Hi Sangram,
Thanks for your reply, however I am confused. What does 3 ohms have to do with anything when my speakers are 6? Could you please explain?
Thanks,
Greg.
I would be tempted to say the two channels inside a 4780 are not 4ohm speaker capable, but that many use them into this more demanding load and find the results good enough.
6r0 could demand upto 5Apk. A parallel 4780 can output upto 14Apk to a load. A pair of 4780 channels fed from 22+22Vac should have sufficient current capability to drive a reactive 6ohm speaker load without triggering Spike or any other limiting if you can keep the chip cool. There is your problem.
6r0 could demand upto 5Apk. A parallel 4780 can output upto 14Apk to a load. A pair of 4780 channels fed from 22+22Vac should have sufficient current capability to drive a reactive 6ohm speaker load without triggering Spike or any other limiting if you can keep the chip cool. There is your problem.
Last edited:
Parallel will still produce up 60W into 8ohm, to get more power you need to use 4 ohm speakers.
Two tranies are required for dual mono, more info on the kit here: http://www.diyaudio.com/forums/audio-sector/155635-kit-arrived.html
Two tranies are required for dual mono, more info on the kit here: http://www.diyaudio.com/forums/audio-sector/155635-kit-arrived.html
You must also keep in mind that an amplifier output voltage swing is determined by the power supply's RMS voltage minus 2.5v to 3.5v (see data sheet) because it can't swing compleatly to the rail and not the peak voltage as measured at the filter capacitor.
therefore with a 22v transformer it can only produce less than 80 watts into a 6 ohm load.
At least a 26v-0-26v transformer should be considered and a 28v-0-28v would be optimum to get a full undistorted 100watts into 6ohms.
I have seen this question time and time again and this has never been mentioned or discussed.
Yes,a higher power supply voltage means warmer chip temps,just use an adequate heatsink and you shouldn't have any problems. jer
therefore with a 22v transformer it can only produce less than 80 watts into a 6 ohm load.
At least a 26v-0-26v transformer should be considered and a 28v-0-28v would be optimum to get a full undistorted 100watts into 6ohms.
I have seen this question time and time again and this has never been mentioned or discussed.
Yes,a higher power supply voltage means warmer chip temps,just use an adequate heatsink and you shouldn't have any problems. jer
You're in the clouds instead of reading the threads.
This topic is discussed and argued over repeatedly on this Forum.
I stand on one side with a few compartiots and quite a few others stand on the other side. We will never agree, but we can pass on our thoughts and let the builder decide which way he/she wants to go.
But GO only after deciding. Don't guess.
BTW,
my side says the chipamps in general run out of current capability into real speakers. Low impedance speakers and chipamps do not go well together.
I agree.
The statement I made is true for all amplifiers not just chip amps.
If you were running into 4 ohms I'm sure a 22v transformer would suffice.
In order to get 100 watts RMS into 6 ohms an amplifier has to output 24V RMS and a 22V transformer is not going to produce 24V when the amp starts drawing some current.
Instant peaks it might but not rms,sorry but thats just ohms law.
BTW I have read all of the threads on this subject as well as others in their entirety.
I can also tell you that I had called National Semiconductor when the chips where first released and had got my samples months before they even hit the market and I had talked to an engineer about this very subject ,and what I had just described was exactly what he told me. jer
The statement I made is true for all amplifiers not just chip amps.
If you were running into 4 ohms I'm sure a 22v transformer would suffice.
In order to get 100 watts RMS into 6 ohms an amplifier has to output 24V RMS and a 22V transformer is not going to produce 24V when the amp starts drawing some current.
Instant peaks it might but not rms,sorry but thats just ohms law.
BTW I have read all of the threads on this subject as well as others in their entirety.
I can also tell you that I had called National Semiconductor when the chips where first released and had got my samples months before they even hit the market and I had talked to an engineer about this very subject ,and what I had just described was exactly what he told me. jer
I think some of the science has passed you by.
100W into 6r0 requires 34.64Vpk at the load terminals.
Allow your 3V or so as losses when the amp and it's cabling is passing that maximum current.
