Topic: AT822 into balanced preamp how? (plus, any way to mod it balanced? or phantom?)

[spaceboy_psy]

So I plumped for a lovely AT822, many thanks to DSatz et al. convincing me to do the honourable thing!

It's coming with the XLR->3.5stereoplug cable, and an XLR->2x 3.5monoplug cable, with 1/4" adaptors.

I've got a Microtrack II, and I've read that for going straight into that it'll be better to use the 1/8" than 1/4" as the mic's unbalanced and the MTII doesn't like that.

What about if/when I get a balanced input preamp? (thinking about the UA-2) Here - http://taperssection.com/index.php/topic,82553.0.html - Mr Church seems to imply that it's best to short the inputs of the device itself, but is there not a way to do it with a cable...?

OOOORRRRRR, how hard would it be to mod the AT822 and make it balanced?? Google didn't throw anything up, is it impossible? Surely you just need something that will reverse the phase of the signal? And I wondered about phantom power - if you could mod the AT822 to use phantom power, would it improve the sound at all? Sensitivity, signal to noise, anything? If so a balanced/phantom-powered AT822 would be perfect to run with both the UA-2 -> MTII and also straight into the MTII's TRS ins, and make a pretty wholesome setup...

Cheers in advance, I really can't wait to hear your thoughts!

[Roving Sign]

I wouldnt stress too much about the balanced aspect...unless you plan on running 100 feet of cable... - just not much to gain there.

Just use this mic as-is...

If so a balanced/phantom-powered AT822 would be perfect...

Also known as an 825

Not sure what the issues might be with the microtrack...I'd make specific post about that..."AT822 > Microtrack II: HOW?"

[spaceboy_psy]

Ah, I just had a likely-obvious thought! (yay me): doing everything balanced would get rid of noise picked up in the cable and hardware, yes? though that can be easily limited by using the shortest cables possible... BUT, can it reduce noise coming from the actual Mic? ie. improve signal/noise ratio?? I was kind of half thinking it might, but now I actually use a couple more synapses on it, I realise it probably wouldn't! Can anyone confirm this?

Yup yup yup, an 825 would be nice, but considering I got the 822 for better than half the price of an 825, I'm willing to work this mess out 

[spaceboy_psy]

Cool ^_^ cheers

Balanced AT822 mod anyone..?

[DSatz]

roving_sign, the AT825 has other differences from the AT822 in addition to balanced outputs; though the same capsules are used in both microphones, the high frequency response of the AT825 is considerably smoother. A couple of years ago I asked some of the company's chief engineers about this difference at the AES convention, and it turned out that they weren't very familiar with either model because both were designed before almost anyone who works there now had been hired. Both these mikes have been in their catalog for a very long time, and I'm glad to see that they've recently introduced some more up-to-date stereo recording microphones.

Just in general--I haven't had time yet to read this thread so someone else may well have said this already--when connecting an unbalanced, self-powered (or non-powered) microphone to a balanced input, the ideal approach is to balance the connection at the microphone output by connecting the unbalanced output to one modulation lead and imitating the microphone's output impedance as precisely as possible and connecting that impedance between ground and the other modulation lead. It isn't necessary to use a transformer or active circuitry at the microphone end to create balance; all that's needed is for the impedance to ground to be equal between the two legs of the circuit.

--best regards

[illconditioned]

roving_sign, the AT825 has other differences from the AT822 in addition to balanced outputs; though the same capsules are used in both microphones, the high frequency response of the AT825 is considerably smoother. A couple of years ago I asked some of the company's chief engineers about this difference at the AES convention, and it turned out that they weren't very familiar with either model because both were designed before almost anyone who works there now had been hired. Both these mikes have been in their catalog for a very long time, and I'm glad to see that they've recently introduced some more up-to-date stereo recording microphones.

Just in general--I haven't had time yet to read this thread so someone else may well have said this already--when connecting an unbalanced, self-powered (or non-powered) microphone to a balanced input, the ideal approach is to balance the connection at the microphone output by connecting the unbalanced output to one modulation lead and imitating the microphone's output impedance as precisely as possible and connecting that impedance between ground and the other modulation lead. It isn't necessary to use a transformer or active circuitry at the microphone end to create balance; all that's needed is for the impedance to ground to be equal between the two legs of the circuit.

--best regards

Thanks for the in-depth knowledge.

By the way, do you have an opinion on the capsules in the AT822/825?  How do you think they compare to the newer single-point stereo mics AT is producing?  From the pictures, it looks like AT is using something different, perhaps the ones in the 4033 model?

Oh yeah, to all you hackers out there, it is quite easy to rip the head off the AT822/825 mic, add a 4.7k resistor to each capsule and power it with plug-in-power, from something like an Edirol R09.  I've got three of these mics, two AT822 were chopped, and the third, an AT825, was left for phantom powering.  To my ear, these mics are OK, but I prefer other mics now.  Hmm.  So why do I keep these?  Well, two of the three are loaned out...

