dustmop

I'd believe it. I'm in an environment where there are well over 360 (enterprise-grade) drives spinning all the time. Seagate drives seem to have bad models, and it seems to hurt. My failure rate for 500gb drives is quite low: maybe around 3% over 6 years. 1tb Seagate drives on the other hand: closer to 60% over 5 years. I've had good experience with 3tb Hitachi drives, with no failures yet over ~13 months. 2tb Seagate are also treating me well.

That being said, I'm not surprised about the 1.5tb Seagate drives being crap. But their newer drives seem to have overcome MOST of their issues from before.

I've learned that you can do all kinds of things to help prevent damage, but in the event of a direct hit, the camera will most likely fry. I believe that the best thing you can do is protect the head-end of your system from lightning chasing back in.

I have issues where the cameras don't get hit, but lightning manages to sneak into the cabling and chase both ways. After having 2 cameras and one entire cable run get toasted from this, I started using the MMS-PTZ-UTP (protects UTP video, UTP data, and power) from http://www.minutemanups.com/products/lineguard.php for the few outdoor PTZs that are on copper. There are comparable products out there from other manufacturers, too.

Professional.

 

Our security group needed help on the engineering portion of a $8 million dollar heavily integrated security system. I was brought in and spent a year learning/designing special systems portions of the project and along the way picked up quit a bit of CCTV knowledge. This was all right before IP really took off, so it was ~600 or so analog cameras, a lot of fiber equipment, and racks upon racks of N.I.C.E. Systems dvr's.

 

So it all kind of just fell into my lap and I've been stuck with it ever since.

Honeywell front end by any chance?

Another great installation at a local school in North Texas. Someone must have found the box of seal-tite and decided that was "good enough"!

 

 

I bet if you looked on the roof of that building, you can put a ladder on there and still not reach the top of the brick.

More than likely, the handy box is the highest point of entry into the building and thats why that installer

did what he did. It may look horrible, but I have a feeling there was a reason.

I've encountered that more than a few times. But I use rigid conduit then. I only like seal-tite for the < 1 foot from the box at the top of the conduit to the camera. It's also cheaper to use conduit than that much seal-tite!

I believe in both uses, but with a heavy emphasis on deterrent. Sometimes unobtrusive/small/covert cameras are more appropriate, but if you have had vandalism/theft/etc issues in the past, you probably want to make it nice and obvious, while ensuring it is the correct camera/mount for the job (vandal resistant) and is mounted where it's difficult to get to from ground level.

I have found that rectangular/bullet style enclosures on arms outside draw a LOT of attention, where a smaller vandal dome is less noticeable. Arm mounted enclosures can be moved or ripped off by a determined individual pretty easily, where a dome isn't so easily abused. (The previous is just my two cents, your mileage may vary, etc etc)

There's cameras, and then there's cameras. It depends on the weight. We use simple drywall anchors for light plastic mini-dome cameras and have never had a problem except for the rare time a water leak causes the tile to turn to mush. For heavier cameras, we still use either toggle anchors or machine screws with fender washers. Be careful with plywood. In some cases, you will have to use fire retardant plywood or risk the wrath of building inspectors.

 

For heavier cameras like vandal domes and PTZs, a safety chain is highly recommended.

Actually, I do use fire rated plywood for my strips. We always have sheets on hand, and when we have to cut one down for something, I cut the scrap into strips for camera supporting.

Also, there's nothing like having to install a DF8 dome just for a camera that is like 1/8" too long. Or the stupid rules here where the regulators really want cameras in a dome (not just exposed), so you have to stick one of those super-heavy-bludgeon-your-enemy Axis IP fixed cameras in there.

All great ideas above me here.

I use bolts with wingnuts and toggle bolts for really light cameras/mics/hornstrobes. For larger/heavier cameras, I usually use strips of wood cut to the size of the tile (2 feet x 2 inches), and then either let the camera housing mounts catch those strips of wood (Pelco type cans), or use regular screws and just screw into the wood. I usually cut them about 1/8" shorter than a ceiling tile in width, to allow for easy removal of the tile down the road.

