Sweep times

[zerbey]

Has anyone done any testing/have any data on the time it takes popular RDs to sweep the bands they're scanning? I'm not even sure if this can be tested without information from the factory but there are far smarter minds than I on this forum. It would be a very useful indicator of performance.

[MEM-TEK]

I have wondered the same thing for a long time. My best guess is that the average RD takes 1 full second to sweep through all three radar bands. Some RD's, notably many Cobra models, seem to take much longer to sweep through all three radar bands.

Bel/Escort prefers to sweep each band individually. This requires a LO board design with inherently high sensitivity in order to use rather fast sweep rates so that all three bands can be swept in a reasonably short period of time (under a second).

Some RD models sweep two bands at the same time. This allows the overall sweep rate to be slowed since two bands are being swept at once, leaving more time for sweeping the remaining band.

Several Whistler products sweep both X and K bands in one sweep. This allows more time to be spent sweeping the Ka-superwide band, even though sweeping both X and K bands at the same time does lead to alerts on X or K band occasionally and falsely jumping from X to K or vice versa.

The Valentine V1 sweeps all three bands simultaneously. This is one of the reasons why the V1 is able to consistently respond to extremely short I/O radar gun bursts. Yet this sometimes creates false bogey counts and/or false alerts on other bands due to harmonics and ghosting, and of course more false alerts to cheap import radar detectors.

Obviously there is no single and ideal solution unless somebody could combine the features of a variety of LO board patents which are held by different manufacturers. Yet Bel/Escort, Whistler, and Valentine have managed to really get the most out of their specific designs. In other words, there are several different ways to skin a cat. And all three of these RD manufacturers have pretty much perfected their methods for skinning this radar cat. That is not to say that any of the solutions are perfect. None of them are since the manufacturers can't infringe on competitor patents. Yet their solutions are pretty darn good.

[zerbey]

Shame there's no proper way to test this, I turned POP off mainly because it adds a tiny amount of time to the sweep but I'd love to know how much since I've had the occasional POP encounter.

[jdong]

I bet you can measure the sweep time by using a scope to monitor the frequency of the LO(s) (i.e. time-aware DFT)

[nickstath]

I looked my archive and i found that in 2005 speedzones they measure the radar detector's scanning rate. Here are the results:

Detector

Immediacy of Response Time

Bel Vector 995

.60 second

Bel RX 65

.67 second

Whistler ( all models)

.73 second

Bel 895

.77 second

Passport X50

.96 second

Valentine One

1.01 second

Cobra ( both models)

1.24 second

PNI Steel Eye

1.48 second

Radio Shack 3000

2.34 second

[zerbey]

Those V1 results are surprisingly slow, I wonder if it was older firmware or the fact that the V1 scans all three bands balances things out.

[CJR238]

My guess is there is a sweet spot time for scanning a band. Enough time to thoroughly check the band but also not long enough to get all the junk.

[jimbonzzz]

Shame there's no proper way to test this, I turned POP off mainly because it adds a tiny amount of time to the sweep but I'd love to know how much since I've had the occasional POP encounter.

There's no question that there are radar units in Florida that could have the POP feature. However, there is evidence to suggest that POP cannot be used in Florida. I posted about this before, here's an excerpt from the MPH manual:



(Note the "Not Present in Florida Units")
I don't know what implications this has for certain. It might be possible that there is some kind of legal restriction to using POP in Florida, I don't know. Anyways just an FYI...

Jim

[jimbonzzz]

You can measure the sweep rate with a scope as long as you have a good idea of how the detector operates and you know where to probe.

MEM-TEK means well and always tries to help people out on here. But unfortunately, on the other side of the coin he has a looooong history of posting technical information on this forum which is, to put it nicely, "less than accurate", and this post is no different. For example, the V1 does NOT sweep all three bands simultaneously.

However if you throw out the "absolute" things that MEM-TEK posted and take the whole thing in general, there is something valid to take away from his post: different detectors have different methods of sweeping and validating a detected signal. Which is why I don't think the actual sweep rate alone is a good indicator of performance. If detector A sweeps all bands in 200 ms and detector B sweeps all bands in 300ms, you can't automatically conclude that detector A will perform better.

All detectors use software rules (to prevent false alerts) which require the signal to persist for a certain duration of time before an alert is triggered. Technically, the duration can be different based on frequency, signal strength, or any other info the detector has available to it (such as direction in the case of the V1). Ultimately it's these software rules are what determine if the detector alerts or not, not the sweep rate.

In the past, there have been a couple of different (but related) parameters tested, and both have some merit:

1. What is the time from radar signal present to alert?

2. How long must a radar signal persist to trigger an alert?

#1 is difficult to measure accurately. This is essentially what was tested by SML above, but using stopwatches ("Immediacy of response"). A more accurate method might be to film the test sequence on video, and then count the frames after the fact.

#2 is generally done in a lab, where timed bursts of radar can be transmitted to see if the detector will alert. Michael B did this with some detectors a while back, here's a couple of the threads:

http://www.radardetector.net/forums/...e-3-826-a.html
http://www.radardetector.net/forums/...testing-3.html

(You'll note from the Michael B tests above that when POP is ON, the average duration to trigger a regular Ka alert was longer. I believe this IS a side effect of the longer sweep.)

#1 does not provide a clear picture of detector's ability to alert to short-duration signals, while #2 does. However, #2 does not take into account any delays in processing/alerting by the detector (however small) while #1 does. Of course, neither of these provides the whole picture. And in addition, there have been no tests performed to date which take into account the signal strength, frequency, etc like I mentioned above.

Anyways, food for thought...

Jim

[MEM-TEK]

^^ I was hoping that you would jump in and clear things up.