  Richard

[spaceboy_psy]

when connecting an unbalanced, self-powered (or non-powered) microphone to a balanced input, the ideal approach is to balance the connection at the microphone output by connecting the unbalanced output to one modulation lead and imitating the microphone's output impedance as precisely as possible and connecting that impedance between ground and the other modulation lead. It isn't necessary to use a transformer or active circuitry at the microphone end to create balance; all that's needed is for the impedance to ground to be equal between the two legs of the circuit.

Wow, right, let me try and get my head around this!

Just read the wikipedia Balanced Audio article - should have done that sooner 

By the way, I just grabbed a UA-5 off ebay (whoooo!  ), so that's what we're working with, that and the microtrack.

I'm a little confused, will the UA-5/MTII will be subtracting Hot and Cold signals? so we need to reverse the polarity of one signal coming from the AT822? Is this the role of the 'modulation lead'? I couldn't find any info on what this might be. Or are you saying that it's not necessary to do any inverting, just match impedances? Where does the 200 Ohm resistor go exactly?

the signal lead is going to ground through the AT822 (200 Ohms)? And we want the shield to just run to the ground of the preamp. So we need to split the signal at some point between each capsule and the amp input, and make the cold signal go to ground through a 200 Ohm resistor at the preamp end?

That's the best I can do for now, but I'm sure I'm probably way off!    Thanks for bearing with me

it is quite easy to rip the head off the AT822/825 mic, add a 4.7k resistor to each capsule and power it with plug-in-power, from something like an Edirol R09.  I've got three of these mics, two AT822 were chopped, and the third, an AT825, was left for phantom powering.

Just so I'm clear, this is preferable to battery power because (apart from the mic being smaller of course), it's getting more voltage, ~5V rather than 1.5V, and this improves performance, like a battery box does...?

Now I think about the small voltages we're dealing with here, I suppose it must be impossible to run the AT822 on a comparatively massive 48V - would it just fry the electronics?

Wheee, this is fun, can't wait to get my AT822, and UA-5!!! whoo whoo whoo whoo 

[DSatz]

spaceboy_psy, actually it is far better to balance the signal at the source (at the microphone outputs). That allows you to use balanced cable to connect the signal to the balanced input of your preamp/mixer/recorder. That way you'll get the noise immunity across that length of cable, which is where it mostly matters.

Any differential input does the trick of sensing the moment-to-moment difference between two wires, neither of which is the signal ground. Each of those two wires (sorry, I lapsed into jargon and wrote "modulation leads" before) has some particular impedance between it and ground. A balanced circuit has a differential input that's fed by two wires, both of whose impedance is identical to each other. The function of the differential input circuit is to subtract one from the other electronically, which tends to cancel any noise that's due to interference--though in practice that also depends on the precise balance of the cable's impedance and some other factors.

--Before I forget to mention it, occasionally someone does connect an AT822 to phantom powering, which burns out its circuitry in an instant. My wife's voice teacher did it a few years ago so I arranged the repair for him; when I called A-T to arrange the repair, it was like they'd heard the story many times before. The repair cost isn't huge but it's still a mistake worth avoiding.

--I'm assuming that the UA-5 has balanced mike-level inputs and balanced line-level outputs. If that's wrong, please let me know; I'm actually not sure what a UA-5 is. The MicroTrack recorder I know has line-level inputs and can record either balanced or unbalanced signals, but it has severe noise (internal interference) problems when unbalanced signals are used, so it's not a good idea to drive it that way.

[kirk97132]

Edirol UA-5 has balanced mic inputs(xlr/trs) with unbalanced line level outputs(rca/1/4") or digital(coax/optical).  Sorry picture isn't a little better.  kirk

[spaceboy_psy]

UA-5: http://www.core-sound.com/ua-5.html

balanced mic inputs indeed.

DSatz, thank you very much for your help, but I'm afraid I've been getting rather confused as to what you think I should actually, physically, be doing! But after some reading, I'm wondering, are you simply talking about leaving the shield disconnected at the microphone end...?

ie. illustration no. 14 here: http://www.rane.com/note110.html?

or this? ("preamp" referring to an unbalanced signal which is going to the balanced Microtrack inputs, so the same as we're talking about I think):

This is using special unbalanced to balanced adapter that takes unbalanced preamp common ground wired to TRS (-minus) ring connections, sending EACH channel's +hot signal to TRS tips, and leaves the common TRS sleeve cable ground shield connections UNCONNECTED, preferably near the preamplifier output jacks.

