survtech

I thought HD-TVI, HDCVI and AHD also use coax???

When daisy-chaining RS422/485 control for PTZs, best practice is to terminate the end of the chain while leaving termination "Off" on PTZs in the middle of the chain. However, unless you are daisy-chaining a boatload of PTZs, under most circumstances you can terminate all PTZs in the chain.

Not necessarily. When I started my job in the latter 90's, we had around 500 VCRs recording 8 hours on T-160 tapes in SLP and a few others recording the signal from 16-channel multiplexers, also at SLP speed. While the customers at my previous job did use either time lapse or 24-hour "Real Time" VCRs, the casino business was required by law to record at least 20fps, which coincidentally was the frame rate of 24-hour Real Time VCRs. We chose to record 8 hours at SLP and change tapes three times a day (once per shift) since Real Time recordings were lower in quality. Note, we didn't have to actually replace tapes that often. They were good for many record/rewind sessions, as long as they were handled properly and the VCRs were kept clean. That was one of my jobs: cleaning and repairing VCRs. I say record/rewind sessions because the vast majority of tape recordings were never played back. Recordings were mostly retained for a week, then overwritten with new recordings. Real Time recorders also cost more than the consumer-grade VCRs we installed in 1999. After testing a number of brands, we settled on the Mitsubishi HS-U445/446/447/448 models. Of interest, by the time we switched to NVRs in 2003, it took two people 4 hours to prep, change, rewind and file the tapes from the over 700 VCRs we were up to.

Could you test recording in TCP/IP Unicast? That is the recommended mode by IndigoVision.

Those cameras use C-mount interchangeable lenses. I can't say what lens you are actually using with them (or even if you are using a lens at all) but any C-mount lens would work.

Cars with no motors? Like this?

Not sure I understand your problem. There are so many possibilities it's hard to pin down any single one. First, do any cameras drop off the system when in "Live"? Second, what make and model cameras? Third, what stream are you recording, TCP/IP Unicast or something else? Do you get a message saying "No footage is available for the selected camera"?

I agree, at least in part. Coax Seal http://coaxseal.com/products/ would do the job but it does tend to deteriorate in certain climates and can be difficult to remove if you need to access the connection later. An alternate solution would be to place the connection in a waterproof junction box with suitable waterproof glands for the cables.

I use West Penn 815 myself. http://www.westpenn-wpw.com/index.php?option=com_cable_finder&func=PDF_Info&download=815

The only way I know of is via the dipswitch settings on the dome drive itself.

I doubt it. PTZ pan/tilt speed is typically a function of the PTZ itself, including its onboard programming. Some PTZs allow memorized "Tours" or, as Pelco calls it, "Patterns". That would be your best bet for what you want to do. At least with the Pelco Spectra PTZs, you set the PTZ in "Program Pattern" mode, use the joystick to pan, tilt, zoom, etc., then save the pattern. At least on Pelco Spectra PTZs, the memorized pattern will run at the same speed you used when programming it. The complexity of the pattern determines how long it can last but in my experience, a pattern can typically last at least a minute. You at least infer that you have set the PTZ to cycle through a number of presets. If so, that would explain why you are having trouble with speed control. PTZs typically go from preset to preset at a fixed, high, speed. That is a function embedded in the firmware.

Why don't you just ask the installer?

