Cooperman

Rory If I spent a couple of thousand on a piece of equipment, I'd want it to outlast me! Call me old fashioned, but I always thought that one of the most important things in terms of providing security systems, is to design out any possible problems related to component failures. Admittedly, domestic and industrial are two completely different ball games, but their is a school of thought that if you supply stable and reliable systems, you get repeat work based on reputation (rather than a steady revenue stream from constant service calls). Apart from existing postings, it would be really interesting to see what are the most reliable (and unreliable) DVR's and which manufacturers are prepared to offer the best ongoing support?

CCTVINSTALLER Maybe we should charter a flight, and then group book all the Brit CCTV bods, and perhaps go and stay with Rory in the Bahamas .... only joking Rory! If it keeps raining like this, we'll have to uprate all the camera housings to fully submersible, and that's just indoors!! I must admit, it got past amusing when a car floated passed me last week, on Londons North Circular Road. Worst I've ever seen!! Back to Peters problem; how about an 'off the shelf' electrical equipment enclosure, either polycarbonate (screw on cover) or steel case (hinged and lockable), usually rated to IP 55 or 65 (fully weatherproof). I'm not sure who is still in business these days, but I used to use, Sarel, leGrande, Appleby, to name but a few. Electrical Wholesalers should be able to advise on local equivalents. You just need to pop a couple of holes for cable glands, and maybe fit a small circulating fan inside to minimise hotspots. If you were that keen to build your own, marine grade or WBP ply is preferable to MDF, but the material + labour costs are probably way more than buying a suitable box. Anybody got a spare umbrella?

Don't know if this is of any use, but Daiwon manufacture a 1.6 - 3.4mm varifocal, which is claimed to give 172 degrees (horizontal) on a 1/3" CS camera, either manual iris or DD. http://www.dwopt.com

Rikky, The unit they demo'd was a very late pre production prototype. If I named the manufacturer, I'd probably never work again!, or walk come to that!! Rory I'm amazed you have the time to test out so many cameras (I must be doing something very very wrong!!). Quite a few of the cameras you mentioned are not readily available in the UK, and would I be right in guessing that the US sourced models you use, are all NTSC? I've noticed that although the technology in a couple of 'wide dynamic' models is clearly CMOS, the manufacturers are very reluctant to come clean; their is still something of a stigma attached to CMOS, particularly as the earliest examples were absolutely dreadful. As I said before, the sneak preview I had of one model, I'd quite happily have missed out on.

Tim, If you do go for the Day / NIght camera, make sure you check on the IR spectral sensitivity performance before you buy it; otherwise, it may not be any more sensitive with the 940nM illuminator, than your existing camera. You'll probably need to go to Sanyo's technical support for the information.

Rory, That price differential is a bit of a shocker; maybe we all look like mugs on this side of the pond! You don't have to answer that!! Having said that, it's still a nice camera (I've got a bit of a soft spot for Sanyo gear). I'm not aware of them having launched a Wide Dynamic yet, but I suppose it's just a matter of time. The only technical problem I used to come across with Sanyo, was the lack of any short circuit protection on the 12v DC models. Power Supply boards are quite expensive to replace, so I tried to always fit a diode to prevent any damage, resulting from the power supply cable being accidentally damaged. I've no experience of the GE gear, as they are not a major player in the UK.

Nice pictures Rory, What price does the camera sell at locally and in the States? It's currently around the £ 300 mark (equivalent US$ 500) in the U.K.

Rory, I recently saw a demo of a 'leading manufacturers' wide dynamic CMOS camera which they claimed was 480 lines, but basically looked as if it was struggling to reach 330. Have you come across any examples where, what the manufacturer claims, and what the eye tells you, are no way the same?

No problem, I can cut down on my medication now! Talk about coincidence; I just opened a mag. this morning and what did I see? A handheld test monitor used in the broadcast TV industry. I wouldn't dare ask the price, but it looks good though http://www.hamlet.co.uk/products/Flexiscope/inst/menu/frame.html

The 940's are almost (but not quite) invisible to the camera as well! Just a couple of final points. The frequency output quoted for IR sources, actually spreads above and below the quoted figure; so for example the 840nM you mentioned, may actually output light between 790 and 880nM. It would then be the lower frequencies that are 'visible'. IR LEDS have a much tighter emission than bulb based illuminators. On balance, if you can try out the 3574 IR camera (seems to have around 30% response at 950nM, 50% at 850nM, with a 0.02 lux rating @ f1.2), you'd probably be happy with it. Only thing I might suggest though, is see if you can use a fixed focal length lens (perhaps [for example], a 6mm 1/2"), rather than a varifocal. It'll generally produce a better quality image. Best of luck.

