So I have a class A/B car amplifier that I've measured with an oscilloscope to do 100 watts rms per channel with a 1khz sine wave. I also have a tda7498e chip amp that I've measured to do around 107-115 watts rms both on 4 ohm. The car amp power rails are 40v each and I'm currently running the tda7498e off 8 lipos that start at 33.5v yet the chip amp gets much louder than my class A/B with 80vpp. What gives? The class d sounds much better to me as well. I think it's time to start investigating boost converters. If I can find a converter that works I think I'll run 2 of these btl tda7498es
The gain of the chip amp is probably higher, so it will sound louder than the other amp with the same input level.
You can verify by measuring.
Both amps have similar power specs but the higher gain amp will reach full power with smaller input.
There are other factors related to circuit topology but the gain is likely the most significant.
You can verify by measuring.
Both amps have similar power specs but the higher gain amp will reach full power with smaller input.
There are other factors related to circuit topology but the gain is likely the most significant.
Usually, Class D amps are in BTL configuration. They quadruple the power output comparing an amp without BTL with the same rail voltages.
But how can an amplifier with less than half the rail voltage be more powerful? I understand gain which the tda7498e has less gain than the 4 channel amp BTW. Does it have everything to do with efficiency? If both amps had the exact same gain but the one with a lower voltage power supply is producing more. I understand the tda is in btl mode so that makes a difference. When I bridge the A/B amp it produces more power but not much more and it comes at the expense of sound quality. Is the A/B amp just inefficient and possibly have crappy output transistors?
I'm just trying to wrap my head around this so when the time comes to buy a new amp I'll have a better understanding. Right now I'm thinking to stay away from class A/B amps for the car because of size, heat, and power. Looks like class d is 100% the way to go in a car and leave class A/B for the home where sound quality is more an issue than my truck thats less than ideal for any kind of SQ.
I'm just trying to wrap my head around this so when the time comes to buy a new amp I'll have a better understanding. Right now I'm thinking to stay away from class A/B amps for the car because of size, heat, and power. Looks like class d is 100% the way to go in a car and leave class A/B for the home where sound quality is more an issue than my truck thats less than ideal for any kind of SQ.
Class A/B is about 60% efficient and class D is about 85%.
The power supply of the A/B amp may not have enough current capacity to support full power in BTL mode, or the chip could be current limiting to protect itself.
Last edited:
BTL is like bridging the A/B amp if current, voltage, heat was the same on both.
But often the A/B has not enough current. And if it has enough current it will be overheated.
It needs four times the current and generates 4 times the heat.
So does the class D but the differense is 30% of the energy converts to heat versus 5%.
So 100watts total,, 30 watts A/B, 5 watts D
Bridged, BTL. 120 watts heat A/B, 25 watts D
But often the A/B has not enough current. And if it has enough current it will be overheated.
It needs four times the current and generates 4 times the heat.
So does the class D but the differense is 30% of the energy converts to heat versus 5%.
So 100watts total,, 30 watts A/B, 5 watts D
Bridged, BTL. 120 watts heat A/B, 25 watts D
The A/B amp is a classic transistor amp mounted in my truck. My truck has 2 large batteries in parallel, a 250amp alternator, and 2awg wiring to a distribution block which then splits to a 4awg wire to the amp about 16" long. I can confidently say that the amp is getting all the current it needs and more. One of the batteries is actually mounted right next to the amp. I have had this amp hooked up this way for 4 years now and has been in 3 trucks by now. To test the tda7498e I put together a 8 cell 18650 battery pack to provide 33.6vdc fully charged. Audio signal is fed by an equalizer the same as the 4ch. amp and the gain is set at 26db. My 4ch. A/B amp gain is set around 40% on the gain pot.
What is a good way of measuring the gain of each amp? Can I send a 1v rms sine wave of 1khz into each amp and measure the outputs with an oscilloscope or multi meter set to AC volts?
But heat protection on amp/chip is most often the limiting factor on max total watts out
The technical term is SOA. Safe operating area.
https://en.m.wikipedia.org/wiki/Safe_operating_area
The technical term is SOA. Safe operating area.
https://en.m.wikipedia.org/wiki/Safe_operating_area
I suspect the 40V rails have a lot of droop at full load, due to losses in its power supply - have you monitored the actual supply rail voltages under load?
Linear amplifiers (as opposed to class D) can have a lot of voltage drop-out too - perhaps 4V or so with an EF output stage, so even +/-40V supply only allows 35V peak output waveform before clipping, and if the rails droop, perhaps a lot less. Class D will get pretty much right to the supply rail without droop as the output devices are only a few milliohms when on. A linear amp might have 220 milliohms of emitter resistor as well, another 1.5V loss to contend with at 100W/4ohms
Lithium batteries are very low output impedance too, again a few milliohms typically each, which might be another factor.
Linear amplifiers (as opposed to class D) can have a lot of voltage drop-out too - perhaps 4V or so with an EF output stage, so even +/-40V supply only allows 35V peak output waveform before clipping, and if the rails droop, perhaps a lot less. Class D will get pretty much right to the supply rail without droop as the output devices are only a few milliohms when on. A linear amp might have 220 milliohms of emitter resistor as well, another 1.5V loss to contend with at 100W/4ohms
Lithium batteries are very low output impedance too, again a few milliohms typically each, which might be another factor.