Stylis, Covering tables can be both more and less difficult than you might think. The cameras' distance from the table doesn't really have a bearing on whether you can identify the cards and chips or not. Simply use either suitable varifocal lenses or cameras having power zoom/focus lenses and zoom in to obtain a suitable field of view. For varifocal lenses, we typically use the 3.1-8mm lenses that come standard with the IndigoVision Ultra 1k cameras for areas with lower ceilings (under 14 feet high) and Evetar 8-16mm lenses in higher ceilings. We also occasionally use Evetar 12-40mm varifocal lenses where we have to zoom in on something more distant. If the camera has a power zoom/focus lens, 3-9mm works for most applications. The key with table games is usually lighting. The intensity, beamwidth and location of the lights is critical for good camera coverage. Tables should be evenly lit with no "hot spots", which are typically caused by lighting tables with "spot" rather than "flood" lights. The camera and lighting angles are also critical. We typically place the camera approximately over to slightly behind the player opposite the dealer. To prevent glare from the lights reflecting off the cards and back into the camera, we also place the lights on the same side of the table. That makes reflections bounce away from, not toward, the camera. There are a number of theories regarding table game camera placement. We like to locate the cameras above the player side of the table. That allows us to see the actions of the dealer and the chip rack. Other casinos like to place the camera behind the dealer so that they can watch the players or to the side of the table. One key is to not place the camera so far away from the table horizontally that there is a big difference in the distance between the camera and various points on the table. If the camera's angle is too far from perpendicular to the table's surface, you'll find it difficult to get the entire table in focus. Steeper angles make it so when one part of the table is in focus, other part(s) are not. Finally, no matter what type of cameras you get, be aware that with many casino applications, AGC becomes a critical part of the equation. With many cameras, the high contrast found within casinos causes the cameras' AGC to misbehave. The result is a "muddying" of the video, which has practically the same effect as an out-of-focus lens. We often turn AGC off or at least lower the level of the AGC if the camera has that capability. We also often use manual iris lenses and adjust the iris and shutter for best picture. Hope that helps.

Stylis, It's honestly difficult to get much info on casino CCTV systems. General rules about cameras and recording systems apply but casinos often have their own approach and we tend not to tell each other very much. I also don't know what is available in Africa so even steering you to a system manufacturer may not be any help, especially if they don't do business there. We use IndigoVision for our recording system and many of our HD cameras. In the U.S., they are one of the companies that specializes in the casino vertical. Other companies involved in the casino industry are Dallmeier, NICE Systems (although they just sold their entire video section to a private company named Qognify, and to a lesser extent, Pelco, Genetec, Milestone Systems and DVTel. All of those systems are aimed at large-scale deployments like ours so I'm not sure if the companies can be of any help but they may be a good place to start. You can also join IPVM. They are a good source of general knowledge but not cheap to join. I myself let my subscription lapse due to its cost. And, of course, you can ask questions here, although there aren't many on cctvforum.com with casino-specific knowledge.

Stylis, It depends an awful lot on what regulations and regulatory authority the casino falls under. Each area has specific rules regarding what must be covered by cameras and often specifies things like frame rates, retention time of cameras, backup power, Surveillance Room staffing, etc. For instance, many regulatory authorities require "real time" recording of gaming tables, individual slot machines or banks of machines, count rooms, cash cages, etc. "Real time" is also sometimes subject to interpretation. Some authorities will accept 15 frames per second while others require 30 frames per second. The NIGC MICS (National Indian Gaming Commission Minimum Internal Control Standards), published by the Bureau of Indian Affairs of the U.S. federal government, requires 20fps, largely because the regulations were enacted during the days of VCRs and were aimed at so-called Time Lapse VCRs, that typically recorded at lower frame rates to extend the amount of time a tape could hold. Many casinos separate the frame rates for some cameras from others. To save storage, some casinos record non-critical cameras at lower frame rates while recording "critical" cameras at "real time" frame rates (15-30fps). We chose to record every camera at 30fps, in part because our regulations were somewhat vague about what "real time" is and where it is required. That's a major reason why we have 700TB of primary storage when if we had liberally interpreted the regulations, we could have gotten away with half of that, or less. In other areas of coverage, most places require camera coverage of all gaming areas and all cash handling areas. Resolutions typically require things like the ability to determine the value and suit of playing cards on the table and the value of chips in a stack. That typically requires at least 720p resolution. The same may be true of cash handling, where some authorities require the ability to determine the value of currency laid on a counter. Other regulations may include requiring two cameras "criss-crossed" for each craps or roulette table. So in your casino, you would need approximately 26 cameras for the gaming tables, with perhaps additional cameras for pit overviews, slot area overviews, point of sales (cash registers) if any, count room/vault, cash cages (typically one camera looking straight down at the counter for each station plus a room overview / "211" (holdup) camera. Add cameras covering the face and jackpot sign of any "wide area progressive" slot machines (we typically cover the face of, and jackpot sign of, all slot machines that have a "reset jackpot" of $100,000 or more) Reset Jackpot is the beginning jackpot value when a machine is first placed in service or the amount the jackpot "resets" back to after someone wins it. In actuality, we cover every square inch of the gaming floor and most of the "back of house", except for private offices. Our casino has 2,000 slot machines, 40-something regular gaming tables and 20 poker tables, in addition to multiple cash cages, a count room and vault, ATMs and Ticket Redemption Kiosks (where slot players turn their cashout tickets back into cash) and numerous other areas we cover. We also cover parking lots and parking structures, employee hallways, entrances and exits and so on. Our camera count is over 1,000. We still use many analog cameras, although we are slowly switching many of our cameras to HD IP. The mix is approximately 20% HD IP and 80% analog but that changes weekly as we add IP and remove analog cameras. We also have over 150 PTZ cameras, both inside and outside the casino buildings. As far as recording, we record every camera at 30fps using an Enterprise-level NVR/VMS system with multiple servers and over 700TB of RAID6 storage with multiple levels of redundancy for everything, including "auto-failover" backup servers, UPS and generator power backup, redundant HVAC server room cooling and redundant networking. Our camera retention time varies from two weeks to 60 days, depending on its function. We also employ RAID to store evidence clips and have retained all that critical data for over 10 years. That takes up around 10TB of storage.

