BAN Build Thread

Support for JLM Audio Kits

Moderator: Joe Malone

Post Reply
User avatar
Joe Malone
Site Admin
Posts: 2070
Joined: Wed Jan 24, 2007 11:35 pm
Location: Brisbane, Australia
Contact:

Re: BAN Unbalance Output

Post by Joe Malone »

chrisp wrote:Joe, I'm looking at a pair of BANs as a pre for stereo miking. I have a little PCB design to do mid-side decoding that I can switch in and out, but it works on unbalanced signals. Is there any unbalanced take off point on the BAN PCB just prior to the outputs? Will TP3 and ground do it for me? I could then return the signal the 51R and 10K resistors. If I read the thing right, this signal point would be -3db, but I can live with that. And I'd have to be careful about noise ....
HI Chris the BAN is fully balanced until the output or fully balanced output as well with normal DIP8 opamps. So no unbalanced point before the driving output opamp. But you can do switchable M-S decode with a switch and 4 resistors into the middle of 2 x BAN easy. I can draw it up if you are interested?
Joe :-)
JLM Audio
Capturing Audio without Injury

chrisp
Posts: 184
Joined: Sun Mar 04, 2007 11:46 am
Location: Sydney, Australia

Re: BAN Build Thread

Post by chrisp »

Thanks Joe - that would be great.

I've made up an MS encode maxtrix for balanced signals before, but it was 8 resistors and not switchable - yech! So I am fascinated by the ol' JLM design magic...
Chris P
---------------------------------------------------------
I do lots of things. I believe eclectic skills are best.

chrisp
Posts: 184
Joined: Sun Mar 04, 2007 11:46 am
Location: Sydney, Australia

Re: BAN Build Thread

Post by chrisp »

Is the balanced signal take off point around the 2K2 resistors between the opamps? And is the idea that you can use those 2K2s as mix resistors? Or am I barking up the wrong tree as usual?
Chris P
---------------------------------------------------------
I do lots of things. I believe eclectic skills are best.

User avatar
Joe Malone
Site Admin
Posts: 2070
Joined: Wed Jan 24, 2007 11:35 pm
Location: Brisbane, Australia
Contact:

Re: BAN Build Thread

Post by Joe Malone »

chrisp wrote:Is the balanced signal take off point around the 2K2 resistors between the opamps? And is the idea that you can use those 2K2s as mix resistors? Or am I barking up the wrong tree as usual?
You are on the right track. I will finish of the wiring diagram to show how on the weekend as it has been to busy this week to do it.
Joe :-)
JLM Audio
Capturing Audio without Injury

chrisp
Posts: 184
Joined: Sun Mar 04, 2007 11:46 am
Location: Sydney, Australia

Re: BAN Build Thread

Post by chrisp »

No rush, Joe - enjoy your weekend
Chris P
---------------------------------------------------------
I do lots of things. I believe eclectic skills are best.

User avatar
Joe Malone
Site Admin
Posts: 2070
Joined: Wed Jan 24, 2007 11:35 pm
Location: Brisbane, Australia
Contact:

2 BAN kits doing M-S Matrix with 4 resistor and a switch

Post by Joe Malone »

1. 4 pole switch open pres operate as normal separate channels.
2. 4 pole switch closed Left will output L-R Side and Right will output L+R Middle which is reversed to the normal standard.
3. 4 pole switch closed and Flip phase on Right channel so Left will output L+R Middle and Right will output L-R Side so is the normal standard. Sending in L+R Middle on Left and L-R Side on Right with give Left out and Right out on respective channels.

Image
Joe :-)
JLM Audio
Capturing Audio without Injury

chrisp
Posts: 184
Joined: Sun Mar 04, 2007 11:46 am
Location: Sydney, Australia

Re: BAN Build Thread

Post by chrisp »

Thanks Joe - looks very easy.

For those who have not tried it, Mid-Side recording is a stereo recording arrangement that uses an omni or cardiod ("Mid") mic directed to the sound source and a figure 8 ("Side") mic whose two response planes are perpendicular to the sound source (ie one of the "dead zones in the Fig 8 mic points to the source). Mid + Side summed gets panned hard left, and Mid - Side hard right. Its main advantages are the co-location of the 2 mics can minimise phase artefacts, that it sums exactly to mono and by adjusting of the Mid and Side volume prior to decoding (which Joe's set up implements), you can wary the width of the sound stage, making it as wide or narrow as you like. I like it a lot for things like acoustic guitar, some vocals (I did a spoken word recording where it worked particularly well) and drum overheads. If you are willing to ride the gain control, you can have nice effects where vocals can be tightly centred and intimate when the signer is soft, and then widen out into the stereo image as the singer gets louder.

