Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

EPIC FAN NOISE - HEARTBREAKING DURING INTERVIEW

Will Keir said:
Looks like in the next firmware there is a "Post" fan 1 minute fix. Not sure exactly what this means, but I'm hoping you can set the fan to be off completely for a limited time.
Click to expand...

I think "post fan 1 minute" is to cut the camera shot and have the sound guy still rolling for continuous ambient without the cam to interupt.
Think of car, plane, siren in the far that you want to keep for the sound editing.


Pat
 
Patrick Tresch said:
I think "post fan 1 minute" is to cut the camera shot and have the sound guy still rolling for continuous ambient without the cam to interupt.
Think of car, plane, siren in the far that you want to keep for the sound editing.


Pat
Click to expand...

That will also be great for when you have to stop recording to quickly change cards and restart. Then the fan won't start blaring again and being a distraction.
 
I love my leaf blower.
 
Les C. said:
Here's a gentleman's post from today talking about artifacts from removing noise using RX2 - where it didn't quite cut it.
Click to expand...
There's an art to doing noise reduction in post. Two major secrets:

1) one tool can't do it all. Sometimes you have to use combinations of tools -- EQ, filters, noise reduction, all kinds of plug-ins from different companies -- in different combinations, for best results.

2) sometimes, multiple light passes of noise reduction produces far better results than one intense pass.

But broadband noise, which is the essence of what fan noise is, is very difficult to remove completely. Experience, talent, time, and a willingness to experiment can yield good results. I can recall a case where I spent 12 hours removing noise from a 45-minute piece -- for a charitable foundation at no charge -- and was told by the director that the results were "miraculous." They were, but it took time. (In my case, some EQ, iZotope RX Advanced, a little Waves DNS, and some C4 downward expansion in Pro Tools did the trick. There was no magic bullet, just trial and error and listening.)

I really have trouble understanding how the perfectionist group of users in this forum, who shoot in 4K or 5K for mostly 1080p delivery (until the world catches up) to get the absolute best quality, can stomach 'less than the best' audio quality due to the noise of the camera?
Click to expand...
I'm baffled by that, as well as people who want to know which is the best $99 microphone to slap on top of their $40,000 Epics. People will spend this kinda money for cameras, and maybe that much or more on lenses, then all kinds of mounts, tripods, heads, and support equipment... but then short-change the audio? That's just sad to me.

Good sound can make all the difference in the world, especially for modest-budgeted films. If the audience can't hear the dialog, they can't get into the story and characters, which kills the effectiveness of the film. No visuals can overcome bad sound.
 
As discussed earlier, this is not a problem affecting all EPIC cameras, but affected cameras are not usable with sound, Red has changed my agent, so that, actually I don´t know where is the solution of my problem.

I love Red, I'm the proud owner of two cameras, but...

This is not a joke, the problem is serious.
 
Patrick Tresch said:
Anybody knows If RED is working on a solution?
Better power management could help and be a solution?

Pat
Click to expand...
Not sure power management is going to help in this case, as when it's recording, that's when the camera's chips that generate heat are gong to be running no matter what. You might be able to throttle the clock down, but then you'd end up with the Rolling Shutter performance of the Scarlet.

Power management, I would guess, will really help with battery life when the unit is NOT recording.

Vico Martin said:
As discussed earlier, this is not a problem affecting all EPIC cameras, but affected cameras are not usable with sound, Red has changed my agent, so that, actually I don´t know where is the solution of my problem.

I love Red, I'm the proud owner of two cameras, but...

This is not a joke, the problem is serious.
Click to expand...
My perception is a bit different.

It affects all the cameras the same, but some feel that removing the noise in post, which completely butchers the waveforms so that you might as well be using a $10 microphone, is 'acceptable' - as it will be covered up by music or other ambient noise in the the scene.
 
John Marchant said:
A few of things worth noting:

• The trigger point for the fan to run at 100% to save you from frying the camera is 75 degrees as shown on the LCD.
• Anything you can do to clear the flow of air will improve both cooling and noise: It may be just me, but my EPIC-M runs much quieter with the fan guard removed temporarily. Otherwise watch placement of cables and acc's around the inlet and exhaust.
• If you're into manual settings, set the cam to run the fan at 100% while in standby - gives you the biggest bite at the cheese when you roll.
• You can adjust fan settings mid-take, no problem at all. I often ride the fan setting a tiny bit manually in a long interview take. Keep my eye on the rate of rise in temp, and if its looking likely that I will hit a critical temp before the interview is over, just nudge it up 1 or 2 % and keep my eye on things.
Click to expand...

John, I couldn't agree more! I live in Arizona... in case some of you don't know... it gets HOT! I run the fan WIDE OPEN at 100% when not rolling and that has helped tremendously! I do the same things as John and haven't had an issue.
 
Maybe someone can tell me a few things before I open my mouth and make a fool of myself. ( late edit - I've gone and done it below anyway ).

Does the EPIC/SCARLETT, use just the single fan visible in the front under the lens mount ? If so, then a power management alteration may not fix much. The fan is in the same spot as the shutter mirror and some gearing of old reflex film cameras. Much of their noise was propagated forward in an acoustic couple via the lens structure itself. Whilst this will be a much smaller effect if at all with the EPIC/SCARLETT, the position of the fan, under the lens, facing forward towards the talent and the common microphone positions is not an ideal one. There may also be some reflection of sound towards the floor off the lens which may come back up, depending on the liveness of the floor surface.

