Hi there,
I have a 2x2 ohm car subwoofer, and I have just recieved 6 LM4780 power opamps.
Now my question is, how should I drive the subwoofer?
I have a 25-26volt supply from 2 carbatterys, and I was thinking of driving each 2 ohm with 2 LM4780 in a bridged mode, where each of them is paralleled.
But can it handle 2ohm?
According to the datasheet I would get about 100watt in 8ohm brigded, so will that mean 400 watt in 2ohm?
But ofc. I could just lower the gain of the amps to get lower output.
But when they say each channel can handle 60watt, is it peak or rms?
Hope you can help me with solving this problem.
Regards,
Simon H.A.
I have a 2x2 ohm car subwoofer, and I have just recieved 6 LM4780 power opamps.
Now my question is, how should I drive the subwoofer?
I have a 25-26volt supply from 2 carbatterys, and I was thinking of driving each 2 ohm with 2 LM4780 in a bridged mode, where each of them is paralleled.
But can it handle 2ohm?
According to the datasheet I would get about 100watt in 8ohm brigded, so will that mean 400 watt in 2ohm?
But ofc. I could just lower the gain of the amps to get lower output.
But when they say each channel can handle 60watt, is it peak or rms?
Hope you can help me with solving this problem.
Regards,
Simon H.A.
Bridged and parallel are not the same, and you can definitely not use bridged with 2 ohm speakers. If the chips can handle 4 ohms in "single" mode, you double the impedance to minimum 8 ohms in bridged mode. You want parallel mode in order to drive low impedances. With 2 channels parallelled the chips can handle 2 ohm. Each LM4780 is basically 2 LM3886 amps in one chip. If you use one LM4780 (in parallel configuration) per subwoofer voice coil, you should have no problem driving it.
Each chip will put out a bit more than 100W at 25V. So using 1 chip in parallel mode for each voice coil would give you a total of a little more than 200W. Which is RMS, not peak, and a lot more power than people who are used to typical retail "marketing ratings" realize.
Look up the "Overture Design Guide" and app note AN-1192.
Each chip will put out a bit more than 100W at 25V. So using 1 chip in parallel mode for each voice coil would give you a total of a little more than 200W. Which is RMS, not peak, and a lot more power than people who are used to typical retail "marketing ratings" realize.
Look up the "Overture Design Guide" and app note AN-1192.
I would say it's likely to be loud enough. Is the subwoofer in a sealed or ported enclosure?60 watts into 4r0 is the limit from +-26Vdc.
in parallel mode you can get ~120W into 2r0
in bridge mode you can get ~120W into 8r0.
If you parallel three sets of bridged you can get ~360W into 2r7.
But you cannot drive a 2.7ohm speaker to this level. The 6chipamps in bridged parallel can only drive >=2r7 resistor and will not be able to produce any sound.
If you fit 4ohm speakers to a 6chip bridged parallel amplifier you should get ~ 240W into 4ohm.
According to National these 6chips (3 off 4780) requires 2.4C/W for each single chip. That's ~0.4C/W for one channel, if Ta<=25degC.
in parallel mode you can get ~120W into 2r0
in bridge mode you can get ~120W into 8r0.
If you parallel three sets of bridged you can get ~360W into 2r7.
But you cannot drive a 2.7ohm speaker to this level. The 6chipamps in bridged parallel can only drive >=2r7 resistor and will not be able to produce any sound.
If you fit 4ohm speakers to a 6chip bridged parallel amplifier you should get ~ 240W into 4ohm.
According to National these 6chips (3 off 4780) requires 2.4C/W for each single chip. That's ~0.4C/W for one channel, if Ta<=25degC.
Is that 1*25-26 V from two 12 V batteries or 2*25-26 V from two 24 V batteries?Cyberzim said:
If it is from 2*12 V, the LM4780 will handle 2 Ohm speakers easily (both channels in parallel for 1*2 Ohm), but you will only get something like 24 W per voice-coil out of it or 48 W with 2 LM4780 and 2*2 Ohm. Paralleling more of them, won't help. The good thing is, heatsinking will be easy.
