I'd be lying if I said I fully understood all of that. 
So in my lack of understanding I have a couple ignorant questions. Is the 200ASC a truly differential input amp or not? Would there be any benefit to using an OPA1632 to send a differential signal to the amp input when using a single ended source, or should I just connect the negative input to ground as they suggest?
So in my lack of understanding I have a couple ignorant questions. Is the 200ASC a truly differential input amp or not? Would there be any benefit to using an OPA1632 to send a differential signal to the amp input when using a single ended source, or should I just connect the negative input to ground as they suggest?
I didn't state it too well, sorry.
They are differential inputs in the 2000ASC, but not "high common mode impedance" types (meaning you could drive either [+] or [-] alone vs ground and the amplifier would still respond to the drive). In B&O's data book for the module, they show the input circuit, it's a single op-amp type diff amp, better than single ended, but not ideal.
Yeah, there'd still be a benefit provided both ends were driven equally and oppositely from low impedances. But it wouldn't be as good at doing that as if you had used a "THAT" type input circuit or a transformer. In that case, any currents running between chassis grounds or any ground differential voltages would be basically invisible to the amplifier, which is good. That much isolation may not be needed, though, depends on the setup. If you have no buzz/hum problems going single ended, all the differential stuff can be skipped.
The business about the "common mode impedances" isn't too complicated to understand. For the ideal simplest situation -- an input transformer-- imagine if you took a transformer and connected one terminal of its primary winding to a preamp's output hot, but with none of the other transformer terminals connected to anything at all. It would be just a lump of metal hanging on the line, and wouldn't do anything to the preamp's signal, because input signal current can't flow without having a complete path (back to the preamp's ground in this case). Current must pass, even if tiny, to send a signal and that requires a full loop.
You could do the same with either primary terminal as long as no other primary terminals connect anywhere. Or you could connect the primary terminals together to the same preamp drive -- same thing, just a dead hunk of metal hanging there, there's no current load path. That's what a high common mode impedance is -- individually no terminal has a way to pass current through to ground, so it acts like nothing at all to the driving circuit and has no effect on whatever the secondary is connected to. The secondary basically ignores any junk on ground. Only signal currents that go into one terminal of the primary and pass out the other can affect the secondary. The THAT circuit acts similarly (though for more complicated reasons).
Sorry for the long reply, I've been thinking about (and researching) grounding quite a bit lately -- I hate buzzing and hum.
They are differential inputs in the 2000ASC, but not "high common mode impedance" types (meaning you could drive either [+] or [-] alone vs ground and the amplifier would still respond to the drive). In B&O's data book for the module, they show the input circuit, it's a single op-amp type diff amp, better than single ended, but not ideal.
Yeah, there'd still be a benefit provided both ends were driven equally and oppositely from low impedances. But it wouldn't be as good at doing that as if you had used a "THAT" type input circuit or a transformer. In that case, any currents running between chassis grounds or any ground differential voltages would be basically invisible to the amplifier, which is good. That much isolation may not be needed, though, depends on the setup. If you have no buzz/hum problems going single ended, all the differential stuff can be skipped.
The business about the "common mode impedances" isn't too complicated to understand. For the ideal simplest situation -- an input transformer-- imagine if you took a transformer and connected one terminal of its primary winding to a preamp's output hot, but with none of the other transformer terminals connected to anything at all. It would be just a lump of metal hanging on the line, and wouldn't do anything to the preamp's signal, because input signal current can't flow without having a complete path (back to the preamp's ground in this case). Current must pass, even if tiny, to send a signal and that requires a full loop.
You could do the same with either primary terminal as long as no other primary terminals connect anywhere. Or you could connect the primary terminals together to the same preamp drive -- same thing, just a dead hunk of metal hanging there, there's no current load path. That's what a high common mode impedance is -- individually no terminal has a way to pass current through to ground, so it acts like nothing at all to the driving circuit and has no effect on whatever the secondary is connected to. The secondary basically ignores any junk on ground. Only signal currents that go into one terminal of the primary and pass out the other can affect the secondary. The THAT circuit acts similarly (though for more complicated reasons).
