing to an 8R speaker will still cause the amplifier to draw more current, with the volume turned up, than your current BMS can provide.
-
NTC is what's used to limit inrush, no need for resistor and bypassing as this devices do all for you
JS
-
NTC is what's used to limit inrush, no need for resistor and bypassing as this devices do all for you
JS
I'm doubtful an NTC resistor will work in this application. If this were for a motor or incandescent lamp, I'd agree, but I'm not sure it'll work here.
An audio amplifier will draw huge peak currents, along with a very low quiescent current, so I don't think the resistor will heat up enough to reduce its value low enough, to power the circuit. I think the value of the NTC resistor will stay high, after power on, causing the supply voltage to collapse, resulting in lots of distortion, as soon as the beat drops.
-
If it is just a thump that worries you, Add that resistor in series with the speaker and just add a 470uF capacitor in parallel with that. Soles the low resistance DC startup without changing audio much.
-
If it is just a thump that worries you, Add that resistor in series with the speaker and just add a 470uF capacitor in parallel with that. Soles the low resistance DC startup without changing audio much.
How would that help?
The capacitor would just bypass the resistor at AC, causing the thump to happen anyway. The resistor won't do anything!
-
Why not just use a 555 timer?
Thank you very much Sir for your reply. I'm trying to solve this issue with a 555 timer which you have suggested in my other thread( 555 timer circuit for 5 seconds).
In this case, I think this has nothing to do with the speakers. You'll probably find the board still draws a huge inrush current, even with the speakers disconnected. Refer to the circuit on page 5 of the data sheet. It has a 1000µF capacitor (C32, next to J2) across the power rail, which is probably causing the problem.
I had connected an ammeter across the amplifier board to see the current draw. As I mentioned before the current draw was about 1.8A at the start up and then reducing considerably. But when I disconnected one speaker the current draw is only about 1A and the BMS is driving the amp smoothly without any trouble. So I guess there is some rush of current to the speaker during start up or I'm totally wrong??
Another thing about the amp board is that, it plays a music at the startup of the amp and the BMS is shutting down just before that music is played. So does this means that the startup music is drawing all this current??
-
The mute function can be used, as described in the TDA7492P data sheet. Failing that, use the timer relay circuit to bypass the current limiting resistor, once the surge has passed. 5 seconds is probably much longer than necessary. 1 second or less will do.
(https://www.eevblog.com/forum/beginners/problem-with-powering-an-amplifer-board/?action=dlattach;attach=484259;image)
-
The mute function can be used, as described in the TDA7492P data sheet. Failing that, use the timer relay circuit to bypass the current limiting resistor, once the surge has passed. 5 seconds is probably much longer than necessary. 1 second or less will do.
(https://www.eevblog.com/forum/beginners/problem-with-powering-an-amplifer-board/?action=dlattach;attach=484259;image)
Ok got it. but one small thing what would be the wattage rating of the current limiting resistor??
Also would it be wrong to use a dpdt relay to cutoff both the left and right audio output to the drivers??
-
The mute function can be used, as described in the TDA7492P data sheet. Failing that, use the timer relay circuit to bypass the current limiting resistor, once the surge has passed. 5 seconds is probably much longer than necessary. 1 second or less will do.
(https://www.eevblog.com/forum/beginners/problem-with-powering-an-amplifer-board/?action=dlattach;attach=484259;image)
Ok got it. but one small thing what would be the wattage rating of the current limiting resistor??
Also would it be wrong to use a dpdt relay to cutoff both the left and right audio output to the drivers??
The resistor is only used for a short period of time, so can be a fairly low power rating like 1W.
You could add another relay to cut power to the speakers, but it shouldn't be necessary for this circuit, which should limit the inrush current to the amplifier, to a safe level.
-
Another thing about the amp board is that, it plays a music at the startup of the amp and the BMS is shutting down just before that music is played. So does this means that the startup music is drawing all this current??
