Topic: DIY: three-wire batt. box + preamp

[poorlyconditioned]

Here are some pictures of my latest project, a DIY three-wire battery box + preamp:

Preamp/battery box connected to AKG CK91 capsules.  Line in to minidisc:


Innards of battery box/preamp:


The amplifier is based on an LT6234 chip.  This is a low-power, high quality opamp (1.15mA per channel, tiny SOIC package).  There are fixed gains of 12dB, 18dB, and 22dB (slider switch at bottom).  The top switch is for power one way, and LED/battery test the other way.  This is built into an old AT8352 battery module.  I used this because it already has a nice 9V battery holder and one miniXLR jack.  The other miniXLR was added (using a Dremmel tool!).

Details of the CK91 capsules (and larger pics of this project) are available at:
  www.cs.uwaterloo.ca/~mannr/AKG
Schematics of the battery box and opamp are at:
  www.cs.uwaterloo.ca/~mannr/AT853

PS: Please don't ask me to make you one of these.  Soldering small components and fitting them in the case was very difficult!  Anyone good with mechanical and electrical skills is welcome to use the design though.

  Richard

[Lil Kim Jong-Il]

Nice.

I know a place where that would be useful. 

[dgodwin]

Nice!   Will you make me one?     (j/k)  +T

[aberg]

Nice Richard, I'd love to hear some recordings with that setup...

[graemecogger]

Richard,

I hope you don't mind a couple of small suggestions about the design

First, I'd suggest increasing the input impedance.  The current value (works out at 11K) is low enough to give some distortion as the gm of the mic FET varies with output level.  One of the main points of the 3-wire connection is to decrease the dependance of output on gm, thereby reducing distortion.

Second, I'd suggest a change to the voltage divider at the input.  Changing from:



to:



would drastically improve supply rejection (PSRR) which, in the current design, is in the order of 0dB at full gain.  If this was a hi-fi pre-amp, I'd expect that change to give a noticeable improvement to transparency/detail (reduced 'smearing') but I don't know if that would be audible with these mics.

All the best,

Graeme

[RebelRebel]

Graeme, no professionals allowed

How are you Sir?Havent seen you on the other place too much.

teddy

[poorlyconditioned]

Richard,

I hope you don't mind a couple of small suggestions about the design

First, I'd suggest increasing the input impedance.  The current value (works out at 11K) is low enough to give some distortion as the gm of the mic FET varies with output level.  One of the main points of the 3-wire connection is to decrease the dependance of output on gm, thereby reducing distortion.

Second, I'd suggest a change to the voltage divider at the input.  Changing from:



to:



would drastically improve supply rejection (PSRR) which, in the current design, is in the order of 0dB at full gain.  If this was a hi-fi pre-amp, I'd expect that change to give a noticeable improvement to transparency/detail (reduced 'smearing') but I don't know if that would be audible with these mics.

All the best,

Graeme

Thanks for the suggestions.  I didn't expect such detail on this forum.  You know about the "micbuilders" forum on Yahoo groups, right?  I hang out there as well.  I'll make the changes you suggest.

I've got a question about the 3-wire powering:  Do you know the output impedance of a mic when I've got the FET wired as a "source follower".  Consider the case of a FET with a source resistor (to ground) of 10K and a drain resistor (to +9V battery) of 3K.  The output impedance is deterined by the drain resistor, right?

Anyway, thanks again for the help!
  Richard

[graemecogger]

Graeme, no professionals allowed

How are you Sir?Havent seen you on the other place too much.

teddy

I'm not a pro any more - all this electronics stuff is fading into the haze of distant memories
I've not been posting much, as I haven't had a lot to say.  I've acquired the SP-CMC-8 omni mics from Sound Professionals and have the choice of a Hi-MD recorder or iRiver iHP-120 with Rockbox firmware for recording.  Unfortunately I've had little chance to use it yet.  I need to make a pre-amp before the end of the month (my wife's next recital), and I'll probably be doing something very similar to Richard.  Did you get someone to do your mods?

I've got a question about the 3-wire powering:  Do you know the output impedance of a mic when I've got the FET wired as a "source follower".  Consider the case of a FET with a source resistor (to ground) of 10K and a drain resistor (to +9V battery) of 3K.  The output impedance is deterined by the drain resistor, right?

