SBTVideoman

Some other things to consider 1) Bear in mind we're talking about analog cameras here...in other words, the analog signal on the back of the camera. 2) Virtually all cameras can only produce a signal that's higher than 500-540TVL when using a S-VHS output instead of their composite (BNC) output - try this for yourself....with any camera that has a SVHS output...every single one that I've tried (this includes Bosch Dinion, GE UltraView, Siemens CCBX series, Kampro EVR/HDR series) has a noticeably sharper image on the SVHS output than on the BNC/composite output. 3) Length of cable run...if you run 15-20 metres or more of coax from the back of the camera to the DVR/monitor, degradation of the analog signal means you will struggle to tell the difference in absolute resolution between a 480TVL and 540TVL camera (assuming we're talking about equal quality cameras with equal quality lenses). Analog signals degrade with distance and that's just a fact. Finally - consider WHAT happens to the camera signal once it gets to its destination. 1) Is it being recorded on a DVR? If so, at what compression level and resolution? If you record at 2CIF (like most DVR's do), then the playback image quality will be lower than the live signal. I can send you images captured off a 2CIF MPEG4 DVR at max quality that show a 330TVL camera is just as good as a 540TVL one...once you digitise it and compress it, the DVR becomes the limiting factor. 2) What's the resolution of the monitor you're using? Unless the monitor has a SVHS input and can display at least 500 TVL of resolutionl, you'll struggle to see any appreciable difference in resolution. Try this: Get a DVD player (any brand) and grab any recent blockbuster movie, and connect it to your TV with the composite signal, then connect it using the SVHS signal. If you cannot see an appreciable difference in signal quality I'll be surprised. Moral of this rather long discourse? Who cares what the theoretical resoluiton of the camera is....at the end of the day the real-life performance of the system depends on the rest of the signal chain.

Is the lens an "IR-corrected" lens (also known as a true day/night lens)? It sounds like when you focus the camera at night, then it's actually focussed on the IR wavelength. But during the daytime the lens is working primarily with visible wavelenghts. A non-IR-corrected lens cannot focus IR wavelenghts and visible wavelenghts to the same spot. I tried this once with a IR-sensitive B&W camera and two light sources - fluorescent (no IR component) and a dedicated LED-type IR illuminator. If I fitted a normal lens to the camera, I could have good focus EITHER with the fluoro visible light OR the IR illuminator but not both, no matter which one I used to focus the lens. If I then fitted the IR-corrected lens I had good focus on both light sources. Hope that helps.

My experience (having seen GE DVR's with Wavelet & WaveJet side by side) is that at max quality settings, the images were very similar in terms of resolution and quality. However, the WaveJet images had noticeably more artefacts at lower quality settings than the comparable Wavelet images. Note this was observed on a unit working in PAL, but I would expect similar results in NTSC (if anyone can confirm or contradict this, by all means do so). The streaming performance of WaveJet was also noticeably better than Wavelet, although it's still not in the same class as MPEG4 systems. Re image resolution - to the best of my knowledge all models of DVMRE's (whether CT, CTII, Pro, StoreSafe) always recorded one field (NOT a complete frame, which consists of two fields) , therefore they would be classed as 2CIF.

The DM DS2 can record all the way up to 4CIF resolution at 100PPS global framerate. And it can record in either MJPEG or MPEG4 and change on the fly, so (for example) you can record in MPEG4 in normal "time lapse" mode to conserve HDD space, but switch to MJPEG on alarm to get slightly higher image quality. While admittedly configuring a DM DS2 via the front panel is not the most intuitive, it has web browser access that makes config a lot easier and permits remote viewing. Also - DM comes (at least here in Australia) with a 3 year warranty including the hard drives.

