Garmin Edge 705 - How is speed measured?

brazilianhipster

:?:

I’ve recently purchased a Gamin Edge 705 (with the limited use so far, pretty impressed, however, not impressed having to folk out a further £30 for a decent map!)

Can someone explain how speed is captured on the device?

Is speed measured by GPS? If it is, then why did I have to fit a magnet onto one of the back wheel spokes?

If GPS is used to record speed, how accurate is it?

Cheers - Steve

rich_e

GPS.

Unless you fit the GSC-10 Cadence/Speed Sensor (which it sounds like you may have).

It will use the GSC-10 if fitted, as it's more accurate when properly calibrated to your wheel size, as obviously GPS is subject to fluctuations, especially when obstructed.

You can get away without buying any maps.
Search for Openstreetmap, or any of the 705 threads on here. You can install free maps which are just as good, if not better than the City Navigator.

amaferanga

You only need the GSC-10 if you're interested in cadence or if you're cycling indoors on a turbo and you want speed/cadence..

brazilianhipster

Thanks Rich_E

I do indeed have the GSC-10 Cadence/Speed Sensor.

Are you saying that if this is fitted, then the system overrides the GPS i.e. it ignores the GPS data for speed and takes it directly from the sensor?

Cheers - Steve

schweiz

amaferanga wrote:

You only need the GSC-10 if you're interested in cadence or if you're cycling indoors on a turbo and you want speed/cadence..

or if you cycle through a tunnel, or if you cycle in high sided valleys, or densely wooded areas or anywhere else where GPS coverage may be restricted.

There's been this discussion before about where a 705 gets its speed from. I have both a 305 and a 705. The blurb that came with the 305 says that the GSC-10 will only be used for speed when there is no/insufficient GPS data. Apparently, it is written somewhere on the internet (Garmin Forum?) that the 705 will always take speed from the GSC-10 if fitted. Why this is different from the 305 I have no idea. The only thing that makes me think that the 705 does use GPS for speed calculations is that after an auto-pause outside with GPS on, it takes longer to auto restart than on the rollers with the GPS switched off.

rich_e

That's correct.

And what amaferanga said is not really true, as like I already pointed out, the GSC-10 gives you the most accurate speed readings. Obviously it's up to the end user whether they want to spend the extra money for a more accurate reading, cadence readings and the ability to use it indoors.

John.T

There is a lot of discussion on the Garmin site forum as to whether the GCS10 overides the GPS or just backs it up when sat info is poor. As yet no deffinative answer from Garmin. I think they are all keeping their heads down from the flak from the Bing map fiasco.

amaferanga

Rich_E wrote:

That's correct.

And what amaferanga said is not really true, as like I already pointed out, the GSC-10 gives you the most accurate speed readings. Obviously it's up to the end user whether they want to spend the extra money for a more accurate reading, cadence readings and the ability to use it indoors.

Note that I said NEED by which I meant without the GSC-10 the 705 won't give you the information you need at all without it.

kettrinboy

As far as the accuracy of the GPS goes, in the manual for my Edge 305 it gives the accuracy of the speed reading as <0.05 m/s depending on the quality of sat reception, dont know what that is in miles per hour as i,m not good enough at maths to work it out, but on my motorbike an indicated 160 mph dead comes out at 148.2 mph on the 305 which seems reasonable given that most speedo,s overread quite a bit at these speeds.

antfly

The 705 uses the sensor to measure speed if you have it, simply because it's more accurate.

schweiz

antfly wrote:

The 705 uses the sensor to measure speed if you have it, simply because it's more accurate.

How is it more accurate? If you leave the 705 in auto wheel size mode, it's calibrating the GSC-10 off the GPS data anyway!

Also, the reed switches on the GSC-10 are not all that sensitive and account for the GSC-10 missing revolutions of the wheel and the crank. I have no accurate data to show how often wheel revolutions are dropped, but again from data gathered on the rollers with the GPS off there are definitely periods of missing data.

Admittedly, on the other hand there is the issues of the GPS not taking gradient into account but the inaccuracy is minimal (50m in 10km distance error on a 10% slope).

The best solution would be to calculate a hybrid value from both data sources. Spikes and troughs could easily be removed and a more accurate value calculated.

antfly

Don't put it in auto wheel size mode then, calibrate it yourself. GPS only claims to be 97 to 98% accurate and steep hills don't help nor does going around in small circles e.g when you are waiting for somebody. Counting wheel revolutions just gives you the distance your wheels have travelled, it doesn't need complicated algorithms to work out anything. Simple is always best, why do you think that Garmin use it on all their latest models? If you are missing data perhaps you need to put the sensor closer to the magnets or put a new battery in.

schweiz

antfly wrote:

Don't put it in auto wheel size mode then, calibrate it yourself. GPS only claims to be 97 to 98% accurate

That's more accurate than most cars on the road.

antfly wrote:

Counting wheel revolutions just gives you the distance your wheels have travelled, it doesn't need complicated algorithms to work out anything.