You can also take a L R paring and, as per Joe's diagram, get Mid and Side out of it. Why? Sometimes processing is easier better using Mid-Side as it avoids many phase and image problems compared to processing L and R. With Joe's setup you can in fact plug in a line signal (engaging the pad on both channels) and merrily convert LR to MS and back to your heart's content!

Have fun, everyone.
Chris P
---------------------------------------------------------
I do lots of things. I believe eclectic skills are best.

beccogiallo
Posts: 3
Joined: Sun Mar 04, 2012 8:50 pm

BAN Details

Post by beccogiallo »

Hi everybody.
I’m interested in some BAN channels but I’ve got some questions, some of them probably pretty basic since I’m only a musician and unfortunately no EE. Sorry for weak English too.
1.Input transistors: I’m interested in jfets, a pair of 2SK389 would need to be matched? And what about 2SK170s? Are those in the shop matched? In both cases, what’s the right Mu grade? GR, BL etc.
2.I’m curious about your Hybrid Opamp as output device. BA/BAN is a single supply opamp application and I’m not able to figure out if the DC servo in the Hybrid can cancel out the DC bias (half supply voltage) present at opamp input, or instead the output DC filtering capacitor is required anyway.
3.A single Hybrid obviously has an unbalanced output. I don’t understand if using it in BAN as output device produces only an unbalanced output (IC1b doesn’t exist anymore), or if a “double Hybrid” exists and should be used in order to have a balanced output.
4.Getting back to monolithic output opamp. When using the output in unbalanced mode, typical situation (see mono jack in combo XLR/TRS socket) grounds “pin 3” (inverting output): this means that IC1b (which is only an inversor after all) should be able to withstand a permanent short on its output (I believe it’s true for many opamps). Does this apply in BAN too?
5.On location I’d like to power some BAN channels out of a battery and a DC-DC converter (outside preamp’s cabinet). Do you consider useful, and can you suggest, some kind of additional filtering, for example RF filtering at preamp DC power lines inlet? (inductors?)

Thank a lot for helping me out of doubts.

User avatar
Joe Malone
Site Admin
Posts: 2070
Joined: Wed Jan 24, 2007 11:35 pm
Location: Brisbane, Australia
Contact:

Re: BAN Details

Post by Joe Malone »

beccogiallo wrote:Hi everybody.
I’m interested in some BAN channels but I’ve got some questions, some of them probably pretty basic since I’m only a musician and unfortunately no EE. Sorry for weak English too.
1.Input transistors: I’m interested in jfets, a pair of 2SK389 would need to be matched? And what about 2SK170s? Are those in the shop matched? In both cases, what’s the right Mu grade? GR, BL etc.
Yes all devices should be matched. For lowest noise highest gain versions should be used. Shop ones are not matched. I still prefer matched pair of BC550C best which come with the kits.
2.I’m curious about your Hybrid Opamp as output device. BA/BAN is a single supply opamp application and I’m not able to figure out if the DC servo in the Hybrid can cancel out the DC bias (half supply voltage) present at opamp input, or instead the output DC filtering capacitor is required anyway.
Output cap can never be removed. DC servo will no help remove the +24v at the output of the BAN only a cap can.
3.A single Hybrid obviously has an unbalanced output. I don’t understand if using it in BAN as output device produces only an unbalanced output (IC1b doesn’t exist anymore), or if a “double Hybrid” exists and should be used in order to have a balanced output.
With the wire jumper on the DIP8 betwen pin 4 and pin 7 the -out is equal output impedance as the +out with the audio signal so any balanced input device that it plugs into will remove any noise picked up. This quasi-balanced output is used a lot in mastering equipment like Sontec and now even in modern Neumann microphones.
4.Getting back to monolithic output opamp. When using the output in unbalanced mode, typical situation (see mono jack in combo XLR/TRS socket) grounds “pin 3” (inverting output): this means that IC1b (which is only an inversor after all) should be able to withstand a permanent short on its output (I believe it’s true for many opamps). Does this apply in BAN too?
Yes be only use NE5532A and OPA2604AP which with the 51R output impedance resistors can handle continuious short circuit.
5.On location I’d like to power some BAN channels out of a battery and a DC-DC converter (outside preamp’s cabinet). Do you consider useful, and can you suggest, some kind of additional filtering, for example RF filtering at preamp DC power lines inlet? (inductors?)
We do this all the time. Some types of SMPS do not need any extra filtering but other will depending on load of the number of channels. The BAN power in filter of 10R and 470uF RC filter is good at removing high frequency noise of some SMPS. Basically start with no filtering and test. Most DC to DC SMPS have a recommended extra filter stages to try.
Thank a lot for helping me out of doubts.
:D 8)
Joe :-)
JLM Audio
Capturing Audio without Injury

beccogiallo
Posts: 3
Joined: Sun Mar 04, 2012 8:50 pm

Re: BAN Build Thread

Post by beccogiallo »

