musher

Hi, I'm playing around with a few IP cameras here at home and had some questions about best practices with regard to protecting my data and camera power systems from lightning induced surges. So far I've installed several IP (megapixel) cameras at various locations both inside buildings and under the eaves. I'm driving everything off of an altronix 24vac supply with 8 fused outputs. The Altronix is plugged into an APC NS1250 to surge protect the system and keep the cameras alive during power outages. The NVR and switch is plugged in to a different APC NS1250. Everything's powered by a GFCI circuit. The data system is shared with our dsl connection and home computers. I'm considering mounting a camera along my back fence line. This camera would be about 120' from the nearest building. Planning this installation got me thinking about lightning and I realized I don't have a very good grasp of best practices with regard to limiting lightning damage in my system. My focus so far has been on the fence line camera since it would be mounted up on a wooden post away from buildings (probably disguised as a wooden birdhouse for aesthetic reasons). Data and power to that camera would be run above ground along the wooden fence. I have started to wonder a bit about the rest of my system as well. For the outside fenceline camera, I ordered a couple of 100mb Cat5 to Fiber media converters and a fiber patch cord. My plan was to run the data line in outdoor rated cat5 from the camera to a point inside the building, through the fiber to isolate any induced surges, then into my switch. Some questions that I'm puzzling over are: Controlling Surges on the DATA side: 1. I'm pretty sure the fiber interconnect is about as good as I can get for surge isolation on the data side. Is this right? 2. Is it acceptable practice to carry the copper into the building and jump to the fiber interconnect inside, or should I figure out how to make that break outside the building? 3. I've started to wonder about existing buried cat5 between buildings. I’ve currently got a few cat5 cables installed between buildings in PVC buried 3-4’ over 20’-30’ runs. Does the burial depth mitigate any lightning surge danger there, or do I need to seriously consider pulling fiber in those conduits instead? 4. I’m also wondering about outside cameras mounted under the eaves. Should these be considered a risk similar to the fenceline camera with regard to carrying lighting surges into the network? 5. If the answer to #4 is yes then do you have any suggestions for isolating them? Perhaps a dedicated switch for the cameras tied to the main switch with a gigabit fiber interconnect? Controlling Surges on the POWER Side: 1. I’m running 16/2+ground for camera power pretty much direct from the Altronix to the fenceline camera (as well as the others). The ground line is bonded at the altronix to the building ground. Am I exposing all the cameras powered by the Altronix to simultaneous lightning damage if a single camera is struck or has an induced surge in the power line? 2. If the answer to #1 is yes, then are there ways to mitigate this risk? I could swallow toasting a single camera due to a strike, but I'd like to lower any risk of simultaneously smoking all the cameras tied to the power supply if possible or practical. I'd welcome any suggestions you might have on all this. I do worry I've got just enough knowledge here to get me in trouble. Thanks

Your provider may be willing to sell you a public IP for that connection as well. Its worth asking.

Pins 123 and 6 are used for ethernet. The colors of the unused pairs may vary depending on how your cable is wired. Look up tia/eia 586a/b

I think he's talking about using the unused pairs to host an analog signal. Might be a trick to do a neat installation at either end. Need to split out the pairs with different functions. It's done successfully with phone signals, I have no idea whether the analog video signal would induce errors in the data pairs. Test it and let us know! PS-- I assume you're running 10 or 100MB here. Gigabit uses all the pairs.

It probably needs the additional amperage to run the plasma gun for active countermeasures.

Yes, it wasn't clear to me whether he needed 3.6 or 1.5. Curious that the manufacturer supplied a 1.5A wall wart for that camera. If 3.6 is needed, he needs an outlet nearby or some heavy wire.

Yeah the 24vac->12vdc converter I mentioned is only rated to deliver 1.5A at 12vdc. You need to have at least 20vac at the camera for the converter to work. 24vac at .75A should be fine for 200' of 18/2 (<1.8v drop)

Feed your 24vac to the siamese and use something like an AT12015-D01 at the camera end to get your 12vdc.