It is if the count is accurate although using lat/long is hardly a 'complicated' algorithm.

antfly wrote:

Simple is always best, why do you think that Garmin use it on all their latest models?

I'm not sure they do. There's lots of talk on this on the Garmin forums yet I have seen no definitive answer from a Garmin source.

antfly wrote:

If you are missing data perhaps you need to put the sensor closer to the magnets or put a new battery in.

I'd rather have a slightly inaccurate speed than a GSC-10 through the spokes!

antfly

Don't use a sensor then if you can't use it properly, nobody's forcing you.
BTW I have had a definitive answer from a Garmin source.

rich_e

Well it does sound all a bit vague on what gets used.

However, you can have GPS accuracy as one of the fields displayed and you can certainly see if you have that up that it will often end up in a situation where it's a good bit out, therefore your speed can't be quite right. Therefore the GSC-10 is likely to give a much more accurate reading.

It's the same with a car GPS, they often aren't quite at the right speed you are driving. GPS is accurate when unobstructed and driving at a constant speed, but not so good at other times.


I personally have the GSC-10 fitted to my Road Bike, as I wanted the cadence information. However, it's probably my Mountain Bike that needs it more so, as riding in woods and such it's not quite so accurate.

It's the same with my Forerunner 305. I usually run along a path that is shaded by trees and it will show me a route once I've uploaded that isn't totally accurate to where I actually was.

schweiz

antfly wrote:

BTW I have had a definitive answer from a Garmin source.

Care to share that source with the rest of us? This has been discussed several times here and on the Garmin forum and I've not come across anything.

John.T

schweiz wrote:

antfly wrote:

BTW I have had a definitive answer from a Garmin source.

Care to share that source with the rest of us? This has been discussed several times here and on the Garmin forum and I've not come across anything.

+1. I also have not been able to get correct info on this.
My experience with an Edge 500 is that when set to auto wheel size and paired with a GCS10 it seems to maintain a size within 1cm from my measured one. Also when used without my mileage is close to the average of the rest of the group so I am happy with it's accuracy.
It also follows track pretty well other than in trees. ridewithgps does it better than Connect.
http://ridewithgps.com/trips/141160
This was on a MTB treasure hunt. On the faster sections in trees it can be as much as 30ft out but in open areas it is often even on the right side of the road. The unit was working under a perspex map board with a map on it which may have affected accuracy a little.
The only problem I have had has been eratic HR readings which I put down to me wearing the belt too low.

antfly

So what's the confusion?

John.T

antfly wrote:

So what's the confusion?

Your reluctance to show us the scource of and what was the definitive answer.

antfly

My source was Garmin, surprisingly enough, but you could just try reading the manual.
page 28


http://static.garmincdn.com/pumac/Edge5 ... Manual.pdf

Simon G

I always calibrate my bike computer's wheel circumference value using an average of three measurements of the wheel (including the fully inflated tyre), so I trust that my polar computer's speed sensor is accurate. I rode with a friend on a 60 mile route a few weeks back and noticed at the end of the ride, that his Edge 705 quoted quite a different avaerage speed to that of my Polar computer. He's got the Edge speed sensor, but clearly the sensor was not in play during this particular ride. We spent 20 minutes or so doing some rolling calculations using my Polar device as the benchmark and concluded that the Edge705 was only using the speed sensor when explicitly set to do so via one of the menus!

I cannot remember the exact set of menus used to switch the sensor on, but I do remember the menu option being somewhere fairly obvious!

schweiz

The confusion is caused by the following:

The Garmin manuals for the 705 and 500 only refer to cadence being continually recorded with the GSC-10 fitted.

The manual tells you that without the GSC-10 fitted, speed and distance is calculated from the GPS, which we know.

But the manuals don't explicitly say that speed and distance only comes from the GSC-10 when fitted and whether the 705 works like the 305.

When the 305 was sold, the GSC-10 was marketed as a sensor that provided cadence and also provided speed/distance when there was no/weak GPS data. This seems to be the case and despite the question being asked on the Garmin forums, no one from Garmin has answered the question about how the 705 works (in the threads I have read) although I've not looked at the Garmin forums on this subject in the last year.

From personal experience, there does seem to be some difference in how quickly auto restart kicks in with and without the GPS switched on, using a GSC-10 in both cases.


Simon G. all accessories have to be paired with the head unit for them to work. As long as a GSC-10, HRM or Power Meter is paired, data is sent to the 705. I'm not aware of any other page where you can 'switch the sensor on'. Any other 705 users out there know of anything?