You will find BNCs even easier to learn, but remember before crimping the center pin make sure it is properly seated to the dialectric (polythene inner).Don't forget to put the relief boot and ferrule on before trimming the cable ( we have all done that at least once I think? ).

Once? I think I've done it at least 100 times in the last 5 years. I even still make the "ah **** I forgot the strain relief boot, I'm a moron" mistake when I do fiber optic terminations from time to time.

It was in Florida, where the law says you do NOT have to back down/flee from your attacker. Also, the part where he "chased" them is very misleading; the entire engagement occurred within the building. He continued firing upon the perpetrators until they fled the scene. According to Florida law (and many other US states), he was within his right to engage with deadly force if he felt the perpetrators were a threat to those around them.

Frankly, I wish he would have had much more training with his personal carry weapon, so that he would have only needed to fire twice; once each. So in regard to the fact that it was pretty pisspoor weapon discipline, I agree. And had he hit a bystander, he would have been charged, and rightfully so. People need to understand that just because they CAN carry a weapon, you still need to have proper training and judgement before whipping it out and "busting some caps".

Also, carrying subsonic hollow-points, and being well trained and very comfortable with your weapon helps prevent collateral damage. I'm a firm believer that EVERYONE should be trained in the appropriate, safe, and effective use of firearms, and those that wish to carry, should be allowed to after passing a set standard of training (except felons and those otherwise excluded, and also adhering to local/state laws for licensing, of course). But then again, maybe that's just the Marine in me speaking up.

I've actually been asked a few times by people if I had any recommendations for how to do this, but I never have an answer beyond the various mobile dvrs and a dash cam.

Two people in particular want cameras in their side mirrors looking back at the doors to the vehicle. Anybody know of a company that provides such a product?

I hope the backlash they receive from this teaches them something. And they better not screw with the commercial/enterprise grade hardware! For SO/HO hardware, DD-WRT or bust!

Yeah dustmop, we're just starting the process of evaluating systems to replace our aging Honeywell Enterprise purchased in 2003. One option we have been assessing is retaining the Pelco 9780 for Spectra PTZ control and monitoring while using a virtual matrix for the fixed cameras.

 

Do you have totally independent monitor/control systems for the two or what? I envision interfacing between the Pelco CC1's or joysticks and the VMS so that if we call say, Camera #501 (fixed) to Monitor 12, it only appears there while if you call Camera #120 (PTZ) to Monitor 12, it appears on Monitor 12 but also appears on a single analog "PTZ Monitor" on the console in front of the operator for zero latency control purposes. Since an operator can't control more than one PTZ at a time, that would work just fine.

 

I believe someone (manufacturer or integrator) should be able to write an app for that.

For when I do have IP cameras in the system for testing purposes, they only show up in our VMS. Going forward we have looked at single monitor stream decoders or just going to several LARGE format monitors per station and doing monitor walls. In either scenario, we would leave probably 1 or 2 analog monitors per station simply for PTZ control of the existing analog PTZs. This would let us draw our 9780 down from the monster it is to probably just a few bays, along with taking out most of the UTP hubs for the twisted pair. We don't envision adding IP PTZ cameras to the gaming floor or money areas in great numbers for a long time yet, and even if we do, we're looking at possibly just adding some 360 degree cameras to get better overall coverage. IP fixed cameras will definitely be added soon, however, as latency is really a non-issue on a fixed camera, and the benefits (tables and money areas) are fantastic.

A virtual matrix for fixed cameras also seems like a decent compromise for a transition phase, but I'd have to see some pricing before making a decision to go that way or just use the existing analog matrix until the cameras are replaced with IP.

We also looked at some of the IP PTZs that also offer analog outputs, but the amount of cabling required for that makes me cringe, and not many cameras offer both.

