I just hooked up two 15" speaker (8 Ohms each, in parallel) on an open baffle to a Dayton SA240 plate amp. At low gain settings everything is fine. If the gain is past 3 o'clock, and I turn off the main stereo amplifier, the subwoofers start oscillating very very wildly, at about 50 Hz. Does anybody know if the plate amp is broken, or do I need to add some sort of Zobel network? I use the hilevel inputs on the plate amp.
I have the gain at around 1 o'clock, but I'm afraid that it starts oscillating when I'm not around. I'm sure it'll fry the speakers.
Thanks for your help, Thomas
I have the gain at around 1 o'clock, but I'm afraid that it starts oscillating when I'm not around. I'm sure it'll fry the speakers.
Thanks for your help, Thomas
Do you have any way to check exactly at what frequency they are oscillating? (not 60hz)?
What is the resonant frequency of the drivers?
When I first used my QSC1450 on my subs, they hummed like crazy. It got louder with higher gain. Turns out my Sony HT receiver was picking up ground noise from the rear speaker cables that ran next to a the romex in the wall. Isolation transformers in the interconnects got rid of the problem. Tough to do with high level inputs.
Is it an uncontrolled oscillation or just a loud constant hum??
JJ
What is the resonant frequency of the drivers?
When I first used my QSC1450 on my subs, they hummed like crazy. It got louder with higher gain. Turns out my Sony HT receiver was picking up ground noise from the rear speaker cables that ran next to a the romex in the wall. Isolation transformers in the interconnects got rid of the problem. Tough to do with high level inputs.
Is it an uncontrolled oscillation or just a loud constant hum??
JJ
Thanks for your reply, JJ. In normal operation there is no hum, neither with hi volume/no signal, nor with low volume and signal. It only happens when I turn off the main amp that feeds the plate amp via hilevel (speaker) inputs, and the plate amp stays powered on. If I increase the gain on the plate amp it starts around 3 o'clock, very wildly, frequency could be less than 50/60, the cone from the subwoofers is moving back and forth until it hits the pol plate.
The speaker is a cheap Goldwood 1558, with fs 31Hz if I recall directly.
Laters, Thomas
The speaker is a cheap Goldwood 1558, with fs 31Hz if I recall directly.
Laters, Thomas
It could be the loss of input damping when your source is turned off.
Presumably you are taking the low impedance drive from your system.
If yes, try connecting 100 ohms across each plate amp input.
Presumably you are taking the low impedance drive from your system.
If yes, try connecting 100 ohms across each plate amp input.
Graham,
Thanks for looking into it. I tried your proposal, still having problems. Also tried Zobels, no help 😕
Laters, Thomas
Thanks for looking into it. I tried your proposal, still having problems. Also tried Zobels, no help 😕
Laters, Thomas
Subs are also super sensitive to ground loops.
You could try disconnecting the mains safety earth lead for your plate amp, and then rely upon earthing through your main system.
You could try disconnecting the mains safety earth lead for your plate amp, and then rely upon earthing through your main system.
Graham Maynard said:
Graham, that is a completely stupid and unsafe suggestion, you should be ashamed of yourself. Disconnecting safety earth is never an option.
Thomas, what happens if you ground the input, (use a shorting plug), does the plate amp still break into oscillation when you turn it up?
Hi,
cone displacement approaching Xlim and no mains hum.
The frequency is way below 50Hz/60Hz.
Even oscillation at around Fs (32Hz) will be heard as a loud drone.
I suspect the oscillation frequency is significantly below 20Hz.
Is this motorboating?
This is the very problem that almost everyone says does not happen with solid state amplifiers.
Can we get an estimate of the frequency?
cone displacement approaching Xlim and no mains hum.
The frequency is way below 50Hz/60Hz.
Even oscillation at around Fs (32Hz) will be heard as a loud drone.
I suspect the oscillation frequency is significantly below 20Hz.
Is this motorboating?
This is the very problem that almost everyone says does not happen with solid state amplifiers.
Can we get an estimate of the frequency?
Hi Pinkmouse,
Unfortunately some audio equipment designers still connect the mains safety earth directly to PSU/signal ground when it ought not be !!!
Surely you know this ?
When there is more than one safety earth connection to signal ground within a system you get nothing but problems, and then one or more of them must MUST be appropriately disconnected from that system ground, if not from internal primary related connections.