No you're not. Belden's catalog is not well-oriented to simple cable choice because they make so many cables that are only slightly different but still maintain the RG designation. So if you're so smart, which Belden RG6 cable is the best choice for delivering baseband analog video indoors 1,200 feet? Would you use 1694/1695 or 9290 or 533945 or 633948 or what, and why?

Seriously? They are not RG59 by any stretch of the imagination. A cable's "RG" designation is a function of the impedance, the diameter of the center conductor and the spacing between inner and outer wires, not the actual materials used in its manufacture. The RG designations are a vestige of an old military numbering system dating from World War 2. "RG" designates the term "Radio Guide" and the "U" signifies "Uniform" or "Universal". They are mostly used to identify compatible connectors that fit the inner conductor, dielectric, and jacket dimensions of the old RG-series cables. Still, calling CCTV-compatible RG6 "long range RG59" is like calling a Boeing 777 a "long range DC-3". RG59, no matter its construction materials and use, has a center conductor ranging from 23 gauge to 20 gauge, depending on the manufacturer. RG6 has an 18 gauge center conductor and RG11 has a 14 gauge center conductor. The major difference between low frequency cables and high frequency cables is the construction of both the center conductor and the shield. LF cables typically have pure copper center conductors and shield while HF cables typically use copper-clad steel or aluminum or some combination of the two. HF cables also commonly use a foil shield and an integral drain wire. The main reasons for the differences in component materials and type of shield are the cost to produce (and sell) and the properties of the signals the cable is meant to carry. Steel and aluminum are substantially cheaper than copper and even adding the cladding and/or tinning process adds less to component costs than using pure copper. In fact, there are some types of hard cables that have hollow center conductors for that very reason - to save the unnecessary cost of the "missing" copper. The second reason for using clad wire vs. solid copper has to to with DC and AC resistance and the "skin effect". LF signals travel totally within the diameter of a wire, so if the inner part is made of steel or aluminum, which both have far higher resistance per foot than copper (or hollow), the signal would suffer greater attenuation and would deteriorate relatively quickly. HF signals ride on the outside of a conductor, something called the "skin effect". Because of that, it is not necessary to use solid copper conductors for CATV because the inner copper part of each wire (including the individual strands of the shield) is unused. In fact, the hollow center conductor of certain "hard cable" types for HF signals is actually a plus. Due to the skin effect, the signal only travels on or near the outer surface and the higher the frequency, the shallower into the cable cross section the signal travels. Hollow wires have both the inner and outer surfaces to ride on and the inner surface adds surface area, lowering the attenuation. You shouldn't use CATV or RF-rated non-pure-copper-conductor cables for LF signals for that specific reason. It can be acceptable for short distances but not longer. Copper/copper RG59, for instance, can transport video signals up to 1,000 feet but I wouldn't use clad cable for distances longer than maybe 200 feet, if that. The specs of the two types of construction say it all. For instance, West Penn 841 and 6100 are almost exactly the same except for the construction of the center conductor. 841 has a solid bare copper inner conductor, which yields a DC resistance of 6.5 ohms per 1,000 feet while 6100, which has exactly the same specs but with a copper-clad steel inner conductor, has a DC resistance of 28 ohms per 1,000 feet. If a 6.5 ohm/k resistance yields a maximum distance rating of 1,000 feet, a 28 ohm DC resistance would yield a maximum distance rating of around 230 feet. Finally, I use West Penn's catalog as a reference because it is simple, yet comprehensive. Belden and other wire manufacturers agree with the gist of what W/P says, only their catalogs are not laid out as simply. Belden, in particular, manufactures so many variations of cables and their documentation leaves so much to be desired that it is almost impossible to pick a cable without spending a huge amount of time digging through their catalog and translating their specs. I don't know why you equate West Penn's specs with a speaker manufacturer's over-inflated power rating. Do you also compare the horsepower of cars to the flavor of beers?

