Soundy

You can't. You MIGHT be able to see the problem with an oscilloscope, but that won't fix it. Here's the problem: with baluns, your signal is running through a pair of coils on either end. That effectively makes for a very, very long piece of wire. The signal ground is shared with the power ground at the camera, and the power ground then makes for a relatively short ground path. Between the two, you now have to separate paths to ground for each camera, one relatively short, one extremely long. When you tie all the cameras to the same power supply, you're giving all of them a common ground point over the short run. They also have a common ground at the DVR, with the long runs. This makes for a large number of mis-matched ground paths, which leads to all manner of induced noise in all the cameras. So again, the solution is simple: remove central power supply, replace with power bar, and use an individual transformer for each camera. That way, they all have a power ground that's isolated from the other cameras.

What I've done in a similar situation (not as long a distance, mind you) is to run the serial data over a pair in an unused network line, and placed the TVS close to the DVR - keep in mind that the serial data only actually requires a single pair, from the TX and GND pins at the paypoint, to RX and GND on the TVS. Then run the video and power using baluns over another unused pair (even in the same cable). The other thing I've done in some instances, is to put the TVS under the counter close to the paypoint computer, where I can use a short serial cable to connect them, then run video from the camera to that location, and then out to the office. The camera power can then be home-run to the DVR, or just to a point under the counter. Remember that a text-inserter doesn't care what kind of DVR you're running, if at all - it's just a separate box that overlays the text on the video and then spews that video out. It will look the same whether you're recording to DVR, time-lapse tape, or just viewing it.

Hmmm, nobody's paying attention here... the OP wasn't talking about standard CCTV baluns. Passive VGA baluns can be tricky things. Most recommend STP rather than UTP, and most support limited resolutions, especially as the runs get longer. Some have phase-adjust screw, others have a button to force an automatic re-sync once they're connected... and a tap of the LCD's own auto-adjust button may be necessary afterward to align the display. All of them should state in the documentation what resolutions and refresh rates they support over varying cable runs. I've found usually you just have to start at lowest res (800x600) and work your way up to find the best.

Well, it kinda depends where the TVS is positioned in related to the baluns. If you're using the baluns coming OUT of the TVS, then no, you don't need an extra pair, because the text is already a part of the video. If you want to run your serial text info along with the video over your twisted-pair, then yes, you'd need an extra pair on the balun... baluns designed for PTZs would work for this, as they have an extra pair normally intended for serial control signals. However, in my experience, the TVS is usually not located anywhere near the camera, so there'd be little advantage to feeding it over the same line as the camera...

Don't see why it shouldn't work, since that appears to be exactly what it's designed for. It SHOULD work with any DVR (card or standalone) that supports PTZ camera control, since all it's doing is taking the standard serial control output and feeding it over the coax. BTW, I believe "Coaxitron" is the protocol you're talking about - searching with that term may yield you better results

That's not new OR different...

Just a thought on this: if voltage drop proves to be the issue, you can try this method, but should consider putting the regulator in the case WITH the camera... then the voltage can drop up to a full 10V or so over that 100m run, but the camera will still get a solid 12V. Remember too, with an IR camera, you'll see substantially more draw once the IR LEDs turn on, and over a 100m run, that will cause even more voltage drop. Again, a higher supply voltage and a regulator right by the camera will counter this.

He wants to watch the crowds lined up at the gates of his fairground. Presumably, it's either daylight, or the area WILL be quite well-lit while people are attending.

Vigil will handle all your criteria as well - 16- and 32-channel analog systems are the norm, as well as analog/IP hybrid systems. Their software has a fully-functional 30-day demo mode and can be installed as a straight NVR (no capture hardware required). see http://www.3xlogic.com Being PC-based, storage expansion is easy though multiple internal drives, or external USB, Firewire, eSATA, and/or NAS storage solutions. One thing to be aware of with your "record to DVD" requirement: most systems now will support DVD discs as an export target; however, all of them that I'm aware of will simply export files to the disc (whether in a common format like AVI, or a proprietary format). I've not seen one that will actually author a home-playable video DVD directly out of the DVR. I suppose they exist, and it certain IS possible, but it's largely impractical due to the time required to transcode, the inferior compression provided by MPEG-2, and the fact that megapixel video would need to be seriously down-sized to meet DVD spec.

