Assuming the amplifier are perfectly matched (this generally requires some care to be taken) you'll get more current handling ie. You will get higher power only into lower impedance loads.
If the amps can do 60W @ 8 ohms alone then in parallel they'll be able to do 120W @ 4 ohms
You will also get a higher damping factor and typically lower distortion.
If the amps can do 60W @ 8 ohms alone then in parallel they'll be able to do 120W @ 4 ohms
You will also get a higher damping factor and typically lower distortion.
Last edited:
They must match exactly or they will be fighting eachother. I doubt any well designed amp would "blow up" if it's done properly and you could mitigate this with small 5W resistors on the outputs before you sum them (say 0R25 or 0R33) but you'll lose a little D.F.
Directly paralleling would be a a recipe for disaster I would say but the devil is in the detail.
Put some numbers in...
lets say the amps when producing just 1 watt (so in round numbers around 3 volts rms output for 8 ohms) have an imbalance of just 50 millivolts in their output voltage. If these are amps with a low output impedance (lets say 0.05 ohm) then currents of 1 amp will be circulating between the two amplifiers. That is an order of magnitude more than the current the amps supply the load with.
Use of balancing resistors as mentioned in the post above would be the only sensible way to go and even then, connecting the outputs together at the end of a speaker cable run just seems all wrong to me.
Put some numbers in...
lets say the amps when producing just 1 watt (so in round numbers around 3 volts rms output for 8 ohms) have an imbalance of just 50 millivolts in their output voltage. If these are amps with a low output impedance (lets say 0.05 ohm) then currents of 1 amp will be circulating between the two amplifiers. That is an order of magnitude more than the current the amps supply the load with.
Use of balancing resistors as mentioned in the post above would be the only sensible way to go and even then, connecting the outputs together at the end of a speaker cable run just seems all wrong to me.
The result could be two burnt out amplifiers and perhaps ringing ears.
I recommend never doing this. As the outputs are shorted together large currents will flow due to the DC offset and the extremely low output resistance of amplifiers. If you are lucky they have speaker protection and one of them cuts out. Otherwise the output devices or fuses will blow. If both use a DC servo the mismatch might be very small and you get away with it (except immediately after switch-on before the servos have stabilized).
If you are really unlucky the setup will be unstable and oscillate at full power before failure - one amp can look like a very reactive load to another at some frequency - nominally matched amplifiers aren't going to have all their poles and zeroes the same in practice, so the HF responses are unlikely to be matched everywhere.
If you are lucky there is enough resistance in the speaker wires to enable the amps to share safely (if you remember to common at the speaker end). That means you have poor quality speaker cables!
You can deliberatately add output resistance before the commoning, losing some power and damping factor. However if one amp has temporary issues then the mismatch may again lead to something failing, perhaps cascading to the other amp.
The end result is likely to be the destruction of both amplifiers.
As others have pointed out, even the smallest mismatch will cause current to flow between the amplifiers and given the low output impedance of modern amplifiers even a fraction of a volt or a couple of degrees of phase difference could result in surprisingly large cross currents.
General rule in electronics ... you can connect some inputs together, but never outputs.
If these amplifiers are not already BTL amps -- where both sides of the speaker leads are active -- you might (maybe) be able to bridge them with some success.
And bridging nakamichi sr-4a is sort of outside of realm of practical project for a novice/hobbeist?
I added pre outs to sr-4a. Another gentleman here took time to look at the schematics and point out the location on the board. I don't know if he knows this or not but this was very much appreciated)
I added pre outs to sr-4a. Another gentleman here took time to look at the schematics and point out the location on the board. I don't know if he knows this or not but this was very much appreciated)
Bridging isn't too hard. You just need to invert the signal being fed to one channel. Usually with an inverting opamp, or if the amp is inverting, feeding the output via attenuation to match the gain into the other amplifier (of course this means higher distortion)
It only works if the amps are not already bridged (ie. Outputs are referenced to a common ground)
You get more potential voltage swing, so, more power into the same impedance.
But each channel is now effectively driving half the load impedance.
So if your speakers are already low for the rating, they will be straining. ie. for a 6 ohm load, each channel will "see" 3 ohms.
It only works if the amps are not already bridged (ie. Outputs are referenced to a common ground)
You get more potential voltage swing, so, more power into the same impedance.
But each channel is now effectively driving half the load impedance.
So if your speakers are already low for the rating, they will be straining. ie. for a 6 ohm load, each channel will "see" 3 ohms.
Another good reason NOT to parallel two amps -- in case you aren't already dissuaded by the prospect of Total Amplifier Death ..
It won't increase power!
Bet several people on here can lay it out better than me, but here goes ..
Whatever undistorted voltage swing (into that load) the amp is capable of, is going to be about the same, even if you were to lash up several in parallel.
On the other hand, bridging works great -- with a couple caveats (mentioned in earlier posts) -- and isn't too difficult, even for a novice.
- the amp outputs have to be ground referenced (not bridged)
- the amp has to be able to drive half whatever the speaker impedance is (without overheating or triggering the protect circuits)
If the Nak is already unhappy with your speakers it may not be a good option. Otherwise, come back on, maybe with a new thread, or Search this body of work -- likely somebody has already spelled it out.
Regards
It won't increase power!
Bet several people on here can lay it out better than me, but here goes ..
Whatever undistorted voltage swing (into that load) the amp is capable of, is going to be about the same, even if you were to lash up several in parallel.
On the other hand, bridging works great -- with a couple caveats (mentioned in earlier posts) -- and isn't too difficult, even for a novice.
- the amp outputs have to be ground referenced (not bridged)
- the amp has to be able to drive half whatever the speaker impedance is (without overheating or triggering the protect circuits)
If the Nak is already unhappy with your speakers it may not be a good option. Otherwise, come back on, maybe with a new thread, or Search this body of work -- likely somebody has already spelled it out.
Regards
Bra, I have learned something
But it's definitely very pointless to post a pointless comment after couple of pages of detailed and very informative explanations of issues. Achieves nothing .
Any chimp can do that. You seem to be just as useful.
Last edited:
You see the problem here is that he didn't know that when he asked the question. It's a rather interesting part of human behaviour... called "learning".
You should try it sometime.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.