Sounds like we're neighbors then. Temps are almost back up to where you can bend wires again. Here's a pic of the foam install. Not real pretty, but it's not too obtrusive when painted to match the building. I'd advise using a circle cutter instead of a jigsaw like I did. I

If you're stuck with 12vdc for the cameras, you can also run 24vac over the line and rectify to 12vdc at the camera end using something like an AT12015-D01 ac/dc adapter.

Really? I think the credit card thing must be new. I had a grandfathered free account for years (with 4 domains) and went on vacation the week before my "reminder" to verify activity. I got back about 12 hours after the deadline and my account was gone. I was able to set up a new free account with 2 domains and no credit card. This was back in July, I think. I probably ought to just pay them. It's a handy service.

Hi, I'm running a couple sanyo hd3500s and an acti 7411 outside in Fairbanks. I mounted 1 sanyo with a 2" blue foam donut fitted around the outside of the dome casing to hold in the heat a bit, the other sanyo is just surface mounted. Haven't seen any difference between the 2 so far. This will be the 2'nd winter for these cameras. -52F at my house this morning and all cameras reported for duty. I am running everything on 24VAC backed by a UPS to keep things alive (and warm) if the main power should fail for a bit. The temps are well outside the camera specs, but so far so good...

Hi, Does anyone know if there's a way to deactivate the power led on an ACTI 7411? I've set one up in a birdhouse but at night, at the right angle, the led glares out the "bird hole" like a satanic sparrow. I've perused the camera menus but can't find a setting to do this. Figured I'd ask here before I gave up.

Hmm, Korgoth pretty hi-tech for barbarian. Power led now reports 0. Of course, I'm 350 miles away so I can't really see it. Easier than climbing the ladder. Searched all over for LED, but it never occurred to me to search for power_led. Found the document right away with that. Of course, I gotta go up and remove bpzle's solution when I get home. thanks.

HA! As it turns out this is exactly the fix I employed this morning. Different brand of tape though. Hope I didn't screw everything up.

Nah, we've even got lightning up there now. Courtesy of global warming according to the eggheads up at the university.

Sort of. The insulating capacity of air allows charges to build, but at the moment of a strike, the lighting is traveling through air that has been ionized/broken down into plasma which is conductive. One comment here did strike me though is I was definitely thinking about this a little backwards. Seems like a fiber patch between the outside cable and my network would likely stop the surge from following the data into the switch, but clearly I need to also provide a path for that surge to follow otherwise the next step a surge might take could be as undesirable as smoking my network. I hadn't quite thought that through.

There are some trees nearby much taller than the fence. I can always find something to do with the fiber converters--maybe I need a network drop out in the field somewhere... I'll run down some shielded cat5 as you suggest, I think. Thanks for the help on this.

Thanks for the feedback. It's easy to get overly worried reading about this on the internet. Kind of like reading bear attack stories before going to Alaska. I'm mostly trying not to do something really stupid to myself. As to dmillers questions. The fence is 8' wood, the post for the camera will be about 12-14' I'm running unshielded copper cat 5 from the camera back to the building (above ground along the fence). I bought 2 $50 cat5->fiber media converters and a 6' fiber patch cord. I was going to run 120' of copper from the camera into one media converter, through the fiber to the next media converter, then on to my main switch with copper. I figured 6' of glass would stop a surge leaving only the camera and 1 media converter at risk. I wondered whether to put the first media converter outside the building and run the fiber through the wall or whether it was ok to run copper from the camera directly inside to the first media converter. Sounds like my data side is probably ok. On the power side I'm still not clear if a surge would be likely to run back up the power cable to the altronix power supply then back out the other 24vac power feeds to the rest of the cameras. It seems to me that everything tied to that power supply would be at risk. The camera takes 24VAC and has provision for a ground. Manufacturer says use of the ground is optional, but I went ahead and home run tied it to my building ground on the other cameras since it was there. Been using 16 ga for that. Clearly not a lightning protection type of ground.

PM sent on 5105+Arecont lens.