There are adaptor / convertors that fit between the PC and the monitor display, which provide an additional composite video output; you would need to use an RF modulator as well, although the eventual quality may not be that brilliant. It's about three years since I last used them, so I'll have to have a look around for some details ... unless anyone can come up with the answer quicker.

Looks like Rory has hit the nail on the head. As far as I can recall, the chip in the 3512 / 3524 is very poor for any frequencies above 850nM; I think the spectral sensitivity was something like 10% ish. Sanyo actually introduced the VCB 3574IRP as a highly Infra Red sensitive equivalent to the 3524, but that doesn't help with this problem. The bottom line is the IR illuminator is not really the correct one for the camera. So, either change your lighting or .... no, just the lighting. The driveway is fairly short, so could you consider using very low level visible light? PIR activated? If you have to change the IR unit, go for something below 850nM.

cctvfan I think I've been spending too much time in the sun (yes we do get some sun in the summer, here), because you are getting me confused now. Just to clarify, the only two 'lines' that are referred to by manufacturers, are either a) the 'TVL' which directly relates to the number of pixels in the imager, and hence the number of equivalent lines that can be resolved (i.e. image resolution), or b) the number of lines making up the picture, which is a function of the scanning system, e.g. NTSC or EIAJ (525 lines); PAL, Secam or CCIR (625 lines). If you tried an NTSC camera on a PAL monitor, you wouldn't get a picture! If you want to do a simple visual check on a cameras performance, set up the best camera you can, so that looking at a torn out page from a newspaper the page fills the screen on your monitor. If you then test any other camera it's relatively easy to see whether it is comparable or inferior. If you need to actually measure the frequency response, you'll have to buy a half decent oscilloscope, learn to use it properly, and then become a braver and more patient man than I. If you're interested in a thirty five year old Tektronix dual beam scope (with valves!) ...... I should really donate it to a museum!! If none of this makes sense, you can send the men in the white coats over to visit me.

Does anyone have any experience of using the 'cctvcad' software? Does it work, and is it worth the bucks?

I don't really know the lens you're using, I've always stuck to Cosmicar / Pentax myself, but if you are getting away with using a 1/3" varifocal on a 1/2" camera, without the corners of the picture gone awol, then lady luck is really starting to smile on you. Rory is absolutely correct that an f0.95 will increase sensitivity slightly, but as you've already said, the quoted sensitivity figures aren't exactly embarassing for your unit. I haven't used a VCB 3512 for quite a while, simply because when I first bench tested it against the 1/3" VCB 3372 (24v is VCB 3374), I actually found the lower performance 1/3" camera to produce a slightly more pleasing image than the significantly more expensive 1/2", under very low light conditions. The fact that you are getting a very nice image during the day, seems to point towards improving the lighting on the driveway, rather than spending a lot on trying to uprate the cameras performance. What IR illuminator are you using (bulb, LED, wattage, 830nM?), and how long is the driveway?

The standardised method for testing whether a camera is resolving its claimed resolution figure, is to use a test chart. If you set up the chart at the correct distance in front of the camera under optimised conditions, provided you are displaying the image on a significantly higher resolution monitor, you should be able to see which of the smallest set of resolution bars are still clearly definable. Although camera manufacturers quote resolution figures (bench tested) for their products, it's usually the poor optical performance of the lens which has an immediate effect on reducing the quality; of course, if the camera is working in low light, or is generally producing a noisy picture, this will also have an effect on resolving detail. Engineers, particularly in the Broadcast TV industry can certainly use 'scopes', test signal generators, vectorscopes and goodness knows what else to measure and optimise set ups, but for general CCTV work, excessive comparisons can eventually make you paranoid. Come to think of it, I've just realised why whenever somebody mentions lines, I immediately think back to my school days!

Commission .... what's that?? I'm a consultant, you don't think people actually pay me for advice (in my dreams!!). If you do use the studding, place a 'locknut' against the ball and socket head ( just as a safety measure) and if their is any 'air-con' or plant near the camera mounting points, be aware that even slight vibrations will be amplified over a long length of 'thin' rod, so keep it short as possible, otherwise if you use a telephoto lens you may see movement in the picture. I had a look at some split flexi conduit over the weekend and it definitely is called 'Copex', but I'm sure you have loads of nice alternatives Stateside.