Camera filters can be IR CUT or IR PASS depending on application. An IR CUT filter passes visible light but blocks all wavelenghts longer than visible red, meaning no infra-red light is passed. An IR PASS filter will typically pass near infrared light, but may block long-wavelength infrared. These typically pass wavelenghts around 830 - 940nm which is the wavelengths common IR illuminators work at. Now - to the application you discuss, licence plate recognition....and why an IR CUT filter may be useful. Any camera looking at the front of a car will typically have the cars headlights in the field of view. Most cars these days employ halogen headlights, which pump out lots of infrared in addition to the visible light. If you use a camera which is very sensitive to IR (eg an ExView or SuperHAD B&W camera, or an IR-pass colour camera or a Day/Night camera with mechanical filter), you may find that such a camera will flare out abnormally when confronted with car headlights, because it "sees" too much of the IR component. That's where screwing an IR CUT filter into the lens may help the issue, because now the camera will only be seeing visible light. Obviously a good camera should not have a problem....but having the option in the lens to screw on an additional filter is useful. Hope that makes sense.

Below is an extract from the PSU bulletin that they put out back in 2003": Technical Bulletin NU-60 Power Supply, used in the following digital recorders: • DSR Series (DSR2000e, DSR1000eZ) • DVMRe Family (CT, eZT, CS, StoreSafe) You may be experiencing symptoms that your NU-60 power supply unit is about to fail. Your NU-60 power supply may be going bad if you experience one or more of the following symptoms: • Recorder re-boots all by itself for no apparent reason • Recorder re-boots from nearby vibration – or from jarring of the PSU cable • LED on the PSU blinks on and off • Repeated email notification from unit of power up • Repeated WaveWatch notification of power up If you experience one or more of these symptoms, please make arrangements to have your NU-60 power supply replaced by our newer NU-70 power supplies. This replacement policy applies to any and all NU-60 power supplies you wish to replace, even if they have not yet begun to fail, as well as all spare NU-60 power supplies you wish to replace. All replacements will be done free of charge. To order your replacements, please make your request via fax at (541) 752-9097. The hard drive change-over from 5400rpm to 7200rpm took place (as I recall) late 2005.

Just to add more food for thought... Rory (and others) have already mentioned that sunlight contains lots of IR which will cause washed out colours with any camera that has an IR-pass filter - common on many "IR LED bullet cameras" but also a feature of some full-body 1/3" cameras, especially those coming out of SE Asia. NOTE - many of these will state something like "IR-extend" on their specifications or have spec simlar to "0.1 lux @ F1.2, 0 lux under IR" You can run into problems when using such camera inside under mixed ligthing.... Think about the two most common types of lights used inside in typical applications like nightclubs, hotels, motels, shops etc. 1) Fluoroescent tubes....these are a "cold" type of light in the sense that they produce only visible light. 2) Incandescent bulbs (either halogen downlights or the classic lightbulb invented by Mr. Edison) ...which are a "hot" type of light...that produces IR as well as visible...think about it....you're passing electric current through a metal filament...filament gets very hot and starts glowing....which is why incandescent lights are so damn inefficient, since most of the energy is turned into heat...some of it is smack bang in the 830 - 1500nm spectrum that IR sensitive cameras can see. OK..enough about the physics....what does this mean in practice..? I have seen video footage pulled off many a DVR that shows a person walking through a lobby in a hotel....and let's say they're wearing black clothes..and the camera used is an IR-sensitive colour camera (Ie NOT a true day/night unit with a mechanical IR cut filter)....and the person passes from under a fluoro light to under a halogen light...and hey presto you see their clothes turn from black to a green/turquise/aqua/teal colour....purely because as they walk under the halogen downlight, they are suddenly illuminated by both visible light and IR and the camera cannot cope with the mixed lighting. Moral of the story - if you want 100% accurate colours...don't use an IR-extend camera...you're better off with a true colour camera that has an IR-cut filter or a true day/night unit. TIP...with a full-body camera you can usually tell if the filter is an IR-cut or an IR-pass one. The true IR-cut filters tend to be a sea-green glass...with no sheen. The IR-pass variety tend to be an aqua colour with a strong pink dichroic sheen that's visible if you make the filter reflect some daylight. Hope that makes sense...happy to e-mail you some photos of the two filter types.