I know you are at least familiar with the system I have here, but we don't actually use Pelco controllers; we currently utilize a USB controller that plugs into the workstations. The VMS software uses a Moxa ethernet-to-serial interface box (such an awesome piece of hardware), where one Moxa port is assigned a COM port for each workstation. Then you just wire it into the back of the CC1 like you would for a normal Pelco controller. This results in a latency on the order of maybe 30ms while it passes the data from workstation to server to Moxa, to Pelco; in other words, you can't tell at all.

Also, in our VMS here, we have a feature called "follow me", where you can make what your doing on one workstation automatically show up on a specific monitor. Normally used in situations where you want to show security or the gaming authority something, without having to tell them what camera to pull up. You just do it and it shows up to them automatically. You could set it up to where it would just control your analog monitor locally and put up the PTZ you want to control. So I'm sure something could easily be written if Honeywell or whomever doesn't already offer something like this. I'm lucky in that my manufacturer takes my input and requests rather seriously.

EDIT: We don't actually use the "Follow Me" feature here, so I just set it up and tried it to see if it would mirror changes on the digital review/spot/VMS/whateveryouwantcallit monitor to an analog monitor, and it sure does! Looks like at least I have a solution for us for the future.

I can see the biggest benefit for you being in PTZ control, especially as this is one of the main areas where latency in IP cameras IS a problem. On that note though, I'm currently bench-testing an Axis Q1604 (their new highly-touted low-light/WDR toy), and I gotta say... latency might be MEASURABLE, but it's practically UNNOTICEABLE unless you're actually looking for it. I can point it at my screen, then watch on the DVR - moving a window on the screen, it's only intermittently that I can actually see a delay on the DVR.

I've played with some Axis cameras in the past, most recently the Q1755, and I love them. While I wish the Q1604 offered 1080 instead of 720, I think it's time to find one to play with. I can think of a few places here that a camera like this would help.

We can't use either one so I've never even tested them. I've had a couple of vendors approach me (one for HD-SDI and one for HDcctv), but my first question to any vendor is can they demonstrate the product on our system.

 

For our purposes, neither system is ready for casino use: it won't work with our cable infrastructure (70% UTP single-pair w/baluns - 30% coax with many runs far longer than 100m), it won't work with either our matrix or our NVR and it won't display on our monitors.

 

In essence, what's the point?

Actually in my line of work, zero latency would be a plus. However, as I stated above, there's no advantage to a technology that not only requires we replace much, if not all, of our infrastructure, is not capable of seamless integration with the rest of our system. No matrix switches or recorders capable of handling >1000 cameras. No ability to display on tens of monitors, etc.

I'm in the same boat as you, except that EVERY camera here is done over UTP (parking lot cameras are on fiber). The thought of a PTZ that isn't easily compatible with our switching matrix/every monitor AND has noticeable latency makes me shudder. For fixed cameras, latency is a non-issue, so going IP and with whatever resolution is required makes great sense (think face cameras, gaming tables, number plate recognition). So with a rather LARGE system that is compatible with both analog encoders and IP cameras (manufacturer will add support for unsupported cameras on request, and quickly), the transition to IP is much easier to accomplish than to HDcctv.

Add decoders to most of the existing monitors so they can decode IP streams, then leave a handful of monitors on the old analog matrix for use with the analog PTZs to eliminate lag. You can still display all analog cameras on the digital streaming/decoded monitors, but they will have a noticeable latency, hence leaving a few analog monitors. IP PTZ lag is getting to the point where we could tolerate it in our line of work thanks to multiple video streams.

As more and more IP cameras are added and old analog replaced, eventually the analog matrix and the analog encoders would simply dwindle away and disappear, freeing up rack space for more DVR/NVR units to record the IP cameras with. And with 7 racks full of analog switching and encoders, that's a WHOLE LOT of recording.

Also, the system I use is only limited by network bandwidth and disk i/o for the number of IP cameras you can record. What was that limit on HDcctv DVRs again?