Sorry for the long reply, I've been thinking about (and researching) grounding quite a bit lately -- I hate buzzing and hum.
Thanks,
I've went over the datasheet for the THAT1200 and now have a better understanding of the difference between it and other devices like DRV134, OPA1632, THS4131, etc.
I have hundreds of OPA1632 so I think that is what I will use. I haven't noticed any hum or excessive noise from the 200ASC so spending money on the THAT1200 wouldn't be worth it to me.
I've went over the datasheet for the THAT1200 and now have a better understanding of the difference between it and other devices like DRV134, OPA1632, THS4131, etc.
I have hundreds of OPA1632 so I think that is what I will use. I haven't noticed any hum or excessive noise from the 200ASC so spending money on the THAT1200 wouldn't be worth it to me.
That's good to hear, I was kind of suspecting that with the Class II AC power on the modules, ground currents probably aren't likely to be too much of a problem.
BTW, you can make up the Jensen circuit using regular opamps, should you ever have the need. The circuit is patented, but for your own experimental use....
www.jensen-transformers.com/an/ingenaes.pdf
BTW, you can make up the Jensen circuit using regular opamps, should you ever have the need. The circuit is patented, but for your own experimental use....
www.jensen-transformers.com/an/ingenaes.pdf
Some day I'll start crying
Why is it that DIY'ers have this thing about making simple things complicated because they can?
(Alternatively, why is it that I just don't get it?)
Bwaslo mentioned it, but it could have come from anywhere.
Why do people keep talking about chassis currents and ground loops?
Contrary to things like my Cambridge Audio CD-player, the ICE-power modules are Class II units (according to the documentation).
Why is it that people insist on making ground connections, maybe even several?
Accoring to IEC 60065, the safety standard for audio/video equipment, it says:
>>>
2.6.2
CLASS II
design in which protection against electric shock does not rely on BASIC INSULATION only, but in which additional safety precautions, such as DOUBLE INSULATION or REINFORCED INSULATION, are provided, there being no provision for protective earthing or reliance upon installation conditions (see 3.3 of IEC 60536)
<<<
There's supposed to be one chassis-connection, if any. That's on the signal ground side.
A sufficiently closed metal enclosure is a faraday cage. Connecting signal ground to it ensures the potential of the F.C.
No multiple connections. No loops. Period.
I understand and agree with the safety concern when building a mains-connected PSU from scratch, but this is pre-build. Just ensure isolation and/or spacing between the PCB and metallic enclosure parts, and you're all set.
This brings me on to the common-mode signal issue:
How exactly would you have a common-mode issue without any other reference points (potentials) than the signal ground?
(...which you're not supposed to have since it's class II)
Then againm maybe it's just me.
bwaslo said:
Why is it that DIY'ers have this thing about making simple things complicated because they can?
(Alternatively, why is it that I just don't get it?)
Bwaslo mentioned it, but it could have come from anywhere.
Why do people keep talking about chassis currents and ground loops?
Contrary to things like my Cambridge Audio CD-player, the ICE-power modules are Class II units (according to the documentation).
Why is it that people insist on making ground connections, maybe even several?
Accoring to IEC 60065, the safety standard for audio/video equipment, it says:
>>>
2.6.2
CLASS II
design in which protection against electric shock does not rely on BASIC INSULATION only, but in which additional safety precautions, such as DOUBLE INSULATION or REINFORCED INSULATION, are provided, there being no provision for protective earthing or reliance upon installation conditions (see 3.3 of IEC 60536)
<<<
There's supposed to be one chassis-connection, if any. That's on the signal ground side.
A sufficiently closed metal enclosure is a faraday cage. Connecting signal ground to it ensures the potential of the F.C.
No multiple connections. No loops. Period.
I understand and agree with the safety concern when building a mains-connected PSU from scratch, but this is pre-build. Just ensure isolation and/or spacing between the PCB and metallic enclosure parts, and you're all set.
This brings me on to the common-mode signal issue:
How exactly would you have a common-mode issue without any other reference points (potentials) than the signal ground?