Just to add a little more to the discussion....TVL by itself is NOT a guarantee of image quality. The TVL resolution is measured by standardised condions, much like the "miles per gallon" test done on new cars. It is 100% correct to say that a GOOD QUALITY 500 or 600TVL camera will give better image quality under the same conditions as a 330TVL camera....however a good quality "branded" 330TVL camera with a good quality lens will easily provide better image quality than a SE Asian cheapie dome that claims to have 540TVL resolution just because it uses a Sony CCD. Other things to look at... 1) Signal to noise ratio (the higher the better) 2) Sensitivity 3) Lens quality As an example...I have a Siemens CCBC1320 full-body camera in my box of test cameras. This is a 1/3" 330TVL colour camera. WHen mated with a 5-50mm Fujinon lens (nice big glass optics) this camera routinely outperforms virtually every "480TVL" or "540TVL" SE Asian dome I've put it next to. And that's in good light...in low light the domes go pitch black while the 330TVL full-body still provides a usable image. Having said that....my Bosch LTC0495 with the same lens does provide a better image again... Key takeaway....compare apples with apples. I would personally always choose a lower TVL, higher quality camera over a "higher TVL" but overall poorer build quality if the price was the same.

Hard to give a definite answer on this one...but I seem to recall that when you enter the OSD menu of a CyberDome there is an option there for an auto-recalibrate period. Check if that's configured....it may explain why rebooting the cameras seems to solve the problem temporarily.

Give GE Tech support a call while out on site by the unit - if you give them the unit serial number and the time/date that the unit is displaying they can generate a "back door" 4-digit code that's only active for 1 hour....hence the need to be out on site. This will allow you to enter the unit menu and default it back to factory settings. In order to do this you will need to be able to see the unit serial number.

Being in Australia you have several choices for "global brand" DVRs. 1) Dedicated Micros DS2 series (100PPS global) - good image quality, dual encoding (MJPEG/MPEG4) with built in burner, well proven with a number of distributors (Siemens BT Products, DVR Solutions, Security Merchants - check out DM's Australian website) who can provide local support and service. And the warranty is 2 - 3 years depending on model. 2) Siemens SISTORE AX - relatively new to Australia, but well proven in Europe. Global frame rate is 200PPS and has built in burner. Encoding is MPEG4 so can provide good storage times. And Hard Drive expansion kits are available from the manufacturer/supplier (Siemens BT Products). Major difference to the DM DS2 is that the SISTORE has USB archiving as well, so you can copy clips to a standard USB2.0 flash drive - this is typically faster than burning a CD or DVD and you can now get 2GB or 4GB memory sticks pretty cheaply. Both of these are embedded units. Suggest you take a look at both, decide which one you like (user GUI's and remote access clients are different) more since they're both different. Both have motion detection, automatically decide between colour and B&W recording and are supplied with Network Access software. There are others as well, some better some worse.

First question - is the unit still under warranty? If so, I would suggest you contact the manfacturer. If not, then read on. What size were the original hard drives and was there one or two? If the original hard drives were 160GB (and there were two of them), you may be able to get away with a single 300GB or 320GB 7200rpm drive. The issue gets back to power consumption. If you compare the spec sheet of the original 5400rpm drives to the newer 7200rpm drives you'll find that the 7200rpm units draw signifcantly more power, particularly inrush current. Makes sense - it takes more power to start spinning the drive at 7200rpm than at 5400rpm. GE actually changed the DVMREs from 5400rpm to 7200rpm a couple of years back or so - there was something published on their website to that effect. At around the same time, they also changed the external PSU on the DVMRE. The original units were a "N60" 5A device...the replacement units were physically larger "N70". For a while the manufacturer was offering free swap over of the old PSU to the new one. The old PSU will almost certainly struggle with 7200rpm drives due to the inrush. The new one should be OK IF you put in one 7200rpm drive. If you need/want to put two 7200rpm, then you will almost certainly need to send the unit back to the manufacturer and request the installation of an additional "7200rpm hard drive kit" - this is an additional power supply board that's required for two 7200rpm drives. I am not aware if this part is available as a purchasable spare. Hope that helps.

Hello, I am a Product Manager for CCTV based in Melbourne, I work for a major global company...I'll be popping in now and then to say G'day. Look forward to chatting with you all. Piotr D.

