BAN Build Thread

Support for JLM Audio Kits

Moderator: Joe Malone

User avatar
Joe Malone
Site Admin
Posts: 1658
Joined: Wed Jan 24, 2007 11:35 pm
Location: Brisbane, Australia
Contact:

Re: BAN Build Thread

Post by Joe Malone »

greenmanhumming wrote:BC109C is much more expensive, is there an advantage?
No BC109C it just older. BC549C is modern equivalent.
Joe :-)
JLM Audio
Capturing Audio without Injury

greenmanhumming
Posts: 117
Joined: Tue Apr 10, 2007 1:19 am

Re: BAN Build Thread

Post by greenmanhumming »

I have upgraded my BANs with LM394s, seem to be quieter , and a bit more solid sounding, maybe less sibilant, more emphasis on the low mids.should I ground the case of the LM394s?

User avatar
Joe Malone
Site Admin
Posts: 1658
Joined: Wed Jan 24, 2007 11:35 pm
Location: Brisbane, Australia
Contact:

Re: BAN Build Thread

Post by Joe Malone »

greenmanhumming wrote:I have upgraded my BANs with LM394s, seem to be quieter , and a bit more solid sounding, maybe less sibilant, more emphasis on the low mids.should I ground the case of the LM394s?
The LM394 is 2 x 50 parallel matched transistors so will usually drop the noise floor level 1 to 3dB more.

I have never tried to ground a LM394 metal can. You would have to check the LM394 data sheet to make sure the case fully floating from the transistors which I think it would be but it should make no difference to noise floor of the BAN.
Joe :-)
JLM Audio
Capturing Audio without Injury

greenmanhumming
Posts: 117
Joined: Tue Apr 10, 2007 1:19 am

Re: BAN Build Thread

Post by greenmanhumming »

I don't have a schematic for the BANs.

I have just plugged in 2 C414s identical in every way except one has an output transformer and the other doesn't - it has pin 3 grounded through a cap.

the 414 with the output transformer is significantly louder than the other on the BAN, but when plugged into a pre with a transformer input they are the same level.

am i right in thinking that a BAN is essentially 2 channels of amp independent of each other one for each side of the blanaced signal? so that in transformerless output one side goes unused? would this explain the level difference?

whats going on?

User avatar
Joe Malone
Site Admin
Posts: 1658
Joined: Wed Jan 24, 2007 11:35 pm
Location: Brisbane, Australia
Contact:

Re: BAN Build Thread

Post by Joe Malone »

greenmanhumming wrote:I don't have a schematic for the BANs.

I have just plugged in 2 C414s identical in every way except one has an output transformer and the other doesn't - it has pin 3 grounded through a cap.

the 414 with the output transformer is significantly louder than the other on the BAN, but when plugged into a pre with a transformer input they are the same level.
Yes that is correct. The transformerless input BAN, SSL, Amek, Neve Mozart, THAT1510, SSM2017 etc will be 6dB down when not presented with audio signal on both phases of the balanced line.
am i right in thinking that a BAN is essentially 2 channels of amp independent of each other one for each side of the balanced signal?
Yes correct.
so that in transformerless output one side goes unused? would this explain the level difference?
It is only used by the preamp to do the noise canceling.
Joe :-)
JLM Audio
Capturing Audio without Injury

greenmanhumming
Posts: 117
Joined: Tue Apr 10, 2007 1:19 am

Re: BAN Build Thread

Post by greenmanhumming »

Is the signal to noise or other performance degraded by this effect? All my mics are transformerless except the ribbons and the one 414 mentioned and most only output signal one one pin.

greenmanhumming
Posts: 117
Joined: Tue Apr 10, 2007 1:19 am

Re: BAN Build Thread

Post by greenmanhumming »

I'm also getting frustrated with the cheap 10k reverse log pots, I'd like to reoplace them with a stepped attenuator, but reverse log attenuators are not to be found, I could build one, but getting the right resistor values could be tricky. how about a kit?

or can you use a linear attenuator with a resistor across it to make it log?

greenmanhumming
Posts: 117
Joined: Tue Apr 10, 2007 1:19 am

Re: BAN Build Thread

Post by greenmanhumming »

I seem to be having multiple problems with my BANs possibly since changing to LM394, and fitting IC sockets to the input transistor holes for easier swap out. but some of the problems may have been pre-existing.

* one channel is louder than the other, this is partly pot mismatch, but not entirely.

* both channels seem to have high frequency oscillation, the louder channel has it in both + and - sides but it disappears if the gain is above 40% on the knob. The other (quieter) channel its lower level only in one polarity but doesn't disappear until the gain hits about 85%

its not audible, but easily visible on the scope when I was looking for causes of the gain disparity.

also the opa2604 chips seems to be hotter than i would expect (ranges from warm to too hot to touch for more than a couple of seconds), but it's hard to say which is hottest or under what conditions.

BTW when I input a signal generator on only one pin, I do get output from both polarities of the output, phase reversed correctly. this seem to contradict our conversation above.

help! what should I check? I can't find anything incorrectly connected.