Pinkmouse, does your oscilloscope have its mains safety earth connected directly to signal ground when you connect it to audio amplifier circuits ?
I have suggested that Salinus try doing this with everything else connected, so please get off your high horse calling me 'stupid' until some more feedback comes through.
Cheers ........ Graham.
Unfortunately some audio equipment designers still connect the mains safety earth directly to PSU/signal ground when it ought not be !!!
Surely you know this ?
When there is more than one safety earth connection to signal ground within a system you get nothing but problems, and then one or more of them must MUST be appropriately disconnected from that system ground, if not from internal primary related connections.
Pinkmouse, does your oscilloscope have its mains safety earth connected directly to signal ground when you connect it to audio amplifier circuits ?
I have suggested that Salinus try doing this with everything else connected, so please get off your high horse calling me 'stupid' until some more feedback comes through.
Cheers ........ Graham.
Hi Pink & Graham,
I have read both your initial advice and final reply.
The first reading of Graham's advice did/does sound like disconnecting the safety earth.
The reply seems more like breaking the connection from audio ground to safety earth, with the chassis connected safety earth still intact.
On a temporary basis and for a user who is fully aware of the danger involved in disconnecting the audio ground from safety earth then the experiment is valid.
Graham, please add in the safety precautions for your suggestion.
However, on a different tack, many plate amps are double insulated and don't have a safety earth and as a result don't have an audio ground to safety earth connection.
I have read both your initial advice and final reply.
The first reading of Graham's advice did/does sound like disconnecting the safety earth.
The reply seems more like breaking the connection from audio ground to safety earth, with the chassis connected safety earth still intact.
On a temporary basis and for a user who is fully aware of the danger involved in disconnecting the audio ground from safety earth then the experiment is valid.
Graham, please add in the safety precautions for your suggestion.
However, on a different tack, many plate amps are double insulated and don't have a safety earth and as a result don't have an audio ground to safety earth connection.
mine does and it causes a whole series of different problems, often making results difficult to interpret.
Hi Andrew, thanks for your overview.
I have observed and cured earth loop problems in both car-sub and home based systems.
As you say many modern PSUs are double insulated, but an input ground can be chassis earthed, and then the chassis connected to mains safety earth, so step by step problem determination becomes necessary.
The fact that there is a problem (oscillation beyond a threshold gain+phase change which causes maximum amplitude oscillation to build up?) with the source turned 'off', is also likely to mean that the response is somewhat reactive to drive when the source is turned 'on'.
Cheers .......... Graham.
I have observed and cured earth loop problems in both car-sub and home based systems.
As you say many modern PSUs are double insulated, but an input ground can be chassis earthed, and then the chassis connected to mains safety earth, so step by step problem determination becomes necessary.
The fact that there is a problem (oscillation beyond a threshold gain+phase change which causes maximum amplitude oscillation to build up?) with the source turned 'off', is also likely to mean that the response is somewhat reactive to drive when the source is turned 'on'.
Cheers .......... Graham.
Mr. Pinkmouse
While this suggestion certainly violates Diy Audio's safety rules and regulations and we all are aware of this it should be tried if the reader is fully aware of all aspects. I will however disagree in your attitude and manor of stating it with regard to Graham. Proper respect to members by members and moderators alike should be a given.
With respect to if the suggestion is unsafe well every piece of audio equipment manufactured in the last 20 or so years not only sees ground thru the neutral but also the ground leg of the power cord and this usually leads to ground loop situations.
In commercial applications it is necessary to lift the ground by means of an adapter and this practice is fully approved by the system designers and audio consultants. Generally to get away from ground loop situations "ONE" piece in a rack system maintains its working ground and the fact that "ALL" pieces make a ground by contact or mounting in the rack transfers the ground to all pieces and the need to repeat the ground process thru the AC cord is not needed and or desired.
Note* With respect to tube gear....... I would have to agree 100% with policy of always maintaining a good mechanical and electronic ground.
burnedfingers said:
Read what I said, the advice is stupid, unless clarified as above by Andrew. I have respect for Graham as a designer, but general posts like that, without qualification, should not be here because of the danger inherent in that practice.
As for system designers lifting safety earth in pro applications, yes, it may be done as a bodge, but not by any engineer that I know that deserves the term professional. Usually it just means they can't be bothered to find the real problem.