So this is making me imagine a setup like this:

AT822L signal -> TRS tip L -> UA-5 L
disconnected <- cable shield L <- TRS sleeve L -> UA-5 L
AT822 ground  -> TRS ring L -> UA-5 L
                    -> TRS ring R -> UA-5 R
disconnected <- cable shield R <- TRS sleeve R -> UA-5 R
AT822R signal -> TRS tip R -> UA-5 R

Is this anything like what you're talking about?? I hope I'm at least close after all that 

Perhaps I should make clear that although I've done some electronic tinkering here and there, my knowledge of theory is patchy to say the least

Oh and kirkd, nice rig! - is that a massive elastic band??? and what's the black thing? I ebayed that exact same battery last night, is it the highly regarded 'Walmart DVD battery' I've heard so much of?

[kirk97132]

Yep one big ass rubber band.  The black thing is a Beachtek SVU-1.  One of the downsides to the UA-5 is metering.  It only has a peak light which is designed to light up at about +3 .  The SVU-1 provides much easier way to set levels.  There is an SVU-1 for sale in the YS right now.  They are a discontinued item and can only be found used.  And that is one of the "Walmart DVD" batteries that a lot of people use for their gear.  A 5400mAh battery will run the UA-5 for over 6 hours no problem.  Probably much longer but I've never needed it to go longer than that. 

[spaceboy_psy]

hmm, metering, an unforeseen niggle! SVU-1 looks perfect, except I'm in the UK and don't trust international post to necessarily get here before I leave for Nepal next month. Plus possible import duties.

So the peak lights go on at +3db..? as in, louder than 0db..??

[Roving Sign]

hmm, metering, an unforeseen niggle! SVU-1 looks perfect, except I'm in the UK and don't trust international post to necessarily get here before I leave for Nepal next month. Plus possible import duties.

So the peak lights go on at +3db..? as in, louder than 0db..??

I think that should be -3db

[spaceboy_psy]

Ha, ok, so that makes things tricky. I'll have to look into this some more *dons investigative cap...*

[kirk97132]

Oops that is correct it is supposed to light at -3dB.  See that's why I use a meter   Edirol manual says "light in red at the level at which the sound begins to distort (-3dB)."  It also does state that the unit will accept an unbalenced 1/4" input along with 1/4" TRS.  Also my unit has had different knobs installed on the front

[DSatz]

spaceboy_psy, the white papers on Rane's Web site are far better than the usual on-line opinion pieces that masquerade as expertise on this subject. But in the particular article that you linked to, they don't include an example of the type of hookup that I'm suggesting. They show something that's partway similar: the use of a transformer at the output of an unbalanced source component, which then drives a balanced line (cable) with a balanced input at the far end.

That would be a possible solution in your case, though it would be less then optimal in several respects. Just briefly, if you're going to use transformers, you'd get considerably better noise rejection by using input transformers at the inputs to your UA-5 than you would by using output transformers on the microphone end of things. (The two types of transformers are built differently; input transformers can use a type of shielding between windings that blocks the capacitive coupling of RFI, while output transformers can't.)

But that wasn't the approach that I was talking about. I was suggesting a less expensive way to create a balanced output--essentially the way Neumann does it in their transformerless microphones. Only one wire (say, the one connected to XLR pin 2) is directly driven by the microphone's output circuit, which has a certain output impedance. The other wire (say, the one connected to XLR pin 3) is connected to a resistor that matches that output impedance. To improve the match at higher frequencies (which is important, since any interference is rejected only to the extent that the hookup is balanced at the frequency of the interfering signals), a series capacitor can also be used.

I understand why you feel that I haven't said very clearly what you should do--mainly because I haven't! But the more I think about this, the more I wish that one or more of the people on this board who build stuff and modify circuits for people here would look into building small, low-cost, transformerless microphone balancing boxes, because a shielding enclosure and some panel-mounted connectors would be needed, and that takes a little experience to design and assemble. The value of the resistors and capacitors inside the box would also have to match--as closely as possible for maximum noise rejection--the actual output impedance of the microphones that the box would be used with.

So, unfortunately, this couldn't be a "one model works with all types of microphones" kind of thing. Someone would have to find out what the exact, actual output impedance of an AT822 is, for example--not just the spec sheet value. That's not too hard, but it takes a bit of doing (or a bit of asking--I'd expect A-T to cooperate gladly; the engineers that I've met from that company have been very nice and very smart).

Just as an aside I'd say that the majority of the technical problems that come up on this board involve making consumer-type equipment work with professional-type or "semi-pro" equipment. People should be aware that the standards of the two realms weren't designed to "interoperate" and that all these contortions and anxieties result from trying to fudge distinctions that were never designed to be fudged.