I'm not sure why you need a drain resistor in a source follower - your 'blackboard' circuit doesn't show one.  It's not needed and can only degrade the overload and distortion performance.  The output impedance is (to a good approximation) a resistance of 1/gm in parallel with the resistor from source to ground.  Without having the details of the FET used, it's impossible to come up with a value.  I tried some SPICE simulations, and the nearest JFET I could get a model for needed an 18K source resistor to drop 4.5V.  It's actually about 30K on the mic capsules I've bought, but I couldn't find anything closer.  This had a 1/gm value of about 3K, so I'd guess the source follower as in your circuit for the AT853 (with 27K resistors) would have an output impedance a little higher than that.  A capsule that drops 4.5V with 10K would probably be a bit lower.  It's pretty much guesswork, but should be in the ballpark.  For best distortion performance, you need the source resistance and pre-amp input impedance (in parallel) to be as high as possible.  The source resistor is limited by where you want the operating voltage at the FET source (a little under half the supply voltage for best overload levels) so you need to get the pre-amp input impedance a fair bit higher than the source resistor.  You could easily go higher than the 47K I suggested in my circuit - should be fine for most op-amps.

[RebelRebel]

 Did you get someone to do your mods?

Yessir, sure did!! Hope to have it back soon!Good to see you post. I think you have a lot to offer to the community here, from what I know of your EE knowledge.

[graemecogger]

I think I said somewhere that I'd post the circuit when I settled on my pre-amp design, so here it is if anyone is interested.  It's a combined 3-wire battery box (for the AT943 caps) and mic pre-amp.



As you can see, it's very similar to Richard's.  R1, R2 and C3 are shared between the channels to save space.  Some of the values have been chosen because it's what I have lying around.  I've kept it as small as possible, to fit in a small box.  I'm sure it could be improved at the cost of space and power consumption (although I've not worried too much about the latter).  Unfortunately, C2 needs to be an electrolytic since my iRiver iHP-120 has a very low 2K2 input impedance.

The input impedance is high (about 100K), so it wouldn't be too good with longer cables.  The software I used to draw it doesn't do switches (!), but it should be clear what they do.  SW1 is switchable bass cut with a DPDT, centre off switch (gives rolloffs at 3, 94 and 233Hz).  SW2 is for power and SW3 is a -20dB setting - in one position the gain is 0-20dB and in the other 0-40dB.

All I have to do now is build it

[poorlyconditioned]

I think I said somewhere that I'd post the circuit when I settled on my pre-amp design, so here it is if anyone is interested.  It's a combined 3-wire battery box (for the AT943 caps) and mic pre-amp.



As you can see, it's very similar to Richard's.  R1, R2 and C3 are shared between the channels to save space.  Some of the values have been chosen because it's what I have lying around.  I've kept it as small as possible, to fit in a small box.  I'm sure it could be improved at the cost of space and power consumption (although I've not worried too much about the latter).  Unfortunately, C2 needs to be an electrolytic since my iRiver iHP-120 has a very low 2K2 input impedance.

The input impedance is high (about 100K), so it wouldn't be too good with longer cables.  The software I used to draw it doesn't do switches (!), but it should be clear what they do.  SW1 is switchable bass cut with a DPDT, centre off switch (gives rolloffs at 3, 94 and 233Hz).  SW2 is for power and SW3 is a -20dB setting - in one position the gain is 0-20dB and in the other 0-40dB.

All I have to do now is build it

+T, thanks for the drawing!

So, do you think the LT1469 is going to be better than the LT6233 I'm using?  The LT1469 has better specs (and comes in a DIP package!) but takes 3.5mA vs. 1.2mA for each amp.  I might build around the 1469 if it is better.  Also, can you PM a source for LT1469?  I checked Digikey but they don't have any in stock.

Another note is the gain.  I like a gain switch instead of a pot, but that is personal preference.

Good luck, and please bring back photos when you're done!

  Richard

[graemecogger]

+T, thanks for the drawing!

So, do you think the LT1469 is going to be better than the LT6233 I'm using?  The LT1469 has better specs (and comes in a DIP package!) but takes 3.5mA vs. 1.2mA for each amp.  I might build around the 1469 if it is better.  Also, can you PM a source for LT1469?  I checked Digikey but they don't have any in stock.

Another note is the gain.  I like a gain switch instead of a pot, but that is personal preference.

Good luck, and please bring back photos when you're done!