There is probably not much to be done about the fan itself. Scimitar shaped blades are about as good as it gets for quiet running. Desperation measures on noisy automotive engine driven radiator fans before electric radiator cooling fans became the standard, involved uneven spacing of the blades which changed the sound from a single fundimental frequency clean whining sound to a lower pitched fluttering consisting of two or more less intense fundimental frequencies, still loud but less intrusive. Given that the computer style fan in the camera may be a generic product, it may not be economically viable to alter the fans to this arrangement.

Air movement over the cover grille may be a separate component in the noise.

As a starter quickfix, if using the existing hardware, I would try bringing the grille covering the fan ahead about 15mm to 20mm and altering it into a cone or dome shape if there is workspace available in order to reduce pulse interaction between the fan and grille on radial structural elements in the grill pattern. A similar grille style is already apparent on some military helicopters in front of engine intakes.

A better fix might be to eliminate the grille entirely and attach an acoustically padded curved and branched outlet or inlet, styled after the outlets of small turbine engines as used in small aircraft and helicopters, to direct outgoing airflow ( if it is outgoing ) and sound away from the talent and common microphone positions. It would need to be tapered wider towards ther tube ends, to allow warm air to expand and reduce resistence against the fan if it is working in the outlet direction.

Care would have to be taken to ensure that the redirected warmed air is not reingested by the camera. This was a issue if two SI2K recorder units were placed side-by-side when connected by umbilicals to a 3D camera pair. One would feed hot air directly into the other. If mounted close on a cart, they had to be positioned side-by-side, head-to-toe, right sides in and separated about 50mm. Crossfeeding of warm air could therefore become an issue with two EPIC/SCARLETTS in a mirrorbox 3D rig it has not already.

In the P+S SI2K recorder unit body, ( yeah I know, I'm making comparisons again ), there are two pairs of small fans on each side of the body operating under filter-silencer pads. One pair pulls air in, the other pair pushes it out. There is a similar speed controller arrangement where the fans go quiet during recording.

There you go, some useless info. I know Jim is not going to gift me a Scarlett on the strength of this lot but maybe a RED badge I can show off for pose value. ( Stick it on the SI2K body maybe.).

I hope I am not diverting people from more intelligent solutions to the issue with these comments.
 
Last edited:
Les C. said:
Not sure power management is going to help in this case, as when it's recording, that's when the camera's chips that generate heat are gong to be running no matter what.
Click to expand...

I think that less power consumption also means less heat generation.

I wonder why they didn't choose a quiet solution based on the Peltier effect for exemple, it's probalbly not efficient enough...?

http://en.wikipedia.org/wiki/Thermoelectric_effect

Pat
 
I had the fortune of sitting in on an intellectual property rights court-fight over CPU cooling using water heat-exhangers and saw one of the exhibits, a beautifully precise piece of engineering. During the submissions on the development of the product, it was mentioned that Peltier effect had been examined and found to be a dead-end. I imagine the same would apply to sensor and processor cooling in the cameras.
 
Les C. said:
My perception is a bit different.

It affects all the cameras the same, but some feel that removing the noise in post, which completely butchers the waveforms so that you might as well be using a $10 microphone, is 'acceptable' - as it will be covered up by music or other ambient noise in the the scene.
Click to expand...

Maybe we´re talking about another issue, the "EPIC FAN NOISE - HEARTBREAKING" problem, raising the temperature up to 75c in less than 15 minutes, only affect to some cameras.

I use to work up to 10 epics from diferent hire company and only my one have this problem.

During recording, when the fan is not at 100% there isn´t any problem with sound, no more problem than the RedOne; I suppose someone working in ENG style, with the microphone at 2 inches from the camera ...
 
Robert Hart said:
Whilst this will be a much smaller effect if at all with the EPIC/SCARLETT, the position of the fan, under the lens, facing forward towards the talent and the common microphone positions is not an ideal one.
Click to expand...

Traditionally, when I've had to deal with HVAC problems on location, the problem is not only with the air conditioning/fan units, but also the shape and size of the vent blades. In some cases, just removing it entirely cuts down on the "whistle" of the air noise going through the vents.

In the case of camera design, I think the issue was, do we expel the hot air directly into the camera operator's face, or towards the talent? Neither is ideal. The recent upgrade does specifically mention a new fan and new heat sinks, which tells me that the company is trying very hard to solve these problems:

http://www.red.com/news/epic-m-upgrade-program
 
Hey, I'm just glad that it's better today than it was 9 months ago. I'm always grateful for incremental improvements, and I never expect a revolutionary solution. Stuff like this happens a little bit at a time.
 
Honestly, no matter how you try to rationalize this -- 5k is processor intensive, film mags had a similar time limit, takes should not be 10+minutes, mics shouldn't be within 6ft of the camera, Epic's/Scarlet's size is so small, you can get rid of it with (audio) noise reduction, etc. -- this is still a design flaw. There is no reason that digital cinema should succumb to any of the pitfalls of film. Whether you can cope with it or not is subjective, but the fact that the RED1 didn't actually suffer from heat related issues only makes instances of this problem, no matter how rare, that much worse.

Plus, if it's like this with 5k (4k on Scarlet), how loud is the fan going to need to be for Dragon/6k?
 
Last edited:
Patrick Tresch said:
I think that less power consumption also means less heat generation.

I wonder why they didn't choose a quiet solution based on the Peltier effect for exemple, it's probalbly not efficient enough...?
Click to expand...

A peltier element is not a cooler per se, it is a splitter that can help you carry heat away: if you cool down 30 degrees on one side, you have 30 more on the warm side of the peltier. Doesn't help you here I think...
 
You must log in or register to reply here.
Back
Top