Yes, gauge of the center conductor and, of course, diameters. But center conductor size has a big effect on DC resistance, which in turn affects the ability to transport both DC and low frequency signals over distance. - RG59, with a 20 gauge center conductor, has a DC resistance of 10.1 ohms per 1,000 feet and attenuation of 0.68db per 100 feet at 10MHz. - RG6, with an 18 gauge center conductor has a DC resistance of 6.5 ohms per 1,000 feet and attenuation of 0.52db per 100 feet at 10MHz. - RG11, with a 14 gauge center conductor has a DC resistance of 2.6 ohms per 1,000 feet and attenuation of 0.35db per 100 feet at 10MHz. That demonstrates why RG6 will transport video farther than RG59 and why RG11 will transport video farther than RG6. Again, all other values being equal (95% copper shield and solid bare copper center conductor). This has nothing to do with a cable's ability to reject EMI, since all three cable types have essentially the same construction and therefor the same EMI rejection capability. In fact, other than signal attenuation, the big differences between types is cost. RG6 costs more than RG59 and RG11 is substantially more. Part of that is the cost of the copper per foot - RG11 has more copper than RG6 which has more copper than RG59. By the way, the same would apply to so-called "mini-coax". We often use W/P 825 MiniMax for very short runs and it works perfectly well up to maybe 100 feet. We use it for short patches, for instance from our patch panels to our encoders. The big advantage is that its smaller diameter allows us to fit a lot more cables in the same conduit or raceway.

Missed that. Still, you did say That was totally wrong. Try the link to West Penn's catalog. W/P 806, 2806B, AQC806, AQC2806, 25806 and 4806 are RG6 cables designed specifically for baseband video. And they're not the only cable manufacturer who makes baseband video compliant RG6. http://www.scpcat5e.com/bulk-cables-c-93/coaxial-cables-c-93_105/cctv-coax-c-93_105_179/rg6-cctv-coax-c-93_105_179_221/ https://secure.libertycable.com/products/Liberty-Wire-and-Cable/RG6-CCTV/Baseband-video-RG6-non-plenum-coaxial-cable http://www.primuscable.com/store/c/719-RG6-CCTV-Coaxial-Cable.aspx

Funny, that's not what I read. You said: Nowhere did you say there are suitable versions of RG6 and RG11. In fact, your statements that RG6 and RG11 are not designed for CCTV is dead wrong! CATV RG6 and RG11 (and RG59, for that matter) are not suitable for CCTV but copper/copper RG6 and RG11 are perfectly usable. I even presented suitable West Penn model numbers. Did you try looking them up? http://www.westpenn-wpw.com/webcatalog/BULK%20CABLES/CCTV-SECURITY/Analog_CCTV.pdf