  Richard

Hi again

It's difficult to say whether the 1469 will be better than the 6233 - it's a bit of a gamble until you listen to it, although either op-amp is probably better than the mics.  I wanted something that was unity gain stable, and I wasn't all that worried about the current draw.  I doubt I'll need more than about 25hrs between recharges!  I preferred it to the 6233 because of lower input bias current and better supply rejection.  The noise is a little higher, but I believe that the mic noise will the dominant factor.  I also liked the circuit topology from the data sheet - looks like the kind of thing I might design   One small issue is that it is not guaranteed to work within spec below 9V (it will typically work at 5V, but 9V is the worst case).  Most PP3 rechargeables are actually 8.4V, but I managed to find one that is 9.6V (and still 220mAh).

I considered a gain switch but decided on a pot because this may sometimes be used into a minidisc (Hi-MD) recorder.  On that, you can't adjust the record level without pausing the recording.  Pretty daft, I think.

I'm afraid I'm in the UK, so I doubt my sources would be of use to you...

I'll post some pics when it's done.  I've just worked out a layout that lets me use the box I wanted, but it only just squeezes in.  It's a good job that I could only source the LT1469 in surface mount format, I guess

Any thoughts on the values of the bass cut frequencies?  I still don't know why it's needed - if the bass level is so high, surely the gig would sound awful?  Still, I've put them on there just in case, at frequencies mostly decided by the capacitors I had available.

[poorlyconditioned]

+T, thanks for the drawing!

So, do you think the LT1469 is going to be better than the LT6233 I'm using?  The LT1469 has better specs (and comes in a DIP package!) but takes 3.5mA vs. 1.2mA for each amp.  I might build around the 1469 if it is better.  Also, can you PM a source for LT1469?  I checked Digikey but they don't have any in stock.

Another note is the gain.  I like a gain switch instead of a pot, but that is personal preference.

Good luck, and please bring back photos when you're done!

  Richard

Hi again

It's difficult to say whether the 1469 will be better than the 6233 - it's a bit of a gamble until you listen to it, although either op-amp is probably better than the mics.  I wanted something that was unity gain stable, and I wasn't all that worried about the current draw.  I doubt I'll need more than about 25hrs between recharges!  I preferred it to the 6233 because of lower input bias current and better supply rejection.  The noise is a little higher, but I believe that the mic noise will the dominant factor.  I also liked the circuit topology from the data sheet - looks like the kind of thing I might design   One small issue is that it is not guaranteed to work within spec below 9V (it will typically work at 5V, but 9V is the worst case).  Most PP3 rechargeables are actually 8.4V, but I managed to find one that is 9.6V (and still 220mAh).

I considered a gain switch but decided on a pot because this may sometimes be used into a minidisc (Hi-MD) recorder.  On that, you can't adjust the record level without pausing the recording.  Pretty daft, I think.

With the HiMD you *can* adjust record levels while running.  Here is what you do: hit pause-record, enter menu> REC SET >REC VOLUME > MANUAL.  Then hit pause to start and stop recording.  The "wheel" adjusts the record level, both in pause and record mode.  You see meters on the main display.  If you've got RM-MC40ELK remote, you see levels on the remote and *backlit* which is great!!!  Now why can't other stealth gear do that so nicely???

I'm afraid I'm in the UK, so I doubt my sources would be of use to you...

I'll post some pics when it's done.  I've just worked out a layout that lets me use the box I wanted, but it only just squeezes in.  It's a good job that I could only source the LT1469 in surface mount format, I guess

Any thoughts on the values of the bass cut frequencies?  I still don't know why it's needed - if the bass level is so high, surely the gig would sound awful?  Still, I've put them on there just in case, at frequencies mostly decided by the capacitors I had available.

I've never used the bass cut.  I *think* if you use 3-wire mic powering your mics won't overload so you won't need it.  You can fix it in post if you need I guess.

Again, please post pics when you're done.  I love seeing gear innards.

  Richard

[Weazel]

can the switches be combined for left and right? basfilter and the gainswitch  switch? or is induvidual for each channel?
and can batterypower be shared over left and right? or does it need 2 battery's?
will you post your pcb drawing when it;s done?

looks prommising.

[graemecogger]

Richard - I didn't realise you could adjust levels on the fly with the Hi-MD recorders.  I'll have to play with that.  Thanks for the info

Weazel - all the switches are double pole.  In other words, they affect both channels equally.  The battery is also shared.  The op-amp bypass caps (C4 and C7) are shared because it's a dual op-amp with only one pair of supply pins.  This is mostly being made on stripboard/veroboard, so no PCB.  There seems little point for a one-off.