Roushon, you didn't tell us what is the impedance of your speakers ?
If it is 8 ohms you will gain nothing with parallel topology (with your +-35V PS) compared to use of only one half of LM4780 per channel (the output power will be about 60 W in each case). If used on 8 ohms speakers, parallel topology gives you nothing good. It does make sense only with 4 or 2 ohms speakers.
Designers use parallel or bridge topology in order to increase output power (parallel for 4 ohms speakers, bridge for 8 ohms speaker). Parallel connection doubles the output current (every half of the circuit sees the double speaker's impedance on its output) and bridge connection doubles the output voltage (every half of the circuit sees the half of speaker's impedance on its output).
In terms of sound quality, neither bridge nor parallel connection brings no improvement, just more power (at expence of problems of mandatory matching of components, greater complexity, DC offset, greater heat disipation...)
If you have 8 ohms speakers, think it all over again and be sure to carefully read the LM4780 datasheet and the AN-1192 document. Also, you should download the Overture_Design_Guide15.xls (spreadsheet calculation) from National's website to help you with design.
Cheers!
If it is 8 ohms you will gain nothing with parallel topology (with your +-35V PS) compared to use of only one half of LM4780 per channel (the output power will be about 60 W in each case). If used on 8 ohms speakers, parallel topology gives you nothing good. It does make sense only with 4 or 2 ohms speakers.
Designers use parallel or bridge topology in order to increase output power (parallel for 4 ohms speakers, bridge for 8 ohms speaker). Parallel connection doubles the output current (every half of the circuit sees the double speaker's impedance on its output) and bridge connection doubles the output voltage (every half of the circuit sees the half of speaker's impedance on its output).
In terms of sound quality, neither bridge nor parallel connection brings no improvement, just more power (at expence of problems of mandatory matching of components, greater complexity, DC offset, greater heat disipation...)
If you have 8 ohms speakers, think it all over again and be sure to carefully read the LM4780 datasheet and the AN-1192 document. Also, you should download the Overture_Design_Guide15.xls (spreadsheet calculation) from National's website to help you with design.
Cheers!
Power into 8 OHM load
Ohms Law folks its really easy, A*V=W, so a chip amp makes 60 watts into an 8 ohm load with +/-30 rails, thats 1 amp per leg.
Put 2 chips in parellel you double the current to load even at 8 ohms giving an avalable 120 watts to load.
I have built many BPA amps my biggest being a BPA 400+ 8,LM3875 bridged and parellel, tell me it makes only 100watts
Ive had it on the bench at over 600watts peak
Ohms Law folks its really easy, A*V=W, so a chip amp makes 60 watts into an 8 ohm load with +/-30 rails, thats 1 amp per leg.
Put 2 chips in parellel you double the current to load even at 8 ohms giving an avalable 120 watts to load.
I have built many BPA amps my biggest being a BPA 400+ 8,LM3875 bridged and parellel, tell me it makes only 100watts
Ive had it on the bench at over 600watts peak
Re: Power into 8 OHM load
Hi, tiltedhalo
just for the truth's sake, if you look in LM4780 datasheet you will see that it will deliver 45W on +-30V (1% THD, 8 Ohms load) per 1 half of the chip.
If you double the value of the load (16 Ohms) by paralleling two halves of the chip, every half of the LM4780 will give half of the previously achieved power. You can't double the current by increasing the load resistance. The higher the resistance, the less the current (if voltage stays the same, and it does, because we don't change the supply voltage in this case and/or amplification factor).
LM3875's datasheet states that this chip will deliver 43 W on +-30V with 8 Ohms load (THD+N less or equal 0,1% throughout the audio band).
BTW, Ohm's Law is I=U/R and P=I*U (1W=1A*1V) is Joule's Law (these two laws interact since they both take into account certain relationships of voltage and current but from different points of view).
Also, +-30V with 1A per leg would be enough for 60W amplifier only if AB class amplifier had 100% efficiency, but unfortunately it's more like 60%. You surely noticed that amplifier gets hot when it plays the music loud. Where do you think the heat comes from if all the energy of PS goes into music?
Cheers!
tiltedhalo said:Ohms Law folks its really easy, A*V=W, so a chip amp makes 60 watts into an 8 ohm load with +/-30 rails, thats 1 amp per leg.
Put 2 chips in parellel you double the current to load even at 8 ohms giving an avalable 120 watts to load.
I have built many BPA amps my biggest being a BPA 400+ 8,LM3875 bridged and parellel, tell me it makes only 100watts
Ive had it on the bench at over 600watts peak
Hi, tiltedhalo
just for the truth's sake, if you look in LM4780 datasheet you will see that it will deliver 45W on +-30V (1% THD, 8 Ohms load) per 1 half of the chip.
If you double the value of the load (16 Ohms) by paralleling two halves of the chip, every half of the LM4780 will give half of the previously achieved power. You can't double the current by increasing the load resistance. The higher the resistance, the less the current (if voltage stays the same, and it does, because we don't change the supply voltage in this case and/or amplification factor).