If it is from 2*24 V, use the two voice-coils in series. That way you get a 4 Ohm load, which should be manageable with at least 4 channels in parallel, that means also 2 LM4780s for the woofer. The resulting ~120 W means you gain 4 dB over the 2*12 V with 2*2 Ohm version. More LM4780s in parallel could help alleviate the heat dissipation issue.
Both versions will almost certainly drive the subwoofer into its mechanical limits below the port's resonant frequency. Consider a subsonic filter.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=104402&highlight=
Yes, but it is around 25,3 volt.
Wikipedia says a car battery that is 100% charged has 12.65 V and when it drops down to 12.06 it's only 25% charged, so my working voltage will be 25,3 volt and that will give me 112,57 watt according to the overture design guide, but I think I'll lower the gain so I get around 100watt each coil to make it more stable.
Regards,
Simon H.A.
Wikipedia says a car battery that is 100% charged has 12.65 V and when it drops down to 12.06 it's only 25% charged, so my working voltage will be 25,3 volt and that will give me 112,57 watt according to the overture design guide, but I think I'll lower the gain so I get around 100watt each coil to make it more stable.
Regards,
Simon H.A.
+-12.65Vdc = 25.3Vdc.
Page 14 power prediction graph only goes down to 4r0 and down to 28Vdc.
This cannot be used to predict the power of your proposal.
page10 shows graphs for 8r0 and 4r0 "Clipping Voltage vs Supply Voltage".
+-12.65Vdc & 8r0 loses ~2V through the amplifier.
+-12.65Vdc & 4r0 loses ~2.4V through the amplifier.
Expect +-12.65Vdc and 2r0 to lose more than 2.8V through the amplifier.
That leaves a peak output voltage <=9.85Vpk to your 2r0 load. If the supply voltage at the chipamp pins stays at +-12.65Vdc. It won't!
The maximum power from one chipamp channel is 24W into 2r0
and 13W into 4r0.
A bridged pair will get you to 48W into 4r0 and 26W into 8r0.
Now start paralleling them.
Two LM4780 chips (4channels of amplifier) will give a maximum of 96W into 2r0 and 52W into 4r0.
But these are resistive load power predictions.
If you want to drive speakers then put at least three channels in parallel.
i.e a 3chip lm4780 in bridged parallel (6channels) may give as much as 96W into a 2ohm speaker.
I suspect it will be <=60W into 2r0.
That's an awful lot of hardware for 60 to 96W.
Page 14 power prediction graph only goes down to 4r0 and down to 28Vdc.
This cannot be used to predict the power of your proposal.
page10 shows graphs for 8r0 and 4r0 "Clipping Voltage vs Supply Voltage".
+-12.65Vdc & 8r0 loses ~2V through the amplifier.
+-12.65Vdc & 4r0 loses ~2.4V through the amplifier.
Expect +-12.65Vdc and 2r0 to lose more than 2.8V through the amplifier.
That leaves a peak output voltage <=9.85Vpk to your 2r0 load. If the supply voltage at the chipamp pins stays at +-12.65Vdc. It won't!
The maximum power from one chipamp channel is 24W into 2r0
and 13W into 4r0.
A bridged pair will get you to 48W into 4r0 and 26W into 8r0.
Now start paralleling them.
Two LM4780 chips (4channels of amplifier) will give a maximum of 96W into 2r0 and 52W into 4r0.
But these are resistive load power predictions.
If you want to drive speakers then put at least three channels in parallel.
i.e a 3chip lm4780 in bridged parallel (6channels) may give as much as 96W into a 2ohm speaker.
I suspect it will be <=60W into 2r0.
That's an awful lot of hardware for 60 to 96W.
you must read the National datasheet carefully.
It swaps between dual polarity voltages +-xVdc and single polarity voltages 2xVdc.
Read what the data tells you.
reading posts 6, 12 & 18 and the National information gives you all you need.
In my opinion you are wasting good resources using paralleled chipamps to drive 2ohm speakers.
It swaps between dual polarity voltages +-xVdc and single polarity voltages 2xVdc.
Read what the data tells you.
reading posts 6, 12 & 18 and the National information gives you all you need.
In my opinion you are wasting good resources using paralleled chipamps to drive 2ohm speakers.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.