Hi all,
I have 8 * lm3886 with 25-0-25 800va supply ( +_33v no load ) diving a set of linkwitz orions using the appropriate active crossver based on OPA2134 op amps.
I have notice some "breathing" on the bass drivers on some CD's on medium/heavy bass material. i.e a movement of the drivers varying between 1-7 Hz ( this is *very* subjective ! ) of around 2-3 cm displacement. Ok, no problem except if I over drive the XLS10 woofers.
next....an earth return on the input to the RH tweeter amp cracks whilst I am fiddling with things ( these were solid wires at the time ) and cruises over the back of the PCB, duly blows one of the fuses on the PCB +ve supply and creams via low frequency/DC the SEAS Millenium tweeter in the process...<large groan>
replace Millenium cone assembly ...
ok, so now I have hooked all back up ( minus tweeter in RHS ) and found that the "breathing" I noticed on some tracks earlier is now replicated to some degree in the midrange speaker on the RHS. <another groan>, it seems to track the movements in the bass drivers.
If I disconnect the woofer inputs to the amps, the "breathing" disappears in the midrange,
I have experimented with changing inputs/outputs etc on the lm3886's but the problem seems to track the amplifier boards, i.e. it doesn't seem to be coming from the electronic crossover, although the possibility of some interaction cannot be entirely eliminated.
All is ok on the LHS (apart from the bass breathing), the RHS sounds ok (but I haven't hooked up the tweeter as yet ...scared you see!)
DC offsets are 0-1 mV on all amps, ac output on no signal is 0-20mV (I have some injected noise from the crossover .... to be solved at a later date !
I haven't as yet pulled the whole thing onto the workbench and checked any more deeply as this "breathing" in the midrange is a little supprising because the mid/bass is 24 db/octave so there should be *no* low frequency stuff on the midrange,
at this point I would be
a) guessing,
b) trying to analyze and find a logical solution
Any input would be appreciated, I don't mind doing the work (I mapped the whole damn crossover with a signal generator to find one soldering issue!)
So any feedback would be appreciated,
If the info I have given is too vague or doesn't make sense, please feel free to call me all the required adjectives <grin> and ask me to investigate/clarify,
ok,
enough rambling for now and thanks for your time,
cheers,
Ray
I have 8 * lm3886 with 25-0-25 800va supply ( +_33v no load ) diving a set of linkwitz orions using the appropriate active crossver based on OPA2134 op amps.
I have notice some "breathing" on the bass drivers on some CD's on medium/heavy bass material. i.e a movement of the drivers varying between 1-7 Hz ( this is *very* subjective ! ) of around 2-3 cm displacement. Ok, no problem except if I over drive the XLS10 woofers.
next....an earth return on the input to the RH tweeter amp cracks whilst I am fiddling with things ( these were solid wires at the time ) and cruises over the back of the PCB, duly blows one of the fuses on the PCB +ve supply and creams via low frequency/DC the SEAS Millenium tweeter in the process...<large groan>
replace Millenium cone assembly ...
ok, so now I have hooked all back up ( minus tweeter in RHS ) and found that the "breathing" I noticed on some tracks earlier is now replicated to some degree in the midrange speaker on the RHS. <another groan>, it seems to track the movements in the bass drivers.
If I disconnect the woofer inputs to the amps, the "breathing" disappears in the midrange,
I have experimented with changing inputs/outputs etc on the lm3886's but the problem seems to track the amplifier boards, i.e. it doesn't seem to be coming from the electronic crossover, although the possibility of some interaction cannot be entirely eliminated.
All is ok on the LHS (apart from the bass breathing), the RHS sounds ok (but I haven't hooked up the tweeter as yet ...scared you see!)
DC offsets are 0-1 mV on all amps, ac output on no signal is 0-20mV (I have some injected noise from the crossover .... to be solved at a later date !
I haven't as yet pulled the whole thing onto the workbench and checked any more deeply as this "breathing" in the midrange is a little supprising because the mid/bass is 24 db/octave so there should be *no* low frequency stuff on the midrange,
at this point I would be
a) guessing,
b) trying to analyze and find a logical solution
Any input would be appreciated, I don't mind doing the work (I mapped the whole damn crossover with a signal generator to find one soldering issue!)
So any feedback would be appreciated,
If the info I have given is too vague or doesn't make sense, please feel free to call me all the required adjectives <grin> and ask me to investigate/clarify,
ok,
enough rambling for now and thanks for your time,
cheers,
Ray
Just a wild guess. Is this behaviour related to any fluctuation of your mains voltage? I'd stick a meter on the mains in your amp to check, but do be careful. 
LF instability?
IMHO it might not be due to mains fluctuation because you must be using split supplies ( +/- supplies) with an independent ground. So this might not be the reason.
However it could be the grounding scheme. I have seen a friend blow a tweeter with very low LF in the tweeter and no bass on it. We never solved the problem but I could see the ribbon tweeter move in and out at LF.
Maybe something like that. Make sure all grounds are separated and tied at only one point.
Cheers.
