Hi guys,
I've been working on a small boombox based on TD7396 chip.
using 4 Ohm dayton ND140 for lows-mid in parallel with 8 Ohm dayton DC28F-8 for highs.
The system sounds quite good but not as loud as i expected ( and intended, it's a boombox after all)
I checked consumption with a bench power supply, and it draws 640mA @ 14.1V so around 9 watts for a steady 100Hz sinewave
The chip datasheet says that for 14V power is 20W for 4Ohms or 32W for 2Ohms.
As my load is about 2.66Ohms : 1/(1/8+1/4), i would assume to be in the 27ishW. Instead of that, i'm not even at the rated power for 4Ohm !
Chips were buy from farnell, a reliable electronic retailer, i don't think those are fake
What am i missing ? Do someone has any ideas ?
I've been working on a small boombox based on TD7396 chip.
using 4 Ohm dayton ND140 for lows-mid in parallel with 8 Ohm dayton DC28F-8 for highs.
The system sounds quite good but not as loud as i expected ( and intended, it's a boombox after all)
I checked consumption with a bench power supply, and it draws 640mA @ 14.1V so around 9 watts for a steady 100Hz sinewave
The chip datasheet says that for 14V power is 20W for 4Ohms or 32W for 2Ohms.
As my load is about 2.66Ohms : 1/(1/8+1/4), i would assume to be in the 27ishW. Instead of that, i'm not even at the rated power for 4Ohm !
Chips were buy from farnell, a reliable electronic retailer, i don't think those are fake
What am i missing ? Do someone has any ideas ?
Audio "watts" are not calculated in the same way as electrical watts, which is what you have done.
The TDA7396 is a mono bridged amplifier. With a 14 volt power supply and a 4 ohm load you should be drawing around 3 amps peak and around 2 amps RMS when clipping, providing there are no losses. If you put a meter across the speakers what is the maximum RMS voltage you read before obvious distortion occurs? Use a low frequency, 50 60 Hz, as many low cost meters are not accurate at higher audio frequencies. Do you have a large enough input signal to drive the amplifier to clipping?
It is possible you are using only 1/2 of the bridged amplifier. Are your speakers connected to the + and - speaker connections of the chip or from the + and ground connections?
It is possible you are using only 1/2 of the bridged amplifier. Are your speakers connected to the + and - speaker connections of the chip or from the + and ground connections?
Thanks for your replies,
first of all, i made a significant improvement by getting rid of my crossover circuit
her what i've got :
100Hz on the 4Ohm without crossover : 1.02A @ 14.1V, corresponding to the specs
100Hz on the 4Ohm with crossover ( eg. the 8Ohm disconnected) : 640mA @ 14.1V
Does such losses seems normal to you guys ?
I didn't made that crossover which is presumably crappy, i'll investigate that
@multisync
i don't have signals large enough to clip it, even if my keyboard is outputting 4v-pp audio signals, i may try with my oscilloscope demo signal.
i ran your test at 50Hz , still 14.1v supply, it draws 1.323A for 10.7V RMS
I checked my design and it looks fine, the speaker connects to OUT+/OUT-, and OUT- and GND are not shorted
first of all, i made a significant improvement by getting rid of my crossover circuit
her what i've got :
100Hz on the 4Ohm without crossover : 1.02A @ 14.1V, corresponding to the specs
100Hz on the 4Ohm with crossover ( eg. the 8Ohm disconnected) : 640mA @ 14.1V
Does such losses seems normal to you guys ?
I didn't made that crossover which is presumably crappy, i'll investigate that
@multisync
i don't have signals large enough to clip it, even if my keyboard is outputting 4v-pp audio signals, i may try with my oscilloscope demo signal.
i ran your test at 50Hz , still 14.1v supply, it draws 1.323A for 10.7V RMS
I checked my design and it looks fine, the speaker connects to OUT+/OUT-, and OUT- and GND are not shorted
10 volts RMS is about right with a 14 volt power supply. Your current draw will depend on the frequency you are using. The AC impedance of the speaker varies a lot with frequency. At or near resonance a 4 ohm speaker may be 40 ohms or higher and will draw a lot less current. Do you have an inductor in series with the 4 ohm speaker? By knowing the inductors DC resistance and Inductance value you can do some calculations to find out the losses in the crossover. The crossover will have impedance values depending on frequency. Basically at the crossover frequency the inductor in series with the speaker will be approximately equal to the speaker. Put a very small value resistor in series with the 4 ohm speaker ie 0.1 ohm or 1 ohm, sweep the frequency from 30 Hz to 500 Hz, measure the voltage across the resistor. The current will be the voltage divided by the resistor value. You will see the current vary a lot
A "4 Ohm" speaker is nothing like a 4 Ohm resistor. As Multisync saz, it's probably closer to 40 Ohms at 100Hz. And you need a signal generator and scope to do this. If you don't have these, you can probably use your computer sound card and some freeware. You will need a 10:1 (or more) divider for the "scope" input to the soundcard. While you are at it, you can plot your speaker impedance. Again, see internet freeware apps.