Hello,

This cable works with Spectra IV all the time.

i am unable to operate Spectra III with this cable.

I am not sure if it is due to some termination issue??

Any Idea??

I don't have any Spectra III here to try it on, so I'm afraid I can't be of much help with this. Maybe someone else here can be of help (I'm looking at you survtech)

Null modem cables flip over pins 2 and 3 internally, so this means you just need to invert those two pins when you make the adapter/solder your end.

Remember, for end-to-end, you want the pins to do this:

Serial Pin (PC)    RJ45 Pin (camera end)
2                  6 (Green in T568B)
3                  8 (Brown in T568B)
5                  2 (Orange in T568B)

If you are soldering the wires to a male DE9, then just using an 8P8C/RJ45 on the other end of the CAT cable, make sure you terminate the CAT as a T568B (normal patch cables almost always are sold/made that way). Then you want to solder the following 3 SOLID COLORED wires (not the white of the pair) on the "loose" end of the CAT cable to the pins on the DE9:

Because you are using a null modem cable, pins 2 and 3 are flipped when making this connection. The end-to-end result will still be the same.

Color Wire        DB9 Pin
Brown             2
Green             3
Orange            5

If by any chance this cable does NOT work, flip pins 2 and 3, and I bet it does (also, let me know if you have to flip the pins, and I will amend this post for future searchers).

I just made it into the office about an hour ago, sorry about that!

I did mine with a null modem cable and THEN a serial to RJ-45, so I had to modify it slightly to get it working without the null modem cable (took about a minute).

For end-to-end, you want the pins to do this:

Serial Pin (PC)    RJ45 Pin (camera end)
2                  6
3                  8
5                  2

Using a regular patch cable and a FEMALE serial (DE9) to RJ45 adapter, and assuming your serial to RJ45 has the same color wires (I've never seen one that didn't use the same colors), assemble it like so:

Socket Hole     Color Wire
Pin 2           White wire
Pin 3           Yellow wire
Pin 5           Orange wire

EDIT: Forgot to add that I use a real serial port, and not a USB-to-serial adapter, so I can not vouch for the functionality of a setup like that.

Also, in the Spectra download utility for windows, be sure to pick the correct protocol, the address of the camera (D or P), and the COM port you are plugged into. If you have issues, you may need to adjust your port settings (in device manager, select com port you want to edit) to match the camera settings. I have never needed to do this; Windows is set at 9600,8,N,1, and I just tested a camera with default settings on SW2 (that's 2400).

I just made one recently at work, actually, with a serial to RJ-45 adapter. Too bad i won't be back there until late Monday (EST). Once I get into the office, I can check the pin-outs for you. Unless someone beats me to it, that is!

Assuming you want to only control the lights by using the camera output, my thought is this:

If you know you can trigger the output on the camera in a way that you can control (webpage, DVR, etc), then all you need is a relay, like a 12VDC < 100mA coil relay (the contact rating needs to be greater than whatever your lights are, i.e. 110/277/whatever at whatever amp load you need)

It's an open collector output, and an open collector essentially acts a ground (by means of an NPN transistor). So you need your positive voltage (+12VDC) going to one leg of the relay, then the other leg going to pin 1 of the DB-15. When you trigger the output, it lets the voltage find a ground, and the relay will change states. If they are PoE cameras, and you have no local power, that rules out 12V relays, as the camera only has a 5V output. The problem here is that 5V relays are usually meant for PCB type installations, and not your classic (and super easy) socket type. Doesn't mean you can't make it work, however. But if you have local power, just find yourself a 12VDC power supply.

It looks like it can sink 24V on the open collector, but the issue to watch out for is difficulty finding relays with coils that can switch 24V <50mA, and still have large enough contacts for high voltage.

If you use double throw relays, they will have 2 distinct states, whereas a single throw is just on and off, like a normal light switch. I say this, because you could use DT style to make so if the camera were to up and die, then the lights could be forced into an on state by using the non-powered circuit of the contacts.