The power supply needs to be ~+-37.7V when delivering the 5.77Apk into that 6r0 load.
Now pick an EI transformer rated at 230:22+22Vac and 150VA. It's regulation could be 10%.
Let's see what voltage is stored in the smoothing capacitors just before you apply the test signal for 1second into your dummy load.
nominal mains voltage 240Vac. Actual mains voltage at the time of test is 240Vac.
Secondary voltage of the 22Vac transformer is 25.25Vac when delivering zero current to the load. DC voltage after rectification and smoothing ~+-37Vdc.
Yes you are right +-37Vdc cannot deliver 100W into 6r0 on the assumptions I have taken.
But look at the variables in those assumptions.
Actual Mains voltage.
Transformer regulation.
Rectifier diode voltage drop at near zero current.
Volts drop between PSU and speaker terminal when delivering 5.77Apk.
It would take very little massaging of these variables to ensure that the test on the real amplifier could deliver a real 100W average into a 6r0 load.
But equally, if all the variations stack against maximum current delivery then the test could drop well short of the 100W target/claim.
A blanket statement, as you put it, is patently misleading and further you erroneously claim it is not discussed.
"A blanket statement, as you put it, is patently misleading and further you erroneously claim it is not discussed. "
Yes ,this is very much understood.
I did not mean to imply that this was never discussed,but it has not been mentioned as often ,as it should ,as many times that the question had been asked.
"I think some of the science has passed you by."
You are quite mistaken by that statement as I have just introduced the science.
If one wants to under power their amp and wonder why it doesn't perform like it should and then complain about it ,so be it.
A paralled chip amp with the proper voltage will preform very well with 4ohm to 8ohm load at 100 watts ,and maybe even as low as 3ohms but that would be pushing the limit some.
Needless to say, it would be a very capable amplifier.
If one is worried about the mains fluctuating by 10 % then then extra voltage woud be a benifit in the long run any how. jer
Yes ,this is very much understood.
I did not mean to imply that this was never discussed,but it has not been mentioned as often ,as it should ,as many times that the question had been asked.
"I think some of the science has passed you by."
You are quite mistaken by that statement as I have just introduced the science.
If one wants to under power their amp and wonder why it doesn't perform like it should and then complain about it ,so be it.
A paralled chip amp with the proper voltage will preform very well with 4ohm to 8ohm load at 100 watts ,and maybe even as low as 3ohms but that would be pushing the limit some.
Needless to say, it would be a very capable amplifier.
If one is worried about the mains fluctuating by 10 % then then extra voltage woud be a benifit in the long run any how. jer
Last edited:
Sorry about the newbie questions. what then is the advantage of using a paralleled kit? It seems like it should produce more power having each board joined. Brings me to another question, when using that kit in standard form are each of the amp modules running half power since each is capable of running 2 channels but only drives one channel in that configuration?
Last edited:
The idea behind running parallel amps is to allow more current to flow into a lower resistance load.
Thus, halving the impedence from 8 ohms to 4 ohms doubles the current demand for the same voltage swing that the amplifier has to produce.
Paralleling splits this current demand between the two amplifiers thus allowing double the output power into the 4 ohm load.
Doubling the output voltage into the same load quadruaples the output power.
Bridging two amplifiers together allows the amplifier to output double the voltage swing into the load as that of a siingle amplifier.
This also quadruaples the current demand.
Which brings us to the bridged-parallel configuration meeting the demands of 4 times the current required to maintain double the voltage swing into the same load as that of a single amplifier.
Basic ohms law stuff here.
The amplifiers ability to produce (X) amount of voltage is dependent on the power supply feeding it.
Which has already been discussed earlier.
For an example if you have an amplifier that can produce A 14Vrms into 8 ohms (24.5watts) it will require a minimum supply voltage of +-20V at the filter output to maintain it.
Bridging the amplifier will allow you to produce a 28Vrms into 8 ohms (98 watts) PROVIDING that the amplifier can produce the current demand on the same +-20V (low voltage) supply.