John.T

This states that if no GCS10 is paired GPS data is used for speed and distance. No argument there. But it does not say whether GPS data or GCS10 data has priority when paired. From the Garmin forums it appears that many believe that GPS is only over-ridden when the GPS signal is poor. The question asked of Garmin is what actualy happens but they have not answered it. The problem seems to affect MTB riders on forest trails who would expect poor tracking due to tree cover but still get good speed and distance from the GCS10. They are not getting this thus they asked Garmin why.
I have not had any problems but as I only use the GCS10 on my TT bike so that bridges do not interfere with the data (not fully tested yet) I would not have expected any. As you can see from the Old Town Bank Road section on my ride it looses the plot by up to 30ft in places and this is not heavy tree cover. I am awaiting my next forest event with interest to see how it performs. I do not want to fit a GCS10 to my MTB so may have to fit a cheap computer to get distance for navigation. I ride MB Orienteering events.
schweiz. I have looked recently and people are still asking for clarification. Garmin's response to customers seems very poor. Just see all the 'bring back Google maps' posts. I think they have answered one and just with a bland cut and paste from their Blog.

rich_e

The Edge 800 Manual states:

Cadence data from the GSC 10 is always recorded. If there is no GSC 10 paired, GPS data is used to calculate the speed and distance.
Cadence is your rate of pedaling or “spinning” measured by the number of revolutions of the crank arm per minute (rpm). There are two sensors on the GSC 10: one for cadence and one for speed.


That would lead one to believe that it's the GSC-10 speed data that is used, and not GPS.

schweiz

it can only be inferred, it is not a definitive statement and it remains a simple question that Garmin have not/will not/do not want to answer when directly asked on their forums.

Maybe the programmers keep changing their mind on the preferred source and it depends on what firmware you have. I know I haven't updated my firmware for a long time an given the errors reported in recent firmware loads, I think I'll leave it like that.

amaferanga

I'd have thought that if you input a ridiculously large or small wheel circumference into your Garmin it should be clear when you're getting speed data from the GPS and when it comes from the GSC-10.

FWIW I have a FR310XT and have used it with a GSC-10. I don't bother now, but I can tell you that things changed recently between firmware updates so I wouldn't place too much on what is in the original user manual for the 705.

antfly

I asked the question a few years ago and had the answer that unlike the 305 the 705 uses the sensor to record speed and distance all the time. It says nothing in the 305 manual about using the GPS to record s and d if a sensor isn't attached because the 305 uses the GPS all the time except when the signal is poor or when you turn GPS reception off, as has been said.

You could use one of these on a MTB. http://www.wiggle.co.uk/o-synce-macro-d ... ensor-kit/ it doesn't have any sticky out bits.

Edit: You could always have a look at the 800 manual, I doubt that has had any updates yet. It says the exact same as all the other manuals.

Simon G

I've got the Edge 605 and not the 705, but having looked through the menu options on the 605 I seem to recall that the menu where you can 'enable' the speed sensor was under 'Settings' then 'Profile and Zones', then 'Bike Profile' - I'm pretty sure that there was a speed sensor switch here!

Anyone out there with a 705 who can confirm or deny?

antfly

GARMIN KNOWLEDGE BASE



Just to put it to bed,






Question: How does the Edge 605/705 record speed, cadence and distance when using a GSC 10 Speed/Cadence Sensor?
Answer:

Distance, speed and cadence are calculated using different methods when a GSC 10 Speed/Cadence Sensor is used on an Edge 605 / 705.

The information is calculated as followed:

Distance - The Edge 605 / 705 uses the GSC 10 speed to calculate distance. If the device is not using a GSC 10, then distance is calculated by GPS.

Speed - If the GSC 10 is being used, but it is not calibrated then the speed information will come from GPS. If the GSC 10 is calibrated, then the speed information comes from the Speed/Cadence Sensor.

Cadence - Cadence will always come from the GSC 10.



Did this information answer your question?
Yes - this fully addressed my question
No/partially - I will continue searching
No - I would like to submit a follow-up email

No - I would like to suggest changes or additions


Other recommended links
Edge 605
Edge 705

amaferanga

antfly wrote:

GARMIN KNOWLEDGE BASE .....

Someone should tell the Garmin Knowledge Base that the 605 isn't compatible with the GSC-10.

schweiz

Alleluia! Thanks for finding that Antfly

Now, a new question..how close do you all run the speed sensor arm to the wheel magnet. On both the BMC and the VN, it isn't possible to tuck the arm in behind the chainstay due to the clearance so the arm has to stick up. It's as close as I dare have it, especially on the BMC which has rectangular stays and therefore the GSC-10 isn't seated well.

My 705 is in auto wheel size as I use it on all my bikes and with different wheels and I don't bother to switch the bike profile when I switch bikes. Occasionally I get a 'Wheel Size Calculated' message after a km or so but more often than not I don't. I'm starting to think that I'm losing data although speeds and distances always seem to tally +/- 1% with other non-garmin computers on club rides.