Thanks Joe. That's all very clear.
One last question: to cut on the gain attenuator steps (12 or so) I'd like to have a fine gain pot to cover inbetween (5-6 dB). I suppose this could be done tapping into the feedback loop of IC1 but, again, I'm not up to grasp if the DC present will play havoc with the pot. Should definitely study more :( :D

beccogiallo
Posts: 3
Joined: Sun Mar 04, 2012 8:50 pm

Re: BAN Build Thread

Post by beccogiallo »

No clue anybody? :?: :(

User avatar
Joe Malone
Site Admin
Posts: 2070
Joined: Wed Jan 24, 2007 11:35 pm
Location: Brisbane, Australia
Contact:

Re: BAN Build Thread

Post by Joe Malone »

beccogiallo wrote:Thanks Joe. That's all very clear.
One last question: to cut on the gain attenuator steps (12 or so) I'd like to have a fine gain pot to cover inbetween (5-6 dB). I suppose this could be done tapping into the feedback loop of IC1 but, again, I'm not up to grasp if the DC present will play havoc with the pot. Should definitely study more :( :D
Change the 2k2 on each side of IC2 both to 2 x 1k in series. Then to attenuate place a resistor or pot between the center joins of both 1k. From no RAtten resistor fitted = 0dB a 2k RAtten will give about 6dB attenuation.

original removed 2k2
----------/\2k2\/---------

---/\1k\/-------/\1k\/---
.............|
............ >
.........RAtten
.............<
.............|
---/\1k\/-------/\1k\/---

----------/\2k2\/---------
original removed 2k2

Ignore the decimal points that are used since forums do not support multiple spaces or tabs
Joe :-)
JLM Audio
Capturing Audio without Injury

prhunt
Posts: 14
Joined: Sat Jan 28, 2012 1:08 pm

Re: Switched resistors for gain

Post by prhunt »

Joe, I have 24-position switches, and I'm thinking of implementing 2dB gain steps.

Assuming 1% resistors, how well matched would gains be between channels?

Thanks! Paul

User avatar
Joe Malone
Site Admin
Posts: 2070
Joined: Wed Jan 24, 2007 11:35 pm
Location: Brisbane, Australia
Contact:

Re: Switched resistors for gain

Post by Joe Malone »

prhunt wrote:Joe, I have 24-position switches, and I'm thinking of implementing 2dB gain steps.

Assuming 1% resistors, how well matched would gains be between channels?

Thanks! Paul
HI Paul

Easy one. With all resistors 1% there are 2 gain stages so 4 resistors all with 1% tolerance so +/-4% error. We do find that 1% resistors are mostly within +/-0.5% range so general error is usually only +/-2%.

Easiest way to minimize channel error when building a matched pair is to use a multimeter and match the gain stage resistor values between channels. As a error of +/-0.5% or less can easily be reached then.

Or if the channels are already built you can match the channels gain at a mid gain level by raising the gain of lowest level channel by adding a large value resistors across the 2k2 resistors on each side of IC2 to bring its gain up to the same level as the other channel. Once this mid gain point is matched the gain match will be very good if you match the gain switch resistors with a multimeter. Or you can do nothing to the BAN PCB's and just tweak each gain switch step resistor value to match the gain of the other channel by adding a large value resistor across the switch resistor on the channel that is lowest gain to bring its gain up to match.
Joe :-)
JLM Audio
Capturing Audio without Injury

prhunt
Posts: 14
Joined: Sat Jan 28, 2012 1:08 pm

Re: Switched resistors for gain

Post by prhunt »

Joe Malone wrote:Easy one. With all resistors 1% there are 2 gain stages so 4 resistors all with 1% tolerance so +/-4% error. We do find that 1% resistors are mostly within +/-0.5% range so general error is usually only +/-2%.

Easiest way to minimize channel error when building a matched pair is to use a multimeter and match the gain stage resistor values between channels. As a error of +/-0.5% or less can easily be reached then.
Cool! Thanks Joe.

Paul

Post Reply