Best way to measure amp power output is with audio "dummy load", which is just a big power resistor(s) of 100 Watts or so and 2, 4, or 8 ohms.
Easy to build and a lot quieter than a speaker.
Easy to build and a lot quieter than a speaker.
Thanks folks for your instructions,
I guess we found the main problem which is that i was using a crossover not suited for my configuration.
I checked the crossover, see what it was about :
In+ goes straight to Woofer +
In+ goes through 4.7µF cap in series with a 20mH coil to Tweeter+
In- is straight to Woofer- and Tweeter-
I don't know how it could mess that much but experimentation showed us.
without this crossover power draw and voltage RMS seems quite adequate to tda7396 spec for 4Ohm speaker, unless, it is still much better and closer to what i expected to see and hear
My guess now is to design my own crossover an see how it turns out before investigate any further, don't you think ?
Thank you guys, your inputs are much appreciated
I guess we found the main problem which is that i was using a crossover not suited for my configuration.
I checked the crossover, see what it was about :
In+ goes straight to Woofer +
In+ goes through 4.7µF cap in series with a 20mH coil to Tweeter+
In- is straight to Woofer- and Tweeter-
I don't know how it could mess that much but experimentation showed us.
without this crossover power draw and voltage RMS seems quite adequate to tda7396 spec for 4Ohm speaker, unless, it is still much better and closer to what i expected to see and hear
My guess now is to design my own crossover an see how it turns out before investigate any further, don't you think ?
Thank you guys, your inputs are much appreciated
Hi guys.
I finnaly received the few components i needed to make my own crossover(Butterworth 1st order( i wanted to keep it simple)). The result is quite good and the sound level meet my expectations.
I ran a new test 100Hz sinewave on the 8Ohm + 4Ohm with my new crossover, the system draws 1532mA@14.1V,
for reminder, the same test but with my old random crossover => 640mA@14.1V
Yes this isn't Watt RMS, but it's still an indication of what's happening. My main concern was the sound level, it noticeably increased
In my opinion the problem is solved, i did not expected the crossover to be the root cause of this problem, i guess i learned something today.
@nigelwright7557 Yes you're (w)right ! I planned to change for speakers with higher sensitivity in case the crossover didn't solved the problem. Glad to don't have to do so !(i'm broke)
I finnaly received the few components i needed to make my own crossover(Butterworth 1st order( i wanted to keep it simple)). The result is quite good and the sound level meet my expectations.
I ran a new test 100Hz sinewave on the 8Ohm + 4Ohm with my new crossover, the system draws 1532mA@14.1V,
for reminder, the same test but with my old random crossover => 640mA@14.1V
Yes this isn't Watt RMS, but it's still an indication of what's happening. My main concern was the sound level, it noticeably increased
In my opinion the problem is solved, i did not expected the crossover to be the root cause of this problem, i guess i learned something today.
@nigelwright7557 Yes you're (w)right ! I planned to change for speakers with higher sensitivity in case the crossover didn't solved the problem. Glad to don't have to do so !(i'm broke)
Oops! What's that? A series resonant circuit = band pass filter in series with a tweeter? Rather I'd expect that in a squawker...
Best regards!
Why not make it biamplified adding a smaller amp (TDA2003 for example) for tweeter and make a passive crossover? I have my system did in such a form.
No, not seriously that high THD TDA2003 of Emperor William's era just for the tweeter 😵 !
Best regards!
Best regards!
Hi Osvaldo, this is a good idea, i gave it a thought in case i didn't manage to get better sound level. It may be overkill but i was even thinking about to use a second TDA7396 as i made some pcbs ready for stereo applications, it'll need just a few tweaks to get it to a biamplified system. I'll give it a try in near future, i'll make sure to give you guys a feedback. Just wondering, what do you mean by passive crossover ? At first glance, i would go for a crossover on input side rather than outputs, am i missing a better way ?
Hello Naresh, i didnt try it since i've changed the crossover, i intend to make it battery operated so i didn't dig this way. Still a good idea, i may try it, thanks
Passive crossover may be simple RC networks and perhaps associated to some double opamp to make an active one. If the set will be stereo, a TDA2822 may contain both highs drivers.