Yes, but you will need load sharing resistors in series with the load from each amp and you would want the output of each amp to be the same voltage.
yeah thats what i thaught... so say if i have 2 identicle amps with the same signal, level and so on all i need is a .5R???
i have seen it done with Gainclones such as the LM3886 but i just wasnt sure if it was safe to do with a standard "solid state" amplifer...
so it is the the same principol as a gainclone then??
thnx haydn
i have seen it done with Gainclones such as the LM3886 but i just wasnt sure if it was safe to do with a standard "solid state" amplifer...
so it is the the same principol as a gainclone then??
thnx haydn
Parallel the inputs. Trim the DC offset to negligible levels and cross your fingers 🙂
Needless to say, test it with current limiting, gradually increasing PSU current. (lamp at AC mains is one way to limit current).
Needless to say, test it with current limiting, gradually increasing PSU current. (lamp at AC mains is one way to limit current).
The gain must also be as equal as possible.h3ad said:
Let me ask you: Do you plan to do this, and why?
i would like to help a friend run a single coil sub off a 4ch amplifier
my idea was parallel ch1+3 and ch2+4 then bridge them
thanx haydn
my idea was parallel ch1+3 and ch2+4 then bridge them
thanx haydn
You get points for trying to wring out every pea out of the amp, however extensive testing is required. Strapping channels is tricky, doing 4 is exponentially trickier.
i have tested the channels and they all have a DC offset within 1mv of each other if that is any help
regards haydn
regards haydn
probably not safe without connecting inputs "inside" the likely DC blocking Caps - then the DC offset has to be revisited
Yes, strap whatever is feasible, 1 mV is not bad... what are the rails of this amp? I'd use a minimum of 0.68R strapping resistors. Also is there an internal bridging switch that you will use or just feed one channel the inverted signal? Using an internal switch needs to be looked at for any issues with your proposed strapping topology.
i actually never bothered 2 check... its pretty low voltage tho... from memory the caps were 35v'ers... but 2 out of the 4 channels r permanently inverted... because ch 1 & 3 have thier GND marked as positive and ther live marked as negative where 2 & 4 r opposite so yeah
ill try it tonight with a few 1R's first just as failsafe... is that going 2 work properly?? just a little less power i would assume
regards haydn
ill try it tonight with a few 1R's first just as failsafe... is that going 2 work properly?? just a little less power i would assume
regards haydn
It should be ok, However do current limit it if you can while testing. (Lamp in series with AC mains is a favorite, begin with 50w gradually going to 300w bulb and testing the amp at 50-70% power eventually).
First test both parallel bridge channels independantly before testing the bridge set up.
Feed a sinewave to one ( 1 of 4) channel and bring the bulb to a dull amber, measure both output V and input mV, making sure the amp is stable and the brightness is also stable. Then repeat the same test after parallel bridge (2 in 4 channels) , the input and output V should be within 2-3% compared to single channel. Also by consequence the bulb will remain a dull amber, if any brighter, then u need to test for gain differences, DC offset and so on. Now test the other set of parallel bridge channels the same way.
35v rails should be much easier to work with than 80v rails.. 😉
First test both parallel bridge channels independantly before testing the bridge set up.
Feed a sinewave to one ( 1 of 4) channel and bring the bulb to a dull amber, measure both output V and input mV, making sure the amp is stable and the brightness is also stable. Then repeat the same test after parallel bridge (2 in 4 channels) , the input and output V should be within 2-3% compared to single channel. Also by consequence the bulb will remain a dull amber, if any brighter, then u need to test for gain differences, DC offset and so on. Now test the other set of parallel bridge channels the same way.
35v rails should be much easier to work with than 80v rails.. 😉
ok i will try that all tonight when i get a chance thnx for the great pointers😀
and also another question... because when u bridge channels it shares the load, for example an 8R speaker will appear as 4R on either channel... will it appear as say a 16R per ch when paralelled???
regards haydn
and also another question... because when u bridge channels it shares the load, for example an 8R speaker will appear as 4R on either channel... will it appear as say a 16R per ch when paralelled???
regards haydn
Kind of yes , the each paralleled channel will drive an 8 ohm load as if loaded with 16 ohms but when you bridge, it goes back to 8 ohms. 😉
But that 8 ohm load will tax the amp as if four 8 ohm loads were being driven simultaneously on 4 channels. i.e. if each channel can do 50 watts into 8 ohms and 75 watts into 4 ohms, If you drove all 4 channels at clipping, the amp's PSU will be delivering a little over 200 watts, that is the same power it will be delivering to a parallel bridged single 8 ohm load, i.e. little over 200 watts.
But that 8 ohm load will tax the amp as if four 8 ohm loads were being driven simultaneously on 4 channels. i.e. if each channel can do 50 watts into 8 ohms and 75 watts into 4 ohms, If you drove all 4 channels at clipping, the amp's PSU will be delivering a little over 200 watts, that is the same power it will be delivering to a parallel bridged single 8 ohm load, i.e. little over 200 watts.
thats exactly what i thaught
my theory was it would return it to 4x 8ohm load
so it wil basiclally extract every bit of "8ohm power" into 1 coil
thnx haydn
my theory was it would return it to 4x 8ohm load
so it wil basiclally extract every bit of "8ohm power" into 1 coil
thnx haydn
Yes, hence my comment on Post #7.
I once posted a query on the forum that brought some opposition form the forum on somewhat similar lines.
I had a 2 ohm load and an Adcom 555 to drive it. Since I worried about the risks of parallel bridge or the horror of using just 1 channel, I decide to do a serial bridge and lower the amp's rails to 50%, by wiring the trafo to run off 220v instead of the standard 110v. By halving the rails, the amp could deliver 2x the current needed, but now I could have 16 OP devices driving the load as opposed to the 8 devices for the single channel... Also 2x the heatsinking available.
Disclaimer... never tried the above in practice. 😀
Disclaimer #2: Your greatest risk will be running the parallel bridges reliably. You need to test the heck out of it. Serial bridge is easy.
I once posted a query on the forum that brought some opposition form the forum on somewhat similar lines.
I had a 2 ohm load and an Adcom 555 to drive it. Since I worried about the risks of parallel bridge or the horror of using just 1 channel, I decide to do a serial bridge and lower the amp's rails to 50%, by wiring the trafo to run off 220v instead of the standard 110v. By halving the rails, the amp could deliver 2x the current needed, but now I could have 16 OP devices driving the load as opposed to the 8 devices for the single channel... Also 2x the heatsinking available.
Disclaimer... never tried the above in practice. 😀
Disclaimer #2: Your greatest risk will be running the parallel bridges reliably. You need to test the heck out of it. Serial bridge is easy.
More thoughts:
Make sure all components are identical in the 2 channels being paralleled. e.g. if a signal or feedback cap (etc.) has been changed on one channel, replace both with same kind. (No need to mention the same about the active devices). Gain, speed, drift.... all need to be identical.
Make sure all components are identical in the 2 channels being paralleled. e.g. if a signal or feedback cap (etc.) has been changed on one channel, replace both with same kind. (No need to mention the same about the active devices). Gain, speed, drift.... all need to be identical.
- Status
- Not open for further replies.