LM3875's datasheet states that this chip will deliver 43 W on +-30V with 8 Ohms load (THD+N less or equal 0,1% throughout the audio band).
BTW, Ohm's Law is I=U/R and P=I*U (1W=1A*1V) is Joule's Law (these two laws interact since they both take into account certain relationships of voltage and current but from different points of view).
Also, +-30V with 1A per leg would be enough for 60W amplifier only if AB class amplifier had 100% efficiency, but unfortunately it's more like 60%. You surely noticed that amplifier gets hot when it plays the music loud. Where do you think the heat comes from if all the energy of PS goes into music?
Cheers!
better research
You might want to take a closer look at both the data sheets and some electronics books and re-work some common sense into the eqation, each chip will see 8 ohms not 16 there for doubling the current avalable to load period. The load is not split between the two parellel chips, each chip see 8 ohms.
This subject is lightly touched on in National BPA200 notes and more so in there PA100 notes.
Sorry for getting a little over jealous, but Ive been through this a hundred times even had to build a PA to prove the point to a local college Professor, these are not parellel emitter outputs in a discrete amp. These chip see and dump there output to load in a parellel config given proper PS and designs.
You might want to take a closer look at both the data sheets and some electronics books and re-work some common sense into the eqation, each chip will see 8 ohms not 16 there for doubling the current avalable to load period. The load is not split between the two parellel chips, each chip see 8 ohms.
This subject is lightly touched on in National BPA200 notes and more so in there PA100 notes.
Sorry for getting a little over jealous, but Ive been through this a hundred times even had to build a PA to prove the point to a local college Professor, these are not parellel emitter outputs in a discrete amp. These chip see and dump there output to load in a parellel config given proper PS and designs.
Bridge or parallel?
By the way it does make over 400 watts into 8 ohms
By the way it does make over 400 watts into 8 ohms
An externally hosted image should be here but it was not working when we last tested it.
Sorry tiltedhalo but you missinterpreted the documents that you mention. Perhaps you should do it the easier way. Use the Overture_Design_Guide15.xls (you can download it from National's website).
You can check this easily. Take a 9V battery. Connect a 100 Ohms between + and - contacts. Measure the current through resistor. Your DMM will read 90 mA. Now connect another battery in parallel to first one. Measure the current through resistor again. It will still read 90 mA. So, engaged power will stay the same (9V * 0.09A=0.81W) no matter how many batteries you connect because the voltage and the load are the same.
Once again, you can not change the current in the circuit if you don't change the voltage and/or load.
It's true that paralleled amplifiers are able to deliver twice the current, but they'll do it only if you supply higher voltage and/or lessen the load. It's somenthing so basic that I won't discuss it any more.
Cheers !
You can check this easily. Take a 9V battery. Connect a 100 Ohms between + and - contacts. Measure the current through resistor. Your DMM will read 90 mA. Now connect another battery in parallel to first one. Measure the current through resistor again. It will still read 90 mA. So, engaged power will stay the same (9V * 0.09A=0.81W) no matter how many batteries you connect because the voltage and the load are the same.
Once again, you can not change the current in the circuit if you don't change the voltage and/or load.
It's true that paralleled amplifiers are able to deliver twice the current, but they'll do it only if you supply higher voltage and/or lessen the load. It's somenthing so basic that I won't discuss it any more.
Cheers !
To tiltedhalo
To tiltedhalo,
It seems some do not agree with you , even though you seem to
have proved your theory by designing and building and showing
a picture. I am confuse because I am just a student learning all I
can. Pictures say a thousand words. Please elaborate on your
build.
Do you have more pictures, schematics, wireing diagrams
kits parts list?
Thank You
To tiltedhalo,
It seems some do not agree with you , even though you seem to
have proved your theory by designing and building and showing
a picture. I am confuse because I am just a student learning all I
can. Pictures say a thousand words. Please elaborate on your
build.
Do you have more pictures, schematics, wireing diagrams
kits parts list?
Thank You
BPA amps
The designs is loosely based on this schematic, but fixed nulling circuits
current feed through power supply, also LM3875's running on+/- 40VDC rails.
The designs is loosely based on this schematic, but fixed nulling circuits
current feed through power supply, also LM3875's running on+/- 40VDC rails.
An externally hosted image should be here but it was not working when we last tested it.
The schematics above was not the subject of discussion.
That drawing shows a combination of bridged and paralleled topology and it will deliver more power because bridging doubles the voltage on the load. This will result in more current flowing through the load.
So, the previous conclusions stay valid: there will be no increase in power without increasing of voltage or reducing of the load's resistance.
Cheers!
That drawing shows a combination of bridged and paralleled topology and it will deliver more power because bridging doubles the voltage on the load. This will result in more current flowing through the load.
So, the previous conclusions stay valid: there will be no increase in power without increasing of voltage or reducing of the load's resistance.
Cheers!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.