I believe you meant RG11, not RJ11. Also, it's not necessarily true that RG6 is not suitable for CCTV. With any type of coaxial cable, the capability to transport video signals is determined as much by cable construction as it is by cable size/type. For best CCTV results with coax cable, it should have 100% pure copper center conductor, not copper-clad steel or aluminum. The cable should also have a shield made of woven pure copper, covering at least 95% of the inner core. That cable is typically referred to as copper/copper or bare copper. Cables with the designation "CCA" (copper-clad aluminum) or "CCS" (copper-clad steel) have limited usefulness for CCTV. The reason being that an analog CCTV signal needs pure copper for best performance / longest distance. Coaxial cables for CCTV can have additional shielding, but the primary shield must be pure copper and cover at least 95%. You can obtain RG59, RG6 and RG11 with either pure copper or copper-clad construction. Clad cable is primarily used to transport HF (high frequency) signals like Broadcast or Cable TV, radio and other RF signals. Pure copper cable is used to transport low frequency signals like baseband video, audio and some types of data. If you choose bare copper, the distance capabilities without mid-span amplification and/or signal correction are: RG59 - up to 1,000 feet / 300M - 20 gauge center conductor (Example: West Penn 825) RG6 - up to 1,600 feet / 500M - 18 gauge center conductor (Example: West Penn 806) RG11 - up to 2,000 feet / 600M - 14 gauge center conductor (Example: West Penn 811) The above table is for copper/copper cable. Distance capability is drastically reduced for other types. As with any type of cabling, the correct termination (connectors) is required for best results and good practice must be followed when installing the cable (no cuts or kinks, installation following proper "bend radius" requirements, etc.).

What, you don't like purple? My wife would disown you.

Deleted

I would do the calculations slightly differently. Assuming you use 8Mbps (and you will have to decide that by trial and error to select the minimum bit rate that provides acceptable images), 8x60x60x24=691200/8=86400/1024=84.375GB per day. Remember a gigabit is 1024 bits and a terabit is 1024 gigabits (the same applies to bytes). CBR data requirements are the easiest to calculate while VBR can be very complex. The actual disk space used can vary quite a bit depending on the camera design, codec used and amount of motion in the video (including noise in low light situations). There are also differences in how manufacturers treat VBR. Some use a hard cap at the specified setting so the bit rate will never exceed what you specify but will float down when little is happening, whereas others let the bit rate float both below and above any setting. With the latter, bit rates can actually run incredibly high during motion or when the video is noisy. Remember when using RAID6, two drives worth of data are parity drives and they can't be included in storage calculations. Example - a 12-bay chassis running RAID6 with 3TB drives would yield 36TB raw but the net would be around 30TB (slightly less with Windows overhead). Another thing to keep in mind is drive error rates. Typical specs call for one unrecoverable error in 10 to the 15th bits (1petabit). I tend to think that is way over-optimistic. Also, as RAID Group size gets larger, so do rebuild times. In fact, a 12-disk RAID6 group takes at least 48 hours to rebuild with most storage systems. During that time, the system is running in "degraded mode", which limits data throughput and also increases the system's vulnerability to faults. For those reasons, I refuse to utilize RAID groups larger than 12 drives and with larger than 3TB disks, would actually seriously consider smaller (<=8-disk) groups. Of course, the problem there is the point of diminishing returns. A 12-disk group of 3TB disks would have a bit under 30TB of usable space with 6TB, or 1/6 of the total, dedicated to parity. An 8-disk group of 5TB disks would also have 30TB of usable space with 1/4 of the total dedicated to parity. The larger the drives, the worse it gets. 8TB disks in a 6-disk group yields 32TB usable but sacrifices 1/3 of the drives to parity, Etc., etc... You have to decide how long rebuild times you can accept. If a 36TB raw (30TB net) group takes 48 hours to rebuild after a HDD failure, a 72TB raw group would take over 4 days. And you must assume disks WILL fail. Not if, but when...

ELECTRICAL Input Voltage 18-30 VAC; 24 VAC nominal Input Power 25 VA nominal (without heater);70 VA nominal (with heater) You definitely need to use AC.