Yes and no. Resolution in TVL is a measure typically used in the "analog" world - in other words, taking the raw composite video signal and displaying it on an analog CCTV monitor while having the camera pointed at a standard test chart. The more TVL the higher the resolution and (theoretically) the better the image quality. Pixel resolution is typically used when talking about the image displayed on a remote client connected to a DVR (ie when streaming video). This is not the same as TVL resolution...because even when the viewing window has an effective display area of 640x480 pixels (common for some IP cameras), this does not mean you have exactly that level or resolution -don't forget that the image has been digitally compressed (via Wavelet, JPEG2000, MPEG4, H.264 or other compression codecs) - hence some resolution is always lost. You may well end up with a number of the pixels in our 640x480 image being identical - this is the typical "blockiness" artefact you seen when playing back some DVD movies or movie trailers online. My point is that the number of "effective pixels" on the CCD not always an indication of final image quality as seen either on an analog monitor on an a remote client streaming video off the DVR. In fact the latter case can be worse, as you're now depending on two things being done right 1) the camera electronics being good enough to generate a quality image 2) the DVR frame grabber and encoder being of sufficient quality to create a high enough quality image. As an example - many PC-based (and some embedded DVR's) only record at CIF resolution (this is 320x288 in PAL). So - regardless of how many effective pixels the CCD has, or how many TVL the camera can produce, your final playback image will at best have 320x288 effective pixels, since that's the weakest link in the signal chain. Hope that makes sense.

Another option to consider... GE/Kalatel has (I think they still have it) a "fixed dome" option for their CyberDome housings. There is a "extreme weather" option available in their outdoor housings, that has a heater/blower rated down to lower temperatures than their normal PTZ housings. They also have at least two vandal-resistant options....one's a 7" "Rugged" housing that can be ordered with a tougher polycarbonate bubble. The other is a rather tougher 8" Heavy-Duty unit that's kinda octagonal and also available with a tougher polycarbonate dome. It may be worth double-checking if that same option is available in the fixed housing option. Failing that, you could use a "normal" Cyberdome PTZ housing and then get a fixed camera conversion bracket that snaps into a standard PTZ Cyberdome. OK...that gets you the dome/housing. For the camera, you could use any compact full body camera....such as GE's KTC-215CC or a Sanyo VCC5885. Both should fit quite nicely into this type of dome housing, with anything other than a ultra-large varifocal lens. Now this may end up being somewhat more expensive....but at least it gets you a housing with a dedicated heater/fan assembly that's rated for lower temperatures, along with a decent full-body camera that performs well in low light. An added bonus is that you can get a relatively fast lens (like a Fujinon 2.8-8mm F0.95, much larger aperture in low light than a typical F1.6 lens on a fixed dome camera).

Seikou Optical/Avenir makes a 7-70mm varifocal lens, that has its end threaded. GE Security has a similar lens, the KTL-7-70VA - look under http://www.gesecurity.com/portal/site/GESecurity/menuitem.f76d98ccce4cabed5efa421766030730?selectedID=9348&seriesyn=true&seriesID= Siemens also has a similar lens, the CLVD1318/7-70 Look under https://www.hqs.sbt.siemens.com/fs/content/20_20_10_20/20_20_10_20_060en.htm There are others as well. I have one of these lenses and it looks like a standard photo filter will screw in...something like a 46mm (49mm is too big, 43.5mm too small, don't have a filter in-between).

The number of effective pixels refers to just that - how many pixels on the CCD chip are physically collecting light. You will find that most 1/3" cameras today have approximately 440,000 pixels (approx 752 x 582 on a PAL camera). That number will be slightly different on a NTSC model. However...you can have two cameras with an identical pixel count producing quite different TVL resolution. The TVL resolution refers to the FINAL resolution of the camera...once you take into account all the post-CCD capture processing. This is where the overall design and build quality decides. That's why a full-body camera (for example a Bosch Dinion, a Siemens CCBC1337, a Panasonic SDIII or a Sanyo) will almost always produce a much cleaner, sharper image than a cheapie dome camera that uses nominally the same Sony CCD. So....at the end of the day it's the quality of the overall electronics that's driving the CCD which will determine the final TVL resolution of a camera, not just the raw pixel count.

