GTM21 Big improvement? Testers wanted

[inspiredron]

I have been playing today with my GTM21 and had some surprising results.
On a journey from Surrey to Canterbury and back with the incorporated antenna spread across the bottom of the windscreen of my motorhome, I stole several glances at signal strengths on the hidden menu. I found that, in general terms, a signal strength of 30 was too low and 40 was enough for TMC data reception. (Obviously I drove through several different transmitter areas)
On returning home, not far from Croydon and a good TMC reception area, I honed this down (in my lounge) to signal strength 35 - message count stays static; at 38 message count increments. TMC tuned itself to channel 187 and was steady on that channel. I then extended the aerial by sticking a pin through a length of wire and the end of the antenna, just behind the stop. The additional wire made no discernible difference to signal strength.
I then fitted a ferrite core between the 660 connector and the TMC receiver. Signal strength increased reliably by 4 or 5 points with different antenna positions. Fitting another ferrite core just the other side of the TMC receiver made no discernible difference to the signal strength. HOWEVER, with both ferrites fitted I was able to wind the antenna into a tight coil, reducing signal strength to low 20's and still see the message count incrementing, albeit somewhat slowly. Carefully removing the ferrites without disturbing the coiled up aerial stopped message reception.
Clearly I need to do some more testing on this but it seems that the ferrite cores may have two effects:
1. To marginally increase the signal strength
2. To increase the useable signal (ie to reduce the message signal threshold required for data reception)
As I said, I live in a good reception area, so it is always frustrating when TMC reception disappears on a journey. I would be interested if others could test out my findings in their areas.
I do appreciate that my test has been very superficial and in relation to a unit that has already locked on to a signal. I would expect a higher threshold to be needed to establish a lock on to a signal.
Many PC leads have ferrite cores on nowadays and they can be easily borrowed for testing purposes.
Please post your findings.
Note for Mods: Please move this to the TMC forum if appropriate. It could also be applicable to TomTom users and others.

[AlanGPS3]

The ferrites cannot increase the strength of the signal nor alter the sensitivity as they are passive devices. Their purpose is to absorb unwanted radio frequency signals on the cable.

If your results are repeatable then they will be absorbing RF interference energy coming down the cable from the 660. This unwanted RF will be adversely affecting what is called the signal to noise ratio of the signal being presented to the GTM21.

In other words they are not increasing the wanted signal but are reducing the effects of RF noise produced by the 660 on that signal. The result is nevertheless the same - it works better!

Some further experiments for you to try:
1. Put both ferrites on the same (effective) side of the GTM21 and note the result
2. Put one next to the 660 only, note the result.
3. Put both near the 660, note the result
4. Put one near the 660 and one near the GTM21, note the result.
5. Put just one next to the GTM21 as you have done above to provide a base reference, noting the result. That needs to be done again so that as far as possible all the tests are done in the same signal strength from the transmitter.

Please post your results!

I have a 510D and will try it out on that too, hopefully this weekend.

[inspiredron]

AlanGPS3 - Thanks for the comments
What we don't know is what the signal strength indicator in the hidden menu is showing. I realise that the ferrite cores are passive devices and cannot do anything to increase the "real" signal strength. But the number shown as "signal strength" increased consistently. Not only that but the aerial did not have to be spread out.

The tests that I described were in my lounge with no power to the 660. The GTM21 was powered from the 660 battery only Unlike in th ecar with engine running, the only RF interference to filter out will be that generated by the 660. I cannot think of any home induced RF apart from my wireless router as both the TV and our PC's were off.

I have not had time to carry out more tests but last night I was planning a journey for this morning and had a very low signal strength until I spread out the power cable from the unit. It seems that the power cable also acts as an aerial. I did not use the unit on my journey.

I need to do comparative tests in the car with the engine running as well.

PS There is only room for one ferrite between the 660 and the GTM21. Another variant would be to wrap a turn or two of the aerial around a ferrite. I remember from my early days of constructing transistor radios that I used a ferrite rod for the aerial with a winding around the ferrite rod tuned to resonate with a variable capacitor!

[AlanGPS3]

If the signal strength indicator is anything like it usually is in other equipment it's showing the strength after detection and filtering. Reduction in noise will increase that reading, as the receiver will not be so de-sensed.

I've tried it this morning with the unit outside the car, as with yours. In part of my house all signals are weak, due to a local hill, and there the TMC either doesn't work, or takes ages to register. With the ferrite clipped on I would say it was quicker to register, so it looks like it's doing something.

The 510D has no TMC signal strength meter that I can find, even in the hidden configuration and test menu.

Putting the ferrite elsewhere had no effect.

The only way to test this properly is on a trip out to the M25 just north of the Dartford Tunnel, where I can normally guarantee no signal until at least the M11 exit. However, I've no plans to go that way at the moment.

[inspiredron]

[AlanGPS3]

Thanks for the information on where to find the TMC strength. Results:

In house, no ferrite 20. Errors about 50
In house, with ferrite 24. Errors about 19

Touching the end of the cigar lighter increased the signal from 20 to 32, and from 24 also to about 32. That I would expect as the cable end is free and my body capacitance to earth would be altering the RF properties of the cable.

In car parked on the drive, no ferrite, 27-29
In car, same position, with ferrite 34, and immediate re-tune from channel 83 to channel 183, signal strength the same but errors down from 49 to nil.

The ferrite certainly seems to be having a beneficial effect. I don't have any spare ferrites to test additional locations on the cable.

[philpugh]

[inspiredron]

I've done a little more testing and taken a photo of the screen and ferrite positions.