Check with any local CCTV suppliers (particularly wholesalers)... manufacturers often put on seminars. I know our local ADI/Burtek warehouse has manufacturer training sessions almost on a monthly basis - Pelco one month, National the next, etc. You could also check with some of the manufacturers and see if/when they have any sessions coming up in your area.

Actually, to get 2X "better" than 3MP, you'd have to go to 12MP... double the horizontal AND vertical resolutions.

What can i do to solve that problem? Very helpful thanks a lot! See bolded text above... They do work perfect individualy! What can i do to solve that problem? Just keep it that way... that's the only thing I've found that fixes it 100%. Again, remove the central power can, mount a power bar or two as needed, and just be done with it. I've fiddled with filters (chokes and caps), assorted different grounding strategies (PSU to DVR case, video input shields to case, shields to PSU, any or all of the above to earth ground, and completely lifting all grounds), all to no avail. Different tweaks improve some cameras and not others, but the only thing that clears up ALL of them is having a separate transformer for each one. The only other 100% solution, as noted previously, is to replace them all with 24VAC cameras and a 24VAC power supply.

Well then, what specifically IS a "license plate camera"? Make and model? Special features?

Uhh... search button in the toolbar?

You already have the voltage requirements: 240VAC to 24VAC. Find the current requirements for the camera itself and make sure you get a large enough transformer to supply it (derate by at least 50% to be safe - ie. if the camera wants 1A, use a 1.5A transformer).

And on that note, another consideration: any lens like that is going to be rather pricey on its own - probably figure another $150-$200 *on top of* the cost of the camera itself. Realistically, the only near-100% reliable way to capture plates is with a strobe. That's how border patrol does it, that's how law enforcement does it (photo-radar and red light cameras), and they're pretty much impossible to beat - Mythbusters tried at length to beat the speed cameras, and failed every time. When you're hitting the plate with a 1/30,000th-of-a-second burst of ultra-bright, tightly-focused light, shutter speed and vehicle movement becomes pretty much irrelevant - even a $100 camera should work in that instance.

Define "cooler"? It's been an early summer here in the eastern 'burbs of Vancouver, highs most of the 33C-34C range (that's 91-93 for you 'merkins)

I do not think is is a lack of standards perse as more of the price range that people are willing to spend. Well, to put the two thoughts together: it's the lack of standards, within the price range that people are willing to spend. Upper-end stuff does use established standards... unfortunately those standards are themselves pretty upper-end and thus rather pricey.

I've installed lots and lots of megapixel cameras. They don't REQUIRE "special" lenses, but what you find is that the higher resolution starts to show up the imperfections in cheaper lenses (poor sharpness, etc.). Thus, most manufacturers have a higher-quality "megapixel" lens line that will generally yield crisper, clearer images. Storage requirements do increase, as even a lowly 1.3MP camera is four times the resolution of 4CIF/D1. HOWEVER, consider that a single MP camera can, in some instances, take the place of two or three analog cameras, and the difference diminishes. And really, the space argument is a non-starter - I mean, why record at 4CIF or D1, when it takes so much more space than CIF? Answer: because you get a better quality image with better detail. More detail will always require more data and thus more storage. If it's that big a deal, just record at 160x120 and one frame every 10 seconds... it's all the same, right?