Then you just run the HOT leg of the mains power going to the light through the contact of the relay, and you should have a relay controlled light!

I actually do stuff like this quite a bit in the door access world, and have 2 relays on my desk that fits the bill for everything I do. I use DIN rail sockets to make installs look nice where applicable, or they screwed into the panels.

Omron MY4-12VDC 4PDT - 12vdc 75mA coil, contacts will do 3A @ 250VAC or 3A @ 30VDC

Allen-Bradley 700-HA32Z12 - 12vdc 100mA coil, contacts will do 30A @ 120VAC, 15A @ 240VAC, or 10A @ 30VDC

Typically, they will short out if their rated voltage is substantially exceeded. That could happen with a nearby lightning strike.

I've had this EXACT failure on 4 boards over the past 4 years, and all were related to a lightning strike in some way. It's amazing what lightning can do, even in a setting where everything is powered from massive building UPS units.

Personal experience from attempting to repair one of these boards and having our in-house bench technician attempt to repair two, is that it's never just that cap. There's going to be something else wrong with the board, somewhere.

Sometimes you can find the boards or back boxes on ebay for fairly cheap.

That's a Spectra III back box, so you should have no issues with either. Flipping through a manual here for one of the Spectra IV cameras, it says that Spectra III and IV dome drives will fit into either Spectra III or IV back boxes.

Page 6 of this pdf http://195.70.43.12/Pelco/c3417me.pdf.

A few times I've been forced out into the winter weather to replace cameras (regulatory requirements and all).

When it's about 20F, with 20mph winds, no amount of clothing helps when you're 15 or 20 feet up in the air!

Thankfully I have remedied the three offending cameras, and all is quite well now.

Otherwise, I don't do ANYTHING outside when the temp is below 40F.

I'm filing all this information away for when I have to do work over in the giant metal barns. We have some 69kv lines running right through the property. Thanks for a follow up post!

In a dream world, I'd LOVE to have a 360 degree recording camera that can do this. I'd like SEVERAL of them, actually. Sadly, that day is likely still 5 to 10 years away.

According to this http://support.lytro.com/entries/20718762-what-file-formats-does-the-lytro-support:

The light field data, including 3D info about the color, luminosity and direction of 11 million rays, is stored in a format developed by Lytro called the LFP (Light Field Picture) format. When you move your "living pictures" to your computer via the Lytro Desktop software, each LFP file is about 16MB.

They don't list a true resolution; they say it's "11 megarays", whatever that means. (You can currently export images as JPEG, at 1,080px X 1,080px) They say it's 16MB for a single image. Given the sheer amount of data they do store in each image, using I/P frame differences like h.264 does is either highly unlikely, or insanely processor intensive.

Estimated sizes if we assume NO compression:

1 image: 16MB

1 second: 480MB

1 minute: 28.8GB

1 hour: 1.65TB

1 day: 39.6 TB

If they can swing something akin to h.264, then the numbers would obviously change to something a bit more reasonable.

Just checked my recording, and an entire day of analog h.264 (1.5Mbps) is between 15GB and 16GB.

The biggest hurdle will be price. Digging around for some Raytrix camera pricing (the people that really started the whole lightfield thing), they sell their R5 camera for just under €3000. That camera can do 30fps over GigE, or 90fps over USB3. Aperture is lousy, so lighting matters. But the max 2d image size? 1 megapixel (source: http://www.raytrix.de/tl_files/downloads/R5.pdf). Also, the Raytrix cameras require CUDA GPUs to perform a lot of their processing it seems.

My immediate thought is that wherever you would put it, you're going to need some great lighting. Face shots into places with large lines/queues (subways, shopping malls, casinos, amusement parks) would be neat. Not sure where else it would be truly useful, given that lighting really does matter to this type of device.

Also, I bet the recorded stream would be ENORMOUS compared to a typical h.264 stream of a similar resolution. Not to mention the processing power to refocus video in playback as opposed to refocusing just a single frame.