(...which you're not supposed to have since it's class II)
Then againm maybe it's just me.
Re: Some day I'll start crying
Thank you for the refreshing view. I will hook it up like the datasheet says anyway.
It seems the PCBs need a DIY sause for some of us. Probably because it are finished products 😉
Jennice said:
This brings me on to the common-mode signal issue:
How exactly would you have a common-mode issue without any other reference points (potentials) than the signal ground?
(...which you're not supposed to have since it's class II)
Then againm maybe it's just me.
Thank you for the refreshing view. I will hook it up like the datasheet says anyway.
It seems the PCBs need a DIY sause for some of us. Probably because it are finished products 😉
Re: Design considerations regarding class II, safety ground, chassis ground, signal groun
The negative input may not be used for single-ended signal sources. If so, it should be connected to signal ground as stated in the manual. Why would anyone let a signal input remain floating, only to be controlled by some not-well-specified internal input resistance?
I have tested the modules on cheap speakers (mainly to check for shipping damages, and my correct understanding of the documentation regarding connections.
Single-ended works fine. No hiss, no humm. Just music. I've tried it.
Oh, and Wim:
B&O hasn't always had everything under control. In their earlier days, some sarcastically referred to B&O (from the founders, Mr. Bang & Mr. Olufsen, if I recall correctly) as "Brum & Overstyring", which translates into something like "Humm & Clipping".
😀
wimdehaan said:
The negative input may not be used for single-ended signal sources. If so, it should be connected to signal ground as stated in the manual. Why would anyone let a signal input remain floating, only to be controlled by some not-well-specified internal input resistance?
I have tested the modules on cheap speakers (mainly to check for shipping damages, and my correct understanding of the documentation regarding connections.
Single-ended works fine. No hiss, no humm. Just music. I've tried it.
Oh, and Wim:
B&O hasn't always had everything under control. In their earlier days, some sarcastically referred to B&O (from the founders, Mr. Bang & Mr. Olufsen, if I recall correctly) as "Brum & Overstyring", which translates into something like "Humm & Clipping".
😀
Well, one way that loops could still happen is with capacitive coupling. For instance, I have a Hafler amp that has a two-terminal plug, no earth ground to the chassis. But it still picks up a buzz (though very weak) when connected to some other components.
Sure, there is no metal path to ground on the AC wiring, and an in an ideal world, nothing could form a loop. But the transformer will have capacitance from both AC lines to its core and mounting, just because they consist of two bits of metal with some dielectric (wire insulation) between them: a.k.a. a capacitor. So buzz energy high enough in frequency can still flow into the transformer, out the transformer core to the chassis, and then out the shield ground on the audio input cable. That's why turning the plug around can sometimes help get rid of low level buzzing.
Sure, there is no metal path to ground on the AC wiring, and an in an ideal world, nothing could form a loop. But the transformer will have capacitance from both AC lines to its core and mounting, just because they consist of two bits of metal with some dielectric (wire insulation) between them: a.k.a. a capacitor. So buzz energy high enough in frequency can still flow into the transformer, out the transformer core to the chassis, and then out the shield ground on the audio input cable. That's why turning the plug around can sometimes help get rid of low level buzzing.
Re: Re: Design considerations regarding class II, safety ground, chassis ground, signal groun
I don't know why anyone would want to, but I noticed no change in noise or the sound of the amp with it floating vs. being grounded.
I tried using an OPA1632 for a balanced signal into the amp but the only difference was the obvious 6dB of gain over single ended.
Jennice said:
I don't know why anyone would want to, but I noticed no change in noise or the sound of the amp with it floating vs. being grounded.
I tried using an OPA1632 for a balanced signal into the amp but the only difference was the obvious 6dB of gain over single ended.
bwaslo said:
To me, this sounds like a massive capacity, a resistor to power ground, or a supply voltage potential issue.
The capacitive coupling (and hence the loop path) is present regardless which way the power plug is orientated. Somewhere in the SMPS there's the high frequency transformer, but why one of its leads couples so much more to ground than the other makes me wonder. I would suspect something else to be resonsible for the coupling. Maybe one of the feed-back or EMC capacitors (which usually are only on the "live" wire between primary and secondary part of the PSU.)