User avatar
Joe Malone
Site Admin
Posts: 1658
Joined: Wed Jan 24, 2007 11:35 pm
Location: Brisbane, Australia
Contact:

Re: BAN Build Thread

Post by Joe Malone »

greenmanhumming wrote:Is the signal to noise or other performance degraded by this effect? All my mics are transformerless except the ribbons and the one 414 mentioned and most only output signal one one pin.
It depends on if the noise pickup pin3 has the same impedance (resistor to pin 1) as the audio output pin 2 from the microphone as to how good the signal to noise is. The fully floating transformer output will usually always be better signal to noise wise.
Joe :-)
JLM Audio
Capturing Audio without Injury

User avatar
Joe Malone
Site Admin
Posts: 1658
Joined: Wed Jan 24, 2007 11:35 pm
Location: Brisbane, Australia
Contact:

Re: BAN Build Thread

Post by Joe Malone »

greenmanhumming wrote:I'm also getting frustrated with the cheap 10k reverse log pots, I'd like to replace them with a stepped attenuator, but reverse log attenuators are not to be found, I could build one, but getting the right resistor values could be tricky. how about a kit?

or can you use a linear attenuator with a resistor across it to make it log?
Easy to do but will cost you twice the price of the BAN for a 24way attenuator and then you will be stuck with 2dB or 3dB steps which with most same type microphones whose output varies +/-3dB output level wise isn't really going to help. Like all our 75dB mic pres you need a stepped attenuator and a small range trim as nothing is matched in the real audio world when microphones and air distance are included.
Joe :-)
JLM Audio
Capturing Audio without Injury

User avatar
Joe Malone
Site Admin
Posts: 1658
Joined: Wed Jan 24, 2007 11:35 pm
Location: Brisbane, Australia
Contact:

Re: BAN Build Thread

Post by Joe Malone »

greenmanhumming wrote:I seem to be having multiple problems with my BANs possibly since changing to LM394, and fitting IC sockets to the input transistor holes for easier swap out. but some of the problems may have been pre-existing.

* one channel is louder than the other, this is partly pot mismatch, but not entirely.
If working properly can only be due to the +/-1% resistors and +/-5% 10ohm and +/-10% 10k rev pots.
* both channels seem to have high frequency oscillation, the louder channel has it in both + and - sides but it disappears if the gain is above 40% on the knob. The other (quieter) channel its lower level only in one polarity but doesn't disappear until the gain hits about 85%
Sounds like a plated hole could have been open circuited while changing the transistor pair or the bandwidth limiting ceramics need to be a larger value for your configuration. Gain pot wires to long and not very tightly twisted. Also worth checking if it does it with the transformer output mic or only with the transformerless mics. As bad transformerless mics short pin 1 and pin 3 of the XLR together which is bad.
its not audible, but easily visible on the scope when I was looking for causes of the gain disparity.
Is your oscilloscope probe turned to x10 to minimize direct injected capacitance when measuring sensitive mic input stages. As often I see oscillation caused by the oscilloscope probe itself. Is it a pure frequency or noise?
also the opa2604 chips seems to be hotter than i would expect (ranges from warm to too hot to touch for more than a couple of seconds), but it's hard to say which is hottest or under what conditions.
The OPA2604AP running on 48v will always run hot as even at 15mA this is a lot of wattage in a small package.
BTW when I input a signal generator on only one pin, I do get output from both polarities of the output, phase reversed correctly. this seem to contradict our conversation above.
No not at all you just misunderstand how the pre works. It has 2 unbalanced gain amps that are then summed together to noise cancel with a third opamp and then a inverting unity opamp to make the -out pin 3. So you will always get the same signal on pin 3 inverted as +out pin 2.
help! what should I check? I can't find anything incorrectly connected.
You need to do testing with different mics and no mic to see if that changes the oscillation and read the frequency and level at every opamp output pin to narrow down the problems you are having and make sure the oscilloscope isn't actually causing the problem. If the mic pre is oscillating in the gain stage there will almost always be a point on the gain pot where a crack or thump sound happens as the mic pre goes into oscillation at that point.
Joe :-)
JLM Audio
Capturing Audio without Injury

greenmanhumming
Posts: 117
Joined: Tue Apr 10, 2007 1:19 am

Re: BAN Build Thread

Post by greenmanhumming »

thanks that's some things to look at.

yes the board is a mess, there are a couple of jumpers i put in to connect stuff that had been disconnected due to broken traces.

however as far as I can tell it is now connected correctly, but i may have missed a connection.

i suspect the quieter channel is only running one side at some point... but without a schematic I'm not really sure what to look for.

I didn't know about the x10 on the scope, but would this be an issue even with the scope on the output?

i see the oscillation, (its a pure frequency, not noise) with no mic attached or with a signal generator attached or with 120R resistors on the inputs, or with a mic.

all my transformerless mics are properly impedance balanced, no shorted pins 3 and 1

BTW 21 step log attenuators from asia can be had on ebay for US$10-20 depending on quality, I haven't tried the absolute cheapest ones, but the $20 ones work just fine. shame they are not available in reverse log.

greenmanhumming
Posts: 117
Joined: Tue Apr 10, 2007 1:19 am

Re: BAN Build Thread

Post by greenmanhumming »

I replaced the pots with 10k $US10 stepped attenuators from ebay, they work great and have cured the channel imbalance.

(although my gain controls now work in reverse, since reverse log attenuators were not available, i'm living with it in the name of mental flexibility!)

I still have (audible) oscillation at some gain settings, despite all the in and out wires being good quality shielded cable and the gain wires being short and tightly twisted. about to look for other possible causes...

greenmanhumming
Posts: 117
Joined: Tue Apr 10, 2007 1:19 am

Re: BAN Build Thread

Post by greenmanhumming »

I have shortened the legs on the (socketed)LM394 - no chnage.

I have increased the ceramic caps near the LM 394 to 100pF, that reduced the level of oscillation but it's still there. how big can I go? does this tell us anything?

I will try going back to different transistors, as I never noticed this oscillation before the LM394s were added.

greenmanhumming
Posts: 117
Joined: Tue Apr 10, 2007 1:19 am

Re: BAN Build Thread

Post by greenmanhumming »

going back to BC549 made it worse.

played around with the grounding, changed things but didn't eliminate it.

stumped.

Post Reply