It's not fair; even if some people are always going to have less money than others, on some level everyone should have an approximately equal right to be able to use the equipment that they can afford. But the industry standards that try to guarantee that stuff will work together properly only define the expected behavior of professional equipment on the one hand, and consumer equipment on the other hand. There are no standards for how those two "hands" should work together, and different equipment manufacturers make widely different (and sometimes totally conflicting) assumptions about how that should be done. So the solutions are often ephemeral and piecemeal.

--best regards

[chris319]

http://www.dself.dsl.pipex.com/ampins/balanced/balanced.htm

[spaceboy_psy]

 

Incredible DSatz, thanks for the white paper of your own!

Let's get stuck in...

Only one wire (say, the one connected to XLR pin 2) is directly driven by the microphone's output circuit, which has a certain output impedance. The other wire (say, the one connected to XLR pin 3) is connected to a resistor that matches that output impedance. To improve the match at higher frequencies (which is important, since any interference is rejected only to the extent that the hookup is balanced at the frequency of the interfering signals), a series capacitor can also be used.

Aha! Aha Aha ahaha ha! I get it now

reading Chris319's link helped a lot too (cheers chris!). So THIS is what we're looking for (just with an added capacitor?):



So the clever bit (one thereof...) is that pin 3 [cold] is being subtracted from pin 2 [hot], but the cold wire isn't carrying any signal, just noise and interference, which will be cancelled! Great! One question - could both channels share one resistor and capacitor, or would I need one set for each channel?

So yup, alas the consumer/professional audio dialectic has not been throwing up any kind of positive synthesis, only trouble, trouble, and more trouble!

In terms of implementation then, it's got to be the Balance Box doesn't it. No way to really implement this in a cable as such..? Unfortunately I think building a boxed unit is probably above my present ability in these things, and certainly beyond my time and money resources, for now. But great to know theoretically how it would be done, really appreciate the knowledge imparted here, thanks.

I'm thinking I might go for the pseudo-balanced lifted sleeve approach mentioned before. I figure get a 5pin XLR to 2 x 3 pin XLR cable and swap the 5pin for a 3pin with the two sleeves disconnected. Add TRS adaptors for Microtrack. Hopefully, from what I've read, that will avoid the nasty MTII "sprinkler" noise, and give me a bit less noise going into the UA-5.

To be honest I'm even starting to worry whether I can afford to buy cables for this setup (was banking on getting a grant to go and conduct research this summer, but the grant fell through. Still going, just with little to no money!)

Whatever I end up doing, I'm damn glad to know all my options! If I do end up getting into building myself a Balance Box, or even trying to get these cable shenanigans to work, I'll post back to update (and most likely to weep "WHY WON'T IT WORRKKKK???")

Many thanks for the valuable info everyone, particularly DSatz, ya star

[spaceboy_psy]

Hi mshilarious, thanks for chipping in. Apologies for late reply, been very busy finishing uni (and trying to research your PFA), but still, how rude!

This PFA sounds great. I couldn't find a schematic floating around anywhere, don't suppose you'd be willing to share a copy with a friendly scot who wants to modify it slightly? Would want to remove the phantom power section as the AT822 is battery powered of course, and maybe build it in XLR -> 2 x XLR...

However, many of these type of mics have an output impedance in the 1k-2k range, so I don't typically stick with the impedance-balanced arrangement.  I'd prefer a circuit that lowers the output impedance to something more reasonable.

So impedance-balanced would be fine for circa. 200 Ohm AT822?

Cheers!

[spaceboy_psy]

and no capacitor?

[spaceboy_psy]

Right! Again, many thanks for all the help. I jet off for Nepal on tuesday so this little effort will be something to look forward to upon my return in August

Looks like I'm now going without my UA-5!! As the stupid Walmart battery died on me 

So just AT822 and MTII, not gonna be anything spectacular then.

Adios lads

[eman]

Are you aware that the Walmart battery has a sensitive overload/ short circuit switch? To reset it, just plug it back in for recharging. I've tried to use them for powering guitar pedals, but there is something about the pedals that the battery doesn't like. The battery appears dead until you plug it back into the wall. It's too bad- the battery is perfect for that application and the cable is the same for a UA-5 as a Boss pedal. Check your battery cable for shorts when you move it around.

I just got AT853's and I've had an 822 for years. The 853's completely smoke the 822, and cost less. It does make a nice interview mic for using with a video camera, though. People will talk to someone holding a mic much more easily than someone holding a camera, plus you get both sides of the conversation on video.

[spaceboy_psy]

hmm, plugging it in to recharge does seem to make it 'work' normally, but only until you plug somethingin to be powered! Then it dies

As I'm away now I'll have to wait until August for further testing...

Cheers for the help

[Windwman]

I know this is an old post, but I have to give thanks to all who made suggestions....I used the 200-ohm cold to gnd trick & it worked great. I noticed it was noticably quieter on a long 30ft run compared to "stock". Thanks again!