That just means that your BMS/power supply is not suitable (is way too weak) for powering this amplifier. The amplifier has a low quiescent current (when not playing anything after the initial inrush) but then when you actually attempt to play sound on it, it needs to get the energy from somewhere. The same when there are any signal peaks (e.g. a big drum note). Then the current through the circuit will jump even an order of magnitude higher than the average current of the amplifier for a short moment. Your power supply must be able to handle this otherwise the amplifier will clip severely or you will burn something out.
You must fix this before you try to do anything else because it is meaningless to troubleshoot issues with inrush current and what not when your power supply will not be able to handle the power requirements of your amplifier when actually in use anyway.
BTW, I wouldn't trust what your meter shows you on the inrush current - multimeters are not great at measuring short transient events, especially if you have a cheap meter that makes only 2-3 measurements per second. There could well be a high current spike and you will miss it because it falls between two measurement cycles of your meter. A better way to measure things like this would be using an oscilloscope, e.g. by measuring the voltage drop across a small value shunt resistor.
-
The mute function can be used, as described in the TDA7492P data sheet. Failing that, use the timer relay circuit to bypass the current limiting resistor, once the surge has passed. 5 seconds is probably much longer than necessary. 1 second or less will do.
(https://www.eevblog.com/forum/beginners/problem-with-powering-an-amplifer-board/?action=dlattach;attach=484259;image)
Three ideas:
1) Use the suggested circuit, but bypass the current limiting resistor with a normally closed relay contact, and have a timer which energises it for 5 seconds.
2) Put a normally closed relay contact on the power rail between the reservoir capacitor and the load and drive the relay for 5 seconds at power up time.
That way, you have 5 seconds of the battery feeding capacitor only, followed by the battery topping off the capacitor and powering the load.
3) Have the battery circuit permanently connected to the reservoir capacitor and switch the amp on afterwards.
Same as 2 above, but the relay contacts are replaced by the on/off switch - no timer required.
-
YMMV, but in my own experience it works fine to drive a 4 Ohm speaker from an amplifier designed for 8 Ohm speakers. This may not always hold true but I can't think of a case where I've had any trouble with it so long as the volume is kept at a reasonable level.
Something else you can do is drive a pair of 4 Ohm speakers in series on each channel.
-
The mute function can be used, as described in the TDA7492P data sheet. Failing that, use the timer relay circuit to bypass the current limiting resistor, once the surge has passed. 5 seconds is probably much longer than necessary. 1 second or less will do.
(https://www.eevblog.com/forum/beginners/problem-with-powering-an-amplifer-board/?action=dlattach;attach=484259;image)
Three ideas:
1) Use the suggested circuit, but bypass the current limiting resistor with a normally closed relay contact, and have a timer which energises it for 5 seconds.
2) Put a normally closed relay contact on the power rail between the reservoir capacitor and the load and drive the relay for 5 seconds at power up time.
That way, you have 5 seconds of the battery feeding capacitor only, followed by the battery topping off the capacitor and powering the load.
3) Have the battery circuit permanently connected to the reservoir capacitor and switch the amp on afterwards.
Same as 2 above, but the relay contacts are replaced by the on/off switch - no timer required.
1) The problem with that is it takes a finite time for a normally closed relay's contacts to open and it will need to break the full current surge, so will be less reliable, than a normally open contact.
2 & 3) Good ideas, but it might be difficult/impractical to break the connection between the reservoir capacitor and amplifier, if the PCB is already made.
YMMV, but in my own experience it works fine to drive a 4 Ohm speaker from an amplifier designed for 8 Ohm speakers. This may not always hold true but I can't think of a case where I've had any trouble with it so long as the volume is kept at a reasonable level.
Something else you can do is drive a pair of 4 Ohm speakers in series on each channel.
Yes an amplifier designed of a 4R speaker should be fine with an 8R load.
I wouldn't recommend connecting speakers in series, because it will result in a poorer damping factor and more peaks and troughs in the frequency response.
-
Thank you every one once again for all those sincere and detailed explanation. I'll build one such circuit and let you guys know about the result.
-
I wouldn't recommend connecting speakers in series, because it will result in a poorer damping factor and more peaks and troughs in the frequency response.
Players of electric guitars like that awful sound and call it "effects".