Had good luck with Seagate Constellation ES 2 TB 7200 - ST32000644NS so far. Only a sample size of 2 though.

Rory, Ya, me too. Clearly any non-trivial software will have bugs. A closed source model does not necessarily protect software from hostile exploits. It's not clear that hiding the design contributes more to security than opening the design, though the evidence seems to point to actively used open source as being more secure/less buggy. I agree, lots of users base their choice on price alone. That's probably why Walmart does so well. Still, I'm convinced one can sometimes make a sound business case for open source vs commercial software based on 'total costs'--which would include the hidden costs associated with quality problems. From a developer's standpoint it's an interesting challenge to decide whether to free or open source license your work vs keep it as commercial software. Pros and cons on both sides of the fence there as well. A bit dated, but here's an interesting article on the subject of bugs--windows vs linux. http://www.wired.com/software/coolapps/news/2004/12/66022

One of those gentleman types who lets others do his cutting eh?

The view you express is one I find common among managers dealing with software in business. It is based on a bit of a misunderstanding of the open source model. What Quantum refers to by "peer review" is not 10 programmers who happen to be on a development team, but thousands of individuals (or tens of thousands or perhaps hundreds of thousands or millions in the case of software with a large installed base like apache http) who are willing to take the time to look at software before using it or when figuring out problems. This model relies on leveraging these large numbers of reviewers to ensure security (and functionality) by design rather than incorporating security through obscurity as part of the model as in commercial closed source software. At least one flaw in the 'keep it secret' model is that it is susceptible to a well known problem with smaller teams of folks who work with each other--everyone tends to start thinking about the problem in the same way. This has a tendency to allow certain defects to slip past the whole group undetected. You are correct that obscuring the design incrementally increases the difficulty for taking hostile advantage over the system, but I submit that a look at your periodic Windows updates demonstrates that approach is not perfect. Personally, my experience is that open source is generally more reliable and bugs/vulnerabilities are addressed more quickly. It also has the added advantage that you can fix a bug yourself if it's a mission critical problem for your situation. Ultimately, I think the decision about open source/vs commercial closed source is not best answered by asking "what's better". I think it's better answered by asking "what resources do I have to take advantage of the strengths of either". Interestingly (to me anyway), I am dealing with a bug in commercial software interacting with the HD3500 (which is what brought me to this thread). While I'm very happy with the support I'm receiving and with the camera software, I suspect it would be more efficient for everyone involved if I could just debug the code locally since the vendor cannot reproduce my problem. Alas, since it appears the problem likely lies in a Windows driver, that option is not available. And now we've at least brought the discussion back to the Sanyo camera!

Sorry 3RDIGLBL, I didn't make what I meant clear. Your client side javascript will call a php script. Same script for all clients That script will return a jpeg to the client. You do the logic in the php script to decide what file to use/return. pseudocode would look something like this: getcurrentimage.php <?php include_once("flagmon.php"); $fileDir = '/path/to/files/'; if (isset($imageflag) && $imageflag==1){ $file = $_GET['file1.jpg']; } else { $file = $_GET['file2.jpg']; } if (file_exists($fileDir . $file)) { // Note: You should probably do some more checks // on the filetype, size, etc. $contents = file_get_contents($fileDir . $file); // Note: You should probably implement some kind // of check on filetype header('Content-type: image/jpeg'); echo $contents; } ?> Your script to capture the streaming will 1. include flagmon.php 2. if $imageflag=1 then you write the new image to file2.jpg, otherwise write the new image to file1.jpg 3 now overwrite the php file flagmon.php with a new version that flips the value of the imageflag You could alternately just write an empty file called imageflag.jpg and test for its existence. If it's there do one thing, if it's not do the other. Delete the file every other image loaded from the stream and recreate it on the other events. Different users may well be out of sync by 1 image for 1 refresh depending on how timing works out, but you won't be able to tell unless you run them side by side, so I'm not sure it matters. Ran dogs for about 10 years, recreationally. Did one race and spent the whole weekend thinking about how I could be up in the mountains camping instead, so I didn't do it any more. Got out of dogs a few years ago.