mark.sullivan
03-20-2008, 09:56 AM
I'm running Red S/N 294 with Build 15 Version 2.1.4
I do not have the new cable so I did all this with the old straight-through adapter. It is correctly wired per Switchcraft but polarity-reversed according to the pinout in the Red operations manual. I hope the manual is wrong and the camera is right. If not, it's an easy matter to flip the phase in firmware, but, depending on the preamp design, it may cause trouble with single-ended sources that tie pins 1 and 3 together.
Here's Switchcraft's pinout:
http://www.switchcraft.com/products/connectors-23.html
It is NOT the same as the full-size XLR, here:
http://www.switchcraft.com/products/pdf_files/connector-5a_schematic.pdf
I have not found an explanation from Red of why we need the new cable. Anders posted a diagram of it's guts and it seems to have an L-pad in it. Why do I want an attenuator in my mic cable? Is the new cable phase-swapped and is this because the camera's input is backwards, as the manual says?
Anyway, I hooked up a Sennheiser ME66-K6P to channel 1 (using the old adapter). I set the gain to 20dB. I enabled recording only on channel 1.
This mic can use an internal battery or phantom power. Both work and neither has the hiss I had in Build 12.
If I enable recording on input 2, which has no cable connected, with phantom power turned on for that channel, then I get hiss. If I turn off phantom power, even though channel 1 still has it on, the hiss goes away. If I plug another mic into channel 2, the hiss goes away. This makes me wonder if the hiss problem others are having is due to their mic not loading the phantom power enough to stop the hiss. This might also explain the utility of the L-pad. It's not the attenuation that's important, but the DC load?
I don't get any hiss from inputs set to "Line" mode.
When I turn off the phantom power, then I hear a whine, probably the DC to DC converter. It goes away after a few seconds, like maybe there's a capacitor in front of the converter that has to run down. But the DC output stops instantly. Maybe there's one DC-to-DC converter with individual transistors feeding each input jack.
A couple of times, I have noticed a problem initializing the phantom power configuration menu. Phantom power is on - the DC is coming out of the camera and it shows 48V in the level meters - but when I enter the "48V MIC" configuration menu, none of the channels are checked. Checking and unchecking them does turn the DC output on and off.
So the phantom power is working for me and I'm not having hiss problems. I'm being careful to only enable the inputs to which I have mics attached.
I found two problems with the headphone outputs. First, the left and right channels are reversed. I have the mix set to 1=L, 2=R, but channel 1 comes out the right headphone channel and channel 2 comes out the left.
A worse problem is that there is a strong DC bias on both channels of the headphone output. It puts 18 mA into my Beyerdynamic DT 770 Pro and 400 mA into a short. Wow! This bias is independent of the headphone volume setting. The DT 770 is a relatively high-impedence phone but i wouldn't want to use 8 ohm phones with this bias. And the camera amplifier probably isn't designed to put out half a watt of continuous DC, either.
In light of all this, I suggest:
- Make sure unused inputs have phantom power off and rec enable off.
- Use high-impedance phones.
- Don't trust audio phase, if it's important to you.
I have to shoot this afternoon so I'm out of time. Here are some things I would try if I had more time:
- Does a DC or AC load stop the hiss on an unused input? This might mean there is a really high frequency oscillation in the preamp.
- Look at the phase of an actual recorded file to see which pinout of the connector is really the correct one.
- Try a single-ended source with pin 1 strapped to 2 or 3.
- Plug 8 ohm phones into the headphone jack.
- Measure the end-to-end AC gain from input to file.
- Record a shorted input to evaluate preamp equivalent noise.
- Try channels 3 and 4.
I do not have the new cable so I did all this with the old straight-through adapter. It is correctly wired per Switchcraft but polarity-reversed according to the pinout in the Red operations manual. I hope the manual is wrong and the camera is right. If not, it's an easy matter to flip the phase in firmware, but, depending on the preamp design, it may cause trouble with single-ended sources that tie pins 1 and 3 together.
Here's Switchcraft's pinout:
http://www.switchcraft.com/products/connectors-23.html
It is NOT the same as the full-size XLR, here:
http://www.switchcraft.com/products/pdf_files/connector-5a_schematic.pdf
I have not found an explanation from Red of why we need the new cable. Anders posted a diagram of it's guts and it seems to have an L-pad in it. Why do I want an attenuator in my mic cable? Is the new cable phase-swapped and is this because the camera's input is backwards, as the manual says?
Anyway, I hooked up a Sennheiser ME66-K6P to channel 1 (using the old adapter). I set the gain to 20dB. I enabled recording only on channel 1.
This mic can use an internal battery or phantom power. Both work and neither has the hiss I had in Build 12.
If I enable recording on input 2, which has no cable connected, with phantom power turned on for that channel, then I get hiss. If I turn off phantom power, even though channel 1 still has it on, the hiss goes away. If I plug another mic into channel 2, the hiss goes away. This makes me wonder if the hiss problem others are having is due to their mic not loading the phantom power enough to stop the hiss. This might also explain the utility of the L-pad. It's not the attenuation that's important, but the DC load?
I don't get any hiss from inputs set to "Line" mode.
When I turn off the phantom power, then I hear a whine, probably the DC to DC converter. It goes away after a few seconds, like maybe there's a capacitor in front of the converter that has to run down. But the DC output stops instantly. Maybe there's one DC-to-DC converter with individual transistors feeding each input jack.
A couple of times, I have noticed a problem initializing the phantom power configuration menu. Phantom power is on - the DC is coming out of the camera and it shows 48V in the level meters - but when I enter the "48V MIC" configuration menu, none of the channels are checked. Checking and unchecking them does turn the DC output on and off.
So the phantom power is working for me and I'm not having hiss problems. I'm being careful to only enable the inputs to which I have mics attached.
I found two problems with the headphone outputs. First, the left and right channels are reversed. I have the mix set to 1=L, 2=R, but channel 1 comes out the right headphone channel and channel 2 comes out the left.
A worse problem is that there is a strong DC bias on both channels of the headphone output. It puts 18 mA into my Beyerdynamic DT 770 Pro and 400 mA into a short. Wow! This bias is independent of the headphone volume setting. The DT 770 is a relatively high-impedence phone but i wouldn't want to use 8 ohm phones with this bias. And the camera amplifier probably isn't designed to put out half a watt of continuous DC, either.
In light of all this, I suggest:
- Make sure unused inputs have phantom power off and rec enable off.
- Use high-impedance phones.
- Don't trust audio phase, if it's important to you.
I have to shoot this afternoon so I'm out of time. Here are some things I would try if I had more time:
- Does a DC or AC load stop the hiss on an unused input? This might mean there is a really high frequency oscillation in the preamp.
- Look at the phase of an actual recorded file to see which pinout of the connector is really the correct one.
- Try a single-ended source with pin 1 strapped to 2 or 3.
- Plug 8 ohm phones into the headphone jack.
- Measure the end-to-end AC gain from input to file.
- Record a shorted input to evaluate preamp equivalent noise.
- Try channels 3 and 4.