Tim What lens are you using (auto iris? type?), what are you looking at, and what lighting is available at night. Are you getting a reasonable picture during the day? I've used the VCB 3512 (12v version) here in the U.K. and it's a fairly solid camera in terms of overall performance.

Peter, I don't know if this suggestion helps, but in the past we've used this technique for mounting C / CS cameras, with a thread mounting on top. Apologies for any non US terminology, but on this side of the pond, we've bought in 3 feet lengths of threaded metal rod (called 'studding' here) with the correct thread for mounting a small ball and socket head (readily available from photo dealers etc.) The cables are then 'tied' down the pre cut length of studding, and the install is finished off with a split flexible conduit (I think it's called Copex) which has an inner open tube section that covers the studding and cables, then a secondary outer cover is push fit into place, to cover the lot. Looks quite neat and tidy if done properly.

If you are looking out of a window, a) make sure the window is kept clean both sides, b) mount the camera with the lens as close to the window as possible to minimise internal reflections (use a rubber lens hood if the window is subject to vibration, or very bright internal lighting) or c) if the camera / lens is away from the window and there are reflections on the glass, use a polarizer filter to reduce / remove the reflections. If the sun shines directly on the outside of the window, it's exactly the same problem as if the camera were mounted in a weatherproof housing. Remember also that the nearer the camera is to the ceiling the hotter it gets, unless of course it's air conditioned!

P.S. I forgot to mention, I'm married to a Tottenham Hotspur supporter, so please don't mention football whilst I'm sober!

DVR Expert, I won't argue with the logic of what facilities a Digital Recorder can (and of course, not all do) offer over analogue, but I think from reading through some of the other postings on this site, it is clear that things do work differently in different countries (it's a shame not everyone registers what country they are in, as it would save a lot of second guessing when posting replies). You are of course spot on to highlight many of the advantages of Digital, but of course, if I was going to play devils advocate, I could argue that my car has cruise control, ABS brakes, a sunshine roof, and any number of other benefits, which are absolutely useless for sitting in a London traffic jam. Analog is still outselling Digital in the UK, although as the marketing of new products begins to bite, Time Lapse machines will routinely be replaced with DVR's, simply because it's the easiest thing to do. Question for you, can you see any possibility of a 'Blu Ray' format "Time Lapse" being developed in the next couple of years? perhaps as a direct replacement for ageing analog 24 Hour machines. Just curious. I think the point about 'stepper motors' is very relevant on either slow speed (48 Hour+) or where there are a significant number of multiplex fed cameras, which reduce the image uprate to perhaps one picture every three or four seconds (or worse!). The acid test is what the judge considers acceptable in relation to the severity of the offence, on the day. Personally, anything above one picture every two seconds for analogue would start to seriously concern me.

Don't know if your looking to use Colour or B/W but around 600 metres of Unbalanced RG 59 co-ax shouldn't be a problem, with the correct Video Launch Amplifier, or possibly also a cable equalising Amplifier. Another company to look at would be Tecsec at http://www.tecseceurope.co.uk

Physics was never my strong subject, but Gamma radiation is I think the most penetrative, and so it's not so much the material that causes the problems (i.e. lens glass, metal or plastic camera case etc.), but the intensity of the particle beam or proximity of the camera to a strong stray particle stream. I may be wrong but the 'Non Browning' lenses aren't necessarily going to prevent radiation passing through, so much as resist the effects which would be suffered by ordinary glass; in other words, bombard a normal lens with radiation and watch the glass go brown. As regards the degree of shielding required, this depends on the level of radiation which the camera is intended to resist. The idea of mounting the camera at a right angle behind a suitably constructed shield, and then using a mirror to view at a right angle, should ensure that only a relatively small percentage of stray particles will eventually find the camera. If you have to work with affordable but 'disposable' cameras and low tech protection, so be it. It may well work out much cheaper than using a very highly engineered radiation resistant housing, unless of course the budget is available. One link for Non Browning lenses is:- http://www.lens-design.co.uk/miniature_security_camera_lens.htm and I'm not sure if a company like Dennard would have any radiation resistant camera housings up their sleeves. Certainly worth a try.

I think from memory there are about ten or more facial recognition 'engines' currently in use, and all with varying degrees of success (or failure, depending on your point of view). The technology is already in use in quite a wide range of locations, usually related to transport, construction or retail industries. There is a UK based company called Aurora which might be worth a look. http://www.facerec.com I've also got a note about some published facial recognition results, but I haven't actually got round to looking them up yet. http://www.frvt.org/FRVT2002