Did you try contacting GE's tech support? Flash upgrading a DVMRE is not difficult, however it is critical to have the correct version for the unit. They may even be able to do it remotely (assuming your DVR's are accessible over the Internet).

Something else to be aware of - several manufacturers make cameras with a feature called "Peak White Inversion". This feature automatically "darkens" utra-bright sections of the image into a black spot. If used in a LPR application, the feature can automatically darken an automobile headlight, thereby reducing the overall dynamic range in an image. It is not necessarily a purpose-built feature for LPR applications, but can help. Some examples include the CCBC1225 & CCBS1225 1/2" cameras made by Siemens. They are not the only ones. I think Ness in Australia may have has a similar feature on one of their 1/3" models that they call the "Eclipser". Again the intention is to darken the ultra-bright portion of the image to improve the overall result. The Peak White Inversion feature is built into the camera- it does not rely on IR illuminators. It may be something seriously worth considering for LPR applications.

Not all varifocals are created equal. Not all use fully multicoated glass optics Some even use plastic lens elements as it is cheaper. Some varifocals (especially those that go quite wide, eg 2.5 or 2.8mm at the wide end) noticeably distort vertical lines away from the centre of the image. Finally - all varifocals suffer from lens "ramping" - this is where the maximum lens aperture deteriorates as you zoom the lens in. Those of you who do either 35mm or digital SLR photography will be familiar with the "standard zoom" commonly sold, being a 28-80mm F3.5 - F5.6 type lens. The key here is that the maximum aperture is F3.5 @ the wide end, but falls away to F5.6 at the telephoto end. CCTV varifocals suffer the same way...but virtually all lens manufacturers only publish the "best" F stop...being at the wide end. For example, a 5-50mm F1.3 varifocal is NOT F1.3 at the 50mm end. This is NOT intended as a recommendation to NOT USE varifocals....but just be aware there are some inherent compromises in their design. They are great for many applications...but not all are equal.

The Kalatel RS422 Digiplex wiring scheme cannot be run in star configuration unless you use a dedicated signal distributor such as their KTD-83 or KTD-83-16. Otherwise you will get problems. In the absence of a data distributor, the only way to run RS422 Digiplex is with a daisychain configuration, where the telemetry wires are looped from one camera to the next. This way you can run the cables up to 3000 metres, IF you are using the correct grade of unshielded twisted pair. I have seen it done with CAT5 cable, but due to the smaller cable size you are more limited in how far you can run the wires. The other thing to be mindful of is termination - there is a little termination switch inside the CyberDome, literally at the top of the housing (where the little RJ45 connector plugs in). In a star configuration (with the data distributor) you must have each camera terminated. In a daisy-chain configuration, only the last camera in the daisy chain is terminated, the others are not. Finally - check the dip switch settings on the KTD-312 to make sure it's set to be putting out RS422.

The other unit that may be worth looking at is the TransVu made by Dedicated Micros.... In QLD you can get them through a couple of distributors - DVR Solutions and Siemens BT Products. I know the TransVu is used in a number of "Cash in Transit" applications so it's a proven product. As far as cameras it depends what your needs are...I would suggest that any good "vandal resistant" dome is a good starting point. In vehicles the main problem will be vibration...so you want a camera where you can well and truly screw-lock the camera position as well as zoom/focus on the lens in place.

Sorry blowrabbit, I never tried it with a Panasonic camera...have seen it done with Pelco Spectras & the Pelco Probridge - results were acceptable, obviously functionality was somewhat limited to pan/tilt/zoom, focus in/out, iris control & presets...but it worked. I have anecdotally heard of a few Panasonics being driven off DVMRE's but have never seen it done. Sorry.

Hmm....that's a possibility. The CBR-PB2-PA Probridge would have only contained the Panasonic protocols that were "current" at the time. If Panasonic has since changed the functionality of their domes, then yeah, the PB may no longer be able to drive it...unless GE have revised the Probridge. Is it possible to configure the Panasonic dome to different protocols? I know that some PTZ's do have multiple protocols - for example the GE Legend and Siemens CCDA1435 series can be set to talk Pelco for example.

Have you tried putting in a ground loop isolation transformer? I know the post doesn't mention ground loops...but isolating the video signal may be worth a shot.