The worse the design is with regards to EMC, the larger the capacitor tends to be, in an attempt to eliminate the dipole effect of the PSU. Especially SMPS's have a problem with this, which is why many chose a class I (grounded) SMPS. When grounded, it often has ground fed through to the output side, which makes passing mandatory EMC requirements much easier. (and safety, too.)
If I recall correctly, a leakage current of up to 3 mA is allowed (I don't have the standards with me at the moment, so bear with me).
The Hafler I was talking about uses a big iron 60Hz transformer, and one end of the winding is almost sure to be closer to the core than the other.
Probably the SMPS won't have anything like that much capacitance (and the core is ferrrite, besides), so yeah, it's probably a non issue except for maybe RFI.
Probably the SMPS won't have anything like that much capacitance (and the core is ferrrite, besides), so yeah, it's probably a non issue except for maybe RFI.
bwaslo said:
That sounds very reasonable, but I would not think that there is a good capacitive coupling at 50/60 Hz. Inductive maybe, but capacitive?
If the Hafler is this "reverse-the-plug-to-kkill-the-humm" issue, does it have any filters on the mains side, any capacitors between chassis and either mains side or secondary side?
...or caps between mains and secondary side?
This is making me curious.
No, rude and crude, just wires through a switch, a fuse, some series thermal breakers and into the transformer.
The sound isn't so much a hum but a buzz (and at very low levels, not audible unless right up against the speaker or when making electrical measurements at the amp output). Upper harmonics of hum, I imagine.
I've seen similar things happen with old tube preamps (PAS2X type).
The sound isn't so much a hum but a buzz (and at very low levels, not audible unless right up against the speaker or when making electrical measurements at the amp output). Upper harmonics of hum, I imagine.
I've seen similar things happen with old tube preamps (PAS2X type).
Hi,
I have also purchased a few of these modules, and I was planning on using one of them to power a sub (Peerless XLS 10" with 10" passive).
The problem is that once everything is set up, and the amplifier is connected to the sub, it barely makes any sound, and I really have to crank up the volume to get any decent output from the sub.
I am using a BOSOZ preamp for the main speakers, and I get the sub output by summing the lowpass of the left and right active crossovers, using 2 resistances (10K). I also tried connecting the amp straight to the subwoofer output of a Rotel HT processor, but it gives the same results.
Is there any way of increasing the amplifier's gain? Or perhaps any circuits that would do a good job increasing the gain when inserted between the preamp and the 200ASC module?
Thanks for any help.
Paul
I have also purchased a few of these modules, and I was planning on using one of them to power a sub (Peerless XLS 10" with 10" passive).
The problem is that once everything is set up, and the amplifier is connected to the sub, it barely makes any sound, and I really have to crank up the volume to get any decent output from the sub.
I am using a BOSOZ preamp for the main speakers, and I get the sub output by summing the lowpass of the left and right active crossovers, using 2 resistances (10K). I also tried connecting the amp straight to the subwoofer output of a Rotel HT processor, but it gives the same results.
Is there any way of increasing the amplifier's gain? Or perhaps any circuits that would do a good job increasing the gain when inserted between the preamp and the 200ASC module?
Thanks for any help.
Paul
is it wired correctly? I have a pair of modules hooked up to a pair of magneplanar mgIIIs and gettin fairly loud playback levels
hi paul,
thats how I have it wired for single ended. have tried it with your main speakers? bosoz pre - ice - main speakers
thats how I have it wired for single ended. have tried it with your main speakers? bosoz pre - ice - main speakers
I haven't tried it with the BOSOZ and my main speakers, but I did try it with a portable CD player and an old speaker when testing it, and it could play pretty loud when the CD player's volume was turned up.
I'm now starting to think that the problem might be with the summing of the 2 signals before the ice module.
Paul
I'm now starting to think that the problem might be with the summing of the 2 signals before the ice module.
Paul
- Status
- Not open for further replies.