When bridging an amplifiers each actualy sees half of the impedence load ,so if the load is 8 ohms each leg is seeing 4ohms.
When paralleling amplifiers, each actualy sees double that of the impedence load ,so if the load is 8 ohms each amplifier is seeing 16 ohms.
I hope this gives you better understanding voltage and current demands of the different configurations as I have been accused of not knowing my science.
If there is any thing that is unclear don't be afraid to ask and I will try to explain it in greater detail including the math equations that go along with it .jer
Thus, halving the impedence from 8 ohms to 4 ohms doubles the current demand for the same voltage swing that the amplifier has to produce.
Paralleling splits this current demand between the two amplifiers thus allowing double the output power into the 4 ohm load.
Doubling the output voltage into the same load quadruaples the output power.
Bridging two amplifiers together allows the amplifier to output double the voltage swing into the load as that of a siingle amplifier.
This also quadruaples the current demand.
Which brings us to the bridged-parallel configuration meeting the demands of 4 times the current required to maintain double the voltage swing into the same load as that of a single amplifier.
Basic ohms law stuff here.
The amplifiers ability to produce (X) amount of voltage is dependent on the power supply feeding it.
Which has already been discussed earlier.
For an example if you have an amplifier that can produce A 14Vrms into 8 ohms (24.5watts) it will require a minimum supply voltage of +-20V at the filter output to maintain it.
Bridging the amplifier will allow you to produce a 28Vrms into 8 ohms (98 watts) PROVIDING that the amplifier can produce the current demand on the same +-20V (low voltage) supply.
When bridging an amplifiers each actualy sees half of the impedence load ,so if the load is 8 ohms each leg is seeing 4ohms.
When paralleling amplifiers, each actualy sees double that of the impedence load ,so if the load is 8 ohms each amplifier is seeing 16 ohms.
I hope this gives you better understanding voltage and current demands of the different configurations as I have been accused of not knowing my science.
If there is any thing that is unclear don't be afraid to ask and I will try to explain it in greater detail including the math equations that go along with it .jer
Thank you so much for posting that. So if I want higher output into 8 ohm speakers then I would want to bridge the amp sections and provide them with a super beefy power supply. Right?
Basicaly you are getting the idea.
If you are bridging you can use a low voltage high current supply.
If you want a higher voltage swing(thus more output power) in any configuration you must then raise the supply voltage in order to increase the voltage swing,further more you must keep in mind that the current demand is increased by the same factor into the same load and you must stay within the power limits of the chip or chips them selves.
What is your target power rating and what impedence range are you planning to use?
Also do you already have a power transformer?If so what is the voltage ?
Knowing these factors would help make good considerations as to what would work for you. jer
If you are bridging you can use a low voltage high current supply.
If you want a higher voltage swing(thus more output power) in any configuration you must then raise the supply voltage in order to increase the voltage swing,further more you must keep in mind that the current demand is increased by the same factor into the same load and you must stay within the power limits of the chip or chips them selves.
What is your target power rating and what impedence range are you planning to use?
Also do you already have a power transformer?If so what is the voltage ?
Knowing these factors would help make good considerations as to what would work for you. jer
Another reason for choosing a bridged configuration is that even order distortion components (and DC offsets) cancel between the two amplifiers. This does not happen with a parallel configuration. The distortion spectrum from a LM3886/LM4780 is dominated by second and fourth order harmonics, indicating that this could be an useful improvement in clarity, as long as the improvement is not eaten up by the opposing distortion increase due to the increased current. That would work best into a high efficiency and/or high impedance loudspeaker, for instance a 16 ohm PHL 1130 midrange with 96 dB @ 1 W.
Even if it were 60 watts I would be happy but I would like something closer to 100 if possible and without adding much more complexity. I once had a Harman Kardon PM655 rated at 65 watts but it was capable of 45 amps and it sounded beautiful. I want this amp to have that "effortless" quality the HK had. I dont have speakers picked out but the target is 8 ohms since I plan on buying some larger Infinity or Klipsch bookshelves. I may even get a wild hair and go for some Kirksaeters, but I will need to get that one past the wife.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.