Hey, all.
I apologize in advance, 'cause I'm a green noob. Greenest of the green, in fact. But I'm tearing my hair out and this looked like the right place to find experts in this field. I hope you don't mind me crashing your party. Also, if I didn't post this in the right place, feel free to move it...given the topic it probably could have gone into one of the amplifier sections.
This is perhaps an unusual situation. We've got some devices that have their own built-in speakers and aren't normally wired up to an external amp (or designed to be). Desktop telephones, actually. In this specific case, IP phones (Polycom SoundPoint IP 335). The built-in speakerphone isn't loud enough (trying to double the phones as a PA system), so we want to hook up an external speaker to many of these phones. I've got an adapter (PC Headset to RJ9/RJ10/RJ22 Headset Plug Adapter - most compatible) that allows me to plug regular headphones (T/R miniplug jack) into the headset jack (RJ9), and I have been testing both with headphones as well as with various cheap, amplified computer speakers. So we're talking consumer-grade unbalanced audio here, where the source is definitely not line-level.
The issue is that when using powered speakers, as long as the phone is on-hook (no active call), there is a terrible 60Hz hum that comes through the speakers. During an actual call, however, even when there is silence on the line, the buzz is completely gone. If I use headphones, no buzz is present (of course), and if I am using speakers that have the option to run their amp off of batteries instead of an AC/DC transformer, when I switch to batteries, the buzz is gone then, too. Cheapie ground loop isolators between the phones and the speakers help (maybe cut about 75% of the noise out) but do not solve the problem completely.
I found one other post in these forums that struck me as being a similar problem, except the poster there is dealing with the headphone output of a smartphone (http://www.diyaudio.com/forums/analog-line-level/244945-ground-loop-type-issue-android-phones-into-class-d-amps.html). And even before I ran across that thread, I talked to an EE friend of mine who suggested a similar course of action (put a resistor between signal and shield/ground). I'm not quite sure I understand the theory behind doing this, but we gave it a shot and it does help. The lower the resistance, the more it helps, which I guess makes sense. So I tried bridging signal and shield directly without a resistor. The buzz went completely away, and everything appears to work fine.
At this point I'm left with some questions.
First, what exactly is happening? I take it when there is no call, the phone's built-in headset amp is off. Does this result in a ground lift on the phone/audio source side, and does that explain the terrible ground-loop-type hum I'm experiencing? And why does a ground loop isolator/transformer only take care of *most* of the problem?
Second, is there any risk of physical harm to either the source or to the speaker amp by shorting signal to shield without using a resistor? (Probably not while the phone is not on a call, but I'm worried about when the phone's headset amp is on.) I can't find a resistance value that completely takes care of the hum, so I'd prefer to do it this way if there is no risk.
Third, I know this is a DIY forum and so I should probably feel ashamed to ask this, but...are there any ready-made, commercial solutions that accomplish what I want to do? Something like an in-line dummy load of sorts? We're talking 40-50 phones here, so I'd just as soon buy a bunch of pre-assembled bits and run around plugging them in rather than take the time to sit down and crank a bunch of custom cables out. I found e.g. things like this -- Etymotic ER4P to ER4S Resistor Adaptor 3 5mm Plug | eBay -- but I believe that these just have resistors wired in serial along the signal path (not between L and R signal and shield), so they won't actually do what I want.
Thanks ever so much for any help or direction you can provide!
-- Nathan
I apologize in advance, 'cause I'm a green noob. Greenest of the green, in fact. But I'm tearing my hair out and this looked like the right place to find experts in this field. I hope you don't mind me crashing your party. Also, if I didn't post this in the right place, feel free to move it...given the topic it probably could have gone into one of the amplifier sections.
This is perhaps an unusual situation. We've got some devices that have their own built-in speakers and aren't normally wired up to an external amp (or designed to be). Desktop telephones, actually. In this specific case, IP phones (Polycom SoundPoint IP 335). The built-in speakerphone isn't loud enough (trying to double the phones as a PA system), so we want to hook up an external speaker to many of these phones. I've got an adapter (PC Headset to RJ9/RJ10/RJ22 Headset Plug Adapter - most compatible) that allows me to plug regular headphones (T/R miniplug jack) into the headset jack (RJ9), and I have been testing both with headphones as well as with various cheap, amplified computer speakers. So we're talking consumer-grade unbalanced audio here, where the source is definitely not line-level.
The issue is that when using powered speakers, as long as the phone is on-hook (no active call), there is a terrible 60Hz hum that comes through the speakers. During an actual call, however, even when there is silence on the line, the buzz is completely gone. If I use headphones, no buzz is present (of course), and if I am using speakers that have the option to run their amp off of batteries instead of an AC/DC transformer, when I switch to batteries, the buzz is gone then, too. Cheapie ground loop isolators between the phones and the speakers help (maybe cut about 75% of the noise out) but do not solve the problem completely.
I found one other post in these forums that struck me as being a similar problem, except the poster there is dealing with the headphone output of a smartphone (http://www.diyaudio.com/forums/analog-line-level/244945-ground-loop-type-issue-android-phones-into-class-d-amps.html). And even before I ran across that thread, I talked to an EE friend of mine who suggested a similar course of action (put a resistor between signal and shield/ground). I'm not quite sure I understand the theory behind doing this, but we gave it a shot and it does help. The lower the resistance, the more it helps, which I guess makes sense. So I tried bridging signal and shield directly without a resistor. The buzz went completely away, and everything appears to work fine.
At this point I'm left with some questions.
First, what exactly is happening? I take it when there is no call, the phone's built-in headset amp is off. Does this result in a ground lift on the phone/audio source side, and does that explain the terrible ground-loop-type hum I'm experiencing? And why does a ground loop isolator/transformer only take care of *most* of the problem?
Second, is there any risk of physical harm to either the source or to the speaker amp by shorting signal to shield without using a resistor? (Probably not while the phone is not on a call, but I'm worried about when the phone's headset amp is on.) I can't find a resistance value that completely takes care of the hum, so I'd prefer to do it this way if there is no risk.
Third, I know this is a DIY forum and so I should probably feel ashamed to ask this, but...are there any ready-made, commercial solutions that accomplish what I want to do? Something like an in-line dummy load of sorts? We're talking 40-50 phones here, so I'd just as soon buy a bunch of pre-assembled bits and run around plugging them in rather than take the time to sit down and crank a bunch of custom cables out. I found e.g. things like this -- Etymotic ER4P to ER4S Resistor Adaptor 3 5mm Plug | eBay -- but I believe that these just have resistors wired in serial along the signal path (not between L and R signal and shield), so they won't actually do what I want.
Thanks ever so much for any help or direction you can provide!
-- Nathan
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.