NotoriousBRK

The differing effective focal length is a factor of different sensor sizes, not lens properties. A lens is rated for the size of the "sweet spot", or area where it achieves a flat focal plane. It's more expensive to make a lens that can focus clearly over a 12" area than a 12 micron area (using extremes to illustrate a point). However the 12" lens would work fine on the 12 micron sensor, it would just be overkill. Same thing for a 2/3" rated lens. It would work fine on a smaller sensor (assuming that it mounts properly, etc.).

Xantech has been the go-to company for IR repeater kits forever. There are some cheaper options, like Sewell, but I've never personally tried them. You can find them on Amazon, or any place that sells A/V equipment for around $100.

It's nowhere near as easy as it used to be. You *can* do this, but it will forever be a headache as the cable companies tend to move and re-allocate frequencies more often now. You'll need a QAM modulator, last time I checked, they were around $1,000 US.

You might (or might not) find this document that I wrote to be helpful: http://support.videoiq.com/docs/MobileBroadband.pdf Note that since I wrote it, it is geared towards our cameras specifically, but most of the concepts are generic to 3G/4G connections. Using h.264 you should be able to get a 2.1 Megapixel 10fps image stream in about 3Mbps of bandwidth, which you probably WON'T get as an upstream on a 3G or 4G connection. Your best option is to record locally and export/copy interesting video out when needed. Use a lower res stream for basic viewing.

A powered splitter takes an incoming signal and essentially regenerates it so that each output port is receiving a "full strength" signal. A passive splitter takes one electrical input (the signal) and splits it into two, which is going to cause some loss of signal strength. It's not always easy to predict when the loss will be visible, but the longer the cable run, the greater the chance the passive splitter won't work.

You can also put a desiccant packet in the dome before you close it up (there is probably one that came in the box with the camera). Just make sure it stays out of the way of the lens. That will absorb any left over moisture.

There are a few companies that claim to do this. I've heard mixed results from all of them, but one that might be worth checking out is: http://www.flonomics.com/ I have no vested interest in them (if anything they are a slight competitor), but it might work in your scenario.

Why? What exactly are you trying to accomplish with this cam? What is the crime risk that you are anticipating? The typical front door camera would capture very little crime footage, so even the sharpest picture isn't going to necessarily provide the evidence footage you think it will in a majority of cases. The front door camera I want to provide facial identification. I fewl my neighbors camera or DVR is not sufficient in quality. This is why I want your expert opinions. 1. How high would you mount the camera? 2. Is TVL beyond 480 really not worth it as mentioned by one of the posters? Thanks. OK, but I think you are over-valuing the facial recognition part is all I was trying to say. You might need more than 1 camera if you *really* want to cover all options. But I would mount at least 1 camera at about 5.5' high, looking slightly up at the point where a person would most likely stand on the porch. And yes, the high TVL count cameras are mostly marketing gimmicks. Personally, I think that if you really think you need such high quality images an analog camera/DVR combo might not be the best option anyway...

Baluns don't change the impedance of the cable. They convert an UNbalanced signal into a BALunced signal (this is where the name "balun" is derived from). This can give you an impedance *match*, but it doesn't change the cable itself. The following is a simplified explanation: In an unbalanced signal, the signal (data) is measured as referenced to ground. The ground side of the cable tends to stay at 0 volts (because it's grounded ), so if you have a source of noise or interference it affects only the signal carrying conductor and not the ground/shield. If you have a REALLY good shield, you can prevent most sources of noise from affecting the signal carrying conductor. If you don't have a perfect shield (and you never really do), then interference can be imparted on the signal, which we see or hear as distortion and noise. In a balanced signal there is no ground. The signal/data is measured as a delta between the two conductors. EG: in data terms a "1" might be a voltage delta of 5 volts, and a "0" might be a voltage delta of 1 volt. Because the wires are unshielded they may pick up more stray signals. However, they wires will both pick up the SAME stray signal. So the noise may impart a 3V spike on the cable, but it will be picked up up by BOTH wires. The delta between the two wires is still the same (eg: 5 volts), but instead of 0V on one wire and 5V on the other it's now 3V and 8V. The twists in a twisted pair cable ensure that the two conductors in the wire pair being used in the signal path stay in close physical proximity to each other so that they are always exposed to the same noise source at the same time. The eHow link you posted is not completely accurate, eg: Cat5 has no limit on length of a run. *Ethernet* on a twisted pair network has a cable length limit of 100M, and this is based on timing (to detect packet collisions), not on the quality of the cable, or signal strength or anything like that.

What is the field of view of the camera, and what level of accuracy is the customer expecting? There are some server-based analytics applications that could give you about an 80-90% accuracy rate if the camera is mounted directly overhead (essentially dedicated to the task of counting). If it's mounted like a "normal" security camera, you can probably get about 60-70% accuracy.

Soundy is correct. You haven't provided enough data for anyone to give you any useful answers. First there is the stream format. It could be MJPEG, MP4, or h.264. Bandwidth variances between those 3 formats could yield results ranging from 25Mbps (MJPEG) to 4Mbps (h.264). Depending on the format, more fps doesn't always equal more (or significantly more) bandwidth. Do you have specific camera models and applications in mind?