IMHO it might not be due to mains fluctuation because you must be using split supplies ( +/- supplies) with an independent ground. So this might not be the reason.
However it could be the grounding scheme. I have seen a friend blow a tweeter with very low LF in the tweeter and no bass on it. We never solved the problem but I could see the ribbon tweeter move in and out at LF.
Maybe something like that. Make sure all grounds are separated and tied at only one point.
Cheers.
Off the wall idea, but are you near a naval base? 🙂
Communication systems for submerged submarines use frequencies between about 6Hz to .01 Hz. Perhaps you might have a screening problem...
Communication systems for submerged submarines use frequencies between about 6Hz to .01 Hz. Perhaps you might have a screening problem...
I tend to agree with AM. I'm in the process of building actively crossed speakers with four home built LM3886 amps (not Orion). Here is what I would try if it was my problem:
1. Disconnect all the amps except one woofer and one tweeter that is having the problem. This is an attempt to minimize the number of variables you have to deal with.
2. While the problem is happening, disconnect the woofer driver from the amp output (not the amp input as you described originally). If this stops the problem you might infer that you have a bad ground and the current flowing through the amp output was dragging a bad ground up and down.
3. You could also check for this by seeing if the "breathing" excursion increases as you increase the volume of the signal. More volume = more current = more ground variation = more "breathing"?
3. Reconnect the driver (if that had stopped the problem) and measure the voltage difference between the ground at the output of the midrange amp and the power supply ground at the center tap of the transformer. See if you can see your slowy modulating signal showing up instead of zero volts between these two point.
4. Filter out very low frequencies from the input to you woofer amplifier. Does the Orion active X-over have a low frequency rumble filter stage? If not, a series capacitor should be enough to filter out 1-7 hz.
My guess would be that some of your CDs have some very low frequency content. This is directly causing the breathing in you woofer. There is a bad connection in your grounding that is causing the ground on your midrange amp to move as you draw current through your woofer amp. This moving ground is impressing the low frequency signal on you midrange driver.
Good luck!
1. Disconnect all the amps except one woofer and one tweeter that is having the problem. This is an attempt to minimize the number of variables you have to deal with.
2. While the problem is happening, disconnect the woofer driver from the amp output (not the amp input as you described originally). If this stops the problem you might infer that you have a bad ground and the current flowing through the amp output was dragging a bad ground up and down.
3. You could also check for this by seeing if the "breathing" excursion increases as you increase the volume of the signal. More volume = more current = more ground variation = more "breathing"?
3. Reconnect the driver (if that had stopped the problem) and measure the voltage difference between the ground at the output of the midrange amp and the power supply ground at the center tap of the transformer. See if you can see your slowy modulating signal showing up instead of zero volts between these two point.
4. Filter out very low frequencies from the input to you woofer amplifier. Does the Orion active X-over have a low frequency rumble filter stage? If not, a series capacitor should be enough to filter out 1-7 hz.
My guess would be that some of your CDs have some very low frequency content. This is directly causing the breathing in you woofer. There is a bad connection in your grounding that is causing the ground on your midrange amp to move as you draw current through your woofer amp. This moving ground is impressing the low frequency signal on you midrange driver.
Good luck!
I have a CD with 1 Hz transient on drum beats. It's Linda Rondstadt, Winter Light a song with heavy beating in the beginning (don't remember the name of the song). This results in big movements in the speaker. What has the recording engineer done?dhenryp said:
Thanks for the informative replies guys,
much appreciated.
The excursion is related to volume level ( in both woofs and mid ) so I will follow the suggestions of AM and dhenryp and see what I find. I had never thought of it as an earth problem.
The crossover does have a 50Hz lowpass filter so I will also enable that and see what the result is.
Mains Power supply is stable and I don't live near a naval base <smile>
cheers, Ray
much appreciated.
The excursion is related to volume level ( in both woofs and mid ) so I will follow the suggestions of AM and dhenryp and see what I find. I had never thought of it as an earth problem.
The crossover does have a 50Hz lowpass filter so I will also enable that and see what the result is.
Mains Power supply is stable and I don't live near a naval base <smile>
cheers, Ray
Hi guys,
well it *was* an earth problem, after checking PSU differences between PSU ground and amp grounds and only found 2-15mV on all channels, hmmm.....
*Then* found that I hadn't connected the -ve outputs on one channel ( 4 amps ) to ground when I rerouted all the speaker wiring ( I won't bore you with the details <groan> ) so no wonder the tweeter gave up the ghost !, feeling a little foolish but no major damage <smile>
thanks for the help,
cheers and regards,
Ray
well it *was* an earth problem, after checking PSU differences between PSU ground and amp grounds and only found 2-15mV on all channels, hmmm.....
*Then* found that I hadn't connected the -ve outputs on one channel ( 4 amps ) to ground when I rerouted all the speaker wiring ( I won't bore you with the details <groan> ) so no wonder the tweeter gave up the ghost !, feeling a little foolish but no major damage <smile>
thanks for the help,
cheers and regards,
Ray
- Status
- Not open for further replies.