It probably maxes out at 50mm because that's the biggest lens that will fit inside the dome housing. Instead, look into a WV-CP484, the "box camera" version of that dome. That will let you use pretty much any available C/CS-mount lens that suits your needs, with all the same features (SDIII, auto-backfocus, etc.). Right now I'm looking at an f/1.8 5.5-82.5mm lens that would suit this use nicely. Mount it and the camera in a good environmental housing (Pelco EH3512, for example) and you're good to go. One consideration: with capturing moving objects like this, you want the highest shutter speed possible to avoid motion blur. Especially at night and in low-light situations, most cameras will need to drop the shutter speed to allow more light in. It's important to find the "fastest" lens (ie. with a larger maximum aperture) that you can - the more light the lens allows through, the faster the shutter you can use. In this case, f/1.8 may not be enough; you may need to try for f/1.4-f/1.2 or better, if they exist. That may also require avoiding varifocals and going for a fixed-focal length lens (as varifocal construction makes for larger, more complex and more expensive designs). Tamron has a handy lens calculator on their site - I plugged in the viewing distance as 125ft. and the field width as 8ft. and got the following:

That's a pretty extreme example. A *single* 4MP camera requires very little computing power (by today's standards) to handle recording it. PoE is used only as a conveniences; it's not a necessity. Setting up a complete network and static IPs is not a necessity, especially for a single camera; neither is QoS. A single IP camera can be very happily added to anyone's home network, plugged straight into any off-the-shelf home-grade broadband router without messing up Internet surfing or VoIP phones, and can be recorded with free or very inexpensive NVR software installed on any existing PC in the house... or all of them, for that matter. Most cameras will pull an IP from the router's DHCP server, so little or no configuration is needed. The only added expense *required* is the camera itself. Well, you could ask yourself the same question about your car, and decide to just walk everywhere you go. I mean, they run out of gas, you have to know how to fill up every few hundred miles, you have to get the oil changed and stuff fixed now and then (mechanic = "tech support")... such a horrible hassle and expense! The problem - as has been covered here many times - is the video standards and encoding used. To do it over your cable, you required new hardware (high-def tuner box, generally)... to do something similar with CCTV would also require new hardware. How much did your HD cable box cost? Around here, the basic ones start at $400. Now multiply that by the number of channels you want to record... Of course, that then requires the camera manufacturers to get onboard with a new standard... and the analog/digital converter manufacturers (ie. the people who make the capture cards). And then you have to get everyone to actually agree to and settle on a new high-definition analog-video standard that they're going to use for both cameras and capture devices, or you end up with a lot of proprietary gear that only works with stuff from the same manufacturer. So let's say, to keep it simple and stick with established standards, you actually decide to use HDTV spec - 1080p at best. That gives you a resolution of 1920x1080, or about 2MP. *At most.* That will also require either a component-video feed - ie. THREE coax cables - or a very expensive HDMI cable that's limited to about 25-30' in length. Or you do it like the cable company, digitize the signal, then modulate it for transmission over RF, only to demodulate it on the other end, extract the digital programming, and then spit out analog video from that... seems kind of extraneous when you can simply send the digital signal directly over IP. Well the difference is, megapixel wasn't right FOR YOUR SITUATION. The way cctv(non)expert goes on about it, you'd think they were never right for anyone, for any purpose, at any time. Would one be right for you if you had a large, brightly-lit area that you needed a lot of detail on? Well, as your criteria change, so do your solutions. $1k+ for 5MP may be overkill for that; $300-$400 for 1.0-1.3MP may suffice. By the same token, a really bad high-contrast backlighting situation may call for a $600 Super-Dynamic III or similar type analog camera, because of its special capabilities (as it did in a situation I just addressed a couple days ago, in fact). We even recently talked a client out of using an IP camera for just such an installation and going with a CP484 SDIII analog camera, because we knew the IP options wouldn't handle the severe backlighting this camera would be faced with. Simply dismissing an entire branch of technology because it doesn't suit *one specific* purpose is, frankly, ridiculous.

So various technical and legal concerns aside, the biggest problem to mixing-and-matching wireless cameras is the complete lack of standards throughout the industry... unless you use WiFi.

In addition to that, some cameras have the capability to send still and video clips by email or ftp (or both), either on a scheduled basis or an event basis.

http://lmgtfy.com/?q=cctv+lens+calculator

What can i do to solve that problem? Very helpful thanks a lot! See bolded text above...