I tried winding the aerial wire around a third ferrite - no difference at all to signal strength so increasing the aerial inductance does nothing.
In the car the signal strength is a little greater than in my lounge. With the aerial wire tightly coiled I got strength varying between 15 and 28 and the threshold for receiving messages seemed to be 18. Ch 187 is my strongest local signal but when it lost that it sometimes locked on to Ch 83. Laying the aerial wire out across the bottom of the screen gave reliable high 40's and into the low 50's for signal strength from 187. Switching on the ignition and running the engine gave a very slight increase, probably due to voltage as internal battery was getting a bit low.

Edit once to add image as per the OP request - MikeAlder

Thanks Mike for adding the image. As you can see there is no room to loop the cable through the ferrite closest to the 660. I have looped it for the one on the cigar lighter side of the TMC. Note particularly the tight coil of the aerial wire.
I forgot to say that the strength does depend on the layout of the power cable - if that is also reduced to a ball then the signal drops right away to low teens or below and the unit hunts for a better signal.

[inspiredron]

A bit more info:
This morning I did a trip from Crawley back home to Carshalton (M23, M25, A217) with the ferrites in place. Traffic was slow to start but once locked on was there for nearly all the journey. Mostly Ch 152 with odd spots of 83 and 187 once I got near to home. Signal strength was good, seldom below 25 and going as high as 60.
But here's the rub: From time to time the signal strength dropped below reception threshold and IMMEDIATELY the TMC starts hunting, first with a seemingly restricted list of channels and then through the whole lot. Two comments on this:
1. The restricted list did not always include the channel just lost.
2. The scanning speed appears to be too fast to allow the unit to lock on. Several times there was a short loss of signal on 152, the unit scanned past 152 with a signal strength in high 30's or 40's but DID NOT STOP.

It seems to me that this is a software issue that needs addressing for the UK market. In the US or mainland EU where signal strengths are high this behaviour would be OK. By the time a signal goes weak enough for the lock to be lost there will be another strong signal waiting to take over and easily locked on to, even with a fast scan - and probably fewer potential signals anyway. Here, we have lower transmitter power. I suspect that the lock gets missed while the unit is scanning if the channel is scanned between the TMC message transmissions. Naturally, it thinks that the signal on that channel is useless. If the scan was slower it would take the units longer to scan the whole spectrum but the chance of a lock on any channel would be enhanced. Also, common sense would suggest that, if the unit has had reception from a particular channel but then loses the lock, theunit should try to re-connect to THAT CHANNEL for a reasonable period (say 1 or 2 minutes) before abandoning the channel for a full scan.

I reckon that software changes to address these two points would significantly enhance performance, not just in fringe areas, but also in urban "tunnels" where multipath reception causes temporary losses. It would not help the Winter Hill issue, of course.

[swing]

Based on my experiences of watching that screen, I would probably agree with you too - it does seem to move onto the next channel too quick to be able to reliably work out whether there was a signal there.

I do wonder if Garmin are working on something along these lines at the moment though.
_________________
Please don't be offended if I do not reply to a PM - please ask questions via the forums.

[Hamie]

I'm not certain the signal strength is telling us anything really useful though...

With an unmodified GTM21 I can get a signal level in the high 20's - low 30's and get no messages... When I connected my useless piece of wet string (YOu know, the bit barmin call an antenna) to my car amplified antenna I get messages with signal levels as low as the high 10's. And rarely see the signal level go much above the 20's. Yet TMC is locked almost always...

Of course the coverage of the sensors and accuracy of data still leaves a lot to be desired...

H

[inspiredron]

Thanks for the feedback, Hamie. That implies that signal to noise ratio is an important parameter (as might be expected) and that your amplified? signal is cleaner and that you can get locks easier at low signal levels, as can I with the ferrites.
I hope that we can all be positive - I keep my piece of wire dry! We all know that UK TMC is not as good as it should/could be. If we can assist Garmin/TomTom and the users to get better results from pooling our experience that has to be a plus. The manufacturers are mainly US where, as I said, the environment is very different. It is not really surprising that the UK tail is having some difficulty in wagging the US dog. That does not make it OK for Garmin or TomTom to ignore the problem but I don't think it helps if we just gripe.
And TrafficMaster Freeway information was always spot on when I used it - always provided that it was interpreted properly. I had to make up my own mind whether it was worth diverting for a 10 minute delay. Less was never worth it, more usually was but the 10 minute one was always a problem and needed to be assessed in th elight of the previous and next reports.

[AlanGPS3]

I've been doing a bit more experimenting on this after downloading the latest TMC update from Garmin at the weekend.

I previously couldn't get traffic details when parked in my garage at home, but with the ferrite and new firmware it works fine now.

When looking at the signal strength it's best when running off the C510D internal battery, with or without the lead being plugged in to the cigar lighter. As soon as I turn on the ignition (lead plugged in), but without starting the engine the signal suffers, although still remains locked. That indicates that RF noise on the cigar power supply is having a detrimental effect too. However, as it seems to work then so what!

I didn't try it with the engine on as that would mean a run out to warm it up properly (cold starting and then turning off leaves neat petrol in the catalytic converter, shortening its life).

Until I get on the M25 NE section, where I've never had a signal, I won't know if this is all now OK although it does look very promising.

[adrian_anastas]

I wondor if this will work with a GTM 12?

The power is supplied straight from the unit so don't no if it will make a difference.

Can you provide details about the specs of the ferrites you're using? I'm thinking that if a 2 or 3 quid purchase could get my traffic happening from my front door it would be well worth it.

[inspiredron]

We see you’re using an ad-blocker. We’re fine with that and won’t stop you visiting the site.
Have you considered making a donation towards website running costs?.
Or you could disable your ad-blocker for this site. We think you’ll find our adverts are not overbearing!