Why? What exactly are you trying to accomplish with this cam? What is the crime risk that you are anticipating? The typical front door camera would capture very little crime footage, so even the sharpest picture isn't going to necessarily provide the evidence footage you think it will in a majority of cases.

You can't adjust it manually, but the iris is typically DC-Driven. You *could* build a little device with a 9V battery and a potentiometer to control the lens externally. You used to be able to buy a lens testing device like this in many places, but I haven't seen one for sale in several years. All the DC iris lenses I've seen default with the iris closed.

That looks like the results of a poor/overly aggressive compression algorithm.

What state/country/province/whatever are you located in?

It absolutely matters. Even if all the materials were free, it still doesn't make sense to make your own patch cables. You never see any of the more professional organizations out there trying to DIY a patch cord. They know that in the long run, the time, materials, and peace of mind are not worth saving $2 on a patch cord.

in most cases you would already have the cable or like most keep good lenth from off cuts from installs. so apart from the ends it costs nothing. But patch cables should be made from stranded cable, and backbone cabling is done with solid-core. So, who has bunches of stranded cable laying around?

Rory, I recognize there are unique factors to business in the Bahamas. I'll admit, I'm not seeing the same trends emerging as quickly with my dealers in the Caribbean, but this is becoming predominant in the US and EMEA markets.

5ft patch cables are $3 in bulk purchase. How can you make one cheaper than that? In bulk, stranded Cat5e cable is about 15 cents a foot (solid is much cheaper, but solid is never used for patch cords). An RJ-45 plug for stranded cable is about 4 cents, and a strain relief is about 2 cents. So a 5ft. patch cable has about a 90 cent parts cost alone. Then there is the time to actually assemble the cable *and* test it. Which is 3 minutes at least. So, a really skilled person could make 20 an hour if you were lucky. Or, a $20/hour tech adds $1 to the cost. This is of course assuming that there are ZERO errors in any of these home-made cables (causing time and materials to be lost). I don't see the justification there. And even if there *is* a slight cost savings, it's going to be one of those "penny-wise, pound-foolish" decisions in the long run.

They're both universal, the benefit of a standards-based structured cabling plant is that you can use it for all kinds of devices. In fact, T568B came about by the "phone company". When this stuff was all being ratified in the late 80's/early 90's time frame, the original wiring color code for 8C twisted pair cabling was just "T568", which is what we now call "T568A". However, by the time the whole thing was settled, AT&T (and this was back when AT&T had clout) had been doing what became known as "T568B" color-pattern punchdowns for several years, across hundreds of thousands of lines. They basically said "we can't be put in a position where a customer could think we did things in a non-standard way", so T568B was shortly introduced as an "alternate acceptable wiring pattern". The US was ahead of the curve on these rollouts, with AT&T at the forefront, so this is why you see 568B primarily in North America, and T568A everywhere else.

You terminate the solid-core cable into a 1-port outlet. Generally inside of a 1-gang box, and then run a weatherproof cable from that to the camera, or mount to camera to the gang box. Because he is in the US (I think). T568B is by far the predominant wiring pattern in the US. T568A is International

No because Beta Max was actually superior to VHS. Where as HDcctv is not superior to IP. Except that beta max couldn't hold most feature length movies. So while it had an arguably slightly better image at times, it didn't fit the needs of what the public actually wanted overall. Hey, that sounds a lot like hdcctv

Here is my suggestion: don't do this (and I'm sure I'll get flak for this). Backbone cable in a structured cable plant is *punched down*, not terminated with RJ-45 plugs. There should be almost no scenarios where you are crimping RJ-45 ends on cables. You buy patch cords for things that need crimped ends. RJ-45 plugs come in many varieties, some are for solid-core cable, some are for stranded. You often end up with the wrong ones, and "universal" plugs aren't. Solid core cable is not meant to be continually flexed. It is meant to be terminated in a stable fashion, secured to supports, conduits, etc. Patch cables are stranded and meant for plugging and unplugging and moving connections around. You don't *make* patch cables, you *buy* them. Granted, it is valuable to know how to properly crimp a cable in a pinch. But if you're doing this more than 4 times per year, *you are doing it wrong*. Flame away...

CCTV doesnt replace a burglar alarm, plus beams are not expensive, alarm systems are peanuts compared to CCTV. Funny, but I had a meeting just yesterday with a rather large US bank, and they disagree with you. Burglar alarms don't offer actionable signals anymore in the US. First question from the police on any alarm (even from a bank) is "is it verified", meaning "do you have video". We are replacing burglar alarms with cameras. Every single day at this point.

Forget doing a wireless IP camera, get an IP camera of your choice (suited to the project) *and* a wifi bridge device that runs on 12VDC (there are many available). I'd probably look at one of the Ubiquiti Nanstation 2 devices. They are outdoor-rated directional wifi devices http://www.amazon.com/Ubiquiti-Nanostation-LOCO-Outdoor-802-11g/dp/B004EGI3CI Plug your camera into the Nanostation and you now have a wifi camera. You can put the camera where it needs to be for the best image, and position the wifi device where it needs to be for best signal. Also, if you want to power the camera off of PoE, check out the Air802 12vdc PoE injectors: http://www.air802.com/dc-dc-converter-poe-injector-with-10-26vdc-input-48vdc-poe-output-din-rail-mount.html