Hi Pinkmouse,
Given that I too share your concerns about safety please advise how you expect (for someone who is unlikely to have safety isolating transformers on hand) to establish whether a phase shifting earth-loop is the problem without first isolating the connection creating that loop so that they can then know whether it needs to be addressed or not.
I did write advising Salinas to rely upon grounding through the rest of his main system, thus I see a repeat of your 'stupid' accusations as being most unhelpful to genuine progress.
We are not talking about bodges either. Salinas' problem might not even be Earth Loop at all, but the possiblity should be investigated, so your SAFE METHOD reply could be helpful instead of merely continuing to attack me.
Given that I too share your concerns about safety please advise how you expect (for someone who is unlikely to have safety isolating transformers on hand) to establish whether a phase shifting earth-loop is the problem without first isolating the connection creating that loop so that they can then know whether it needs to be addressed or not.
I did write advising Salinas to rely upon grounding through the rest of his main system, thus I see a repeat of your 'stupid' accusations as being most unhelpful to genuine progress.
We are not talking about bodges either. Salinas' problem might not even be Earth Loop at all, but the possiblity should be investigated, so your SAFE METHOD reply could be helpful instead of merely continuing to attack me.
Hi
The use of a Full Wave Bridge Rectifier might be helpful in this case.
Using a Block 35 Amp Bridge connect + and _ together feed this
to actual Ground. Then use each of the AC Poles ie one for Audio
Ground the other for Power Ground.
This provides three features
1. Isolation between ground returns which should overcome
propensity for ground loops occuring.
2. Isolation up to the Peak Inverse Voltage rating of the
rectifier block from Ground Voltages/Currents returning
from actual ground as can occur during lightning.
3. Connection via rectifier block diodes to safety earth
Hope this helps / Chris
The use of a Full Wave Bridge Rectifier might be helpful in this case.
Using a Block 35 Amp Bridge connect + and _ together feed this
to actual Ground. Then use each of the AC Poles ie one for Audio
Ground the other for Power Ground.
This provides three features
1. Isolation between ground returns which should overcome
propensity for ground loops occuring.
2. Isolation up to the Peak Inverse Voltage rating of the
rectifier block from Ground Voltages/Currents returning
from actual ground as can occur during lightning.
3. Connection via rectifier block diodes to safety earth
Hope this helps / Chris
Hey guys, I feel bad now that my post caused this stirrup. I'm halfway educated and know what the grounding is for, so my life was never in danger 😎
I found out the problem: The main speaker are hooked up to the HiLevel input from the sub plate amp; if the main receiver is turned off, and the main speaker act as microphones, if the gain is too high I run into this feedback loop
I did a simple test: main receiver off, tapped the main speakers, and the sub replicate the tap - both sub and main coax speaker are 15" and right next to each other on the baffle....
Thanks, Thomas
I found out the problem: The main speaker are hooked up to the HiLevel input from the sub plate amp; if the main receiver is turned off, and the main speaker act as microphones, if the gain is too high I run into this feedback loop
I did a simple test: main receiver off, tapped the main speakers, and the sub replicate the tap - both sub and main coax speaker are 15" and right next to each other on the baffle....
Thanks, Thomas
I'm with Andrew on motor-boating, unless I have missed some other info.
Put your scope probe on a rail and wind it up. If the rail is collapsing at the same frequency, that may be it.
A quick test is to hook in another decopling cap to double the existing value. See what happens.
Put your scope probe on a rail and wind it up. If the rail is collapsing at the same frequency, that may be it.
A quick test is to hook in another decopling cap to double the existing value. See what happens.
then wire up the low level input and isolate that feedback loop.Salinas said:I found out the problem: The main speaker are hooked up to the HiLevel input from the sub plate amp; if the main receiver is turned off, and the main speaker act as microphones, if the gain is too high I run into this feedback loop
I did a simple test: main receiver off, tapped the main speakers, and the sub replicate the tap - both sub and main coax speaker are 15" and right next to each other on the baffle
Graham Maynard said:
???
I wasn't attacking you, or anyone else for that matter, sorry if it seemed that way. I was just pointing out that removing safety grounds is a stupid thing to do when, as in this case, it wasn't the problem, and other options that might result in this LF oscillation had not been explored yet.
- Status
- Not open for further replies.