I found that mine does the same thing. The time difference appears to be close to the amount of time I was stopped at traffic lights.

The timer on my LYC continues for about 30 seconds after I stop which happens to be about the same length of time the traffic light intervals are around here.

jaretj

Yeah but that would affect the displayed time and the downloaded time.

The display time is right - the downloaded isn't.

Plus one of these was a race - so no stopping.

FWIW I called Saris about this same issue. They did not really have an answer.

This sounds like the PT CPU keeps the timer going when you rack your bike. I think the default timeout for the CPU turning off is 4 minutes. Do a trial run of 5 minutes, and see if it reports 9 minutes in PA/WKO+. Then, maybe try changing the sleep time to 2 minutes, and see if that effects it.

A way to prevent the timer from keeping going is to put it in interval mode (hold both buttons for a second), and then at the end of your race, tap both buttons to get all the good race data in a single interval.

cool - any way to fix a file post download.

I have the same problem w/ my LYC. DD coaches a bunch o' folks who also have the same issue. It is not unique.

Even when I do an interval, the time for that interval before and after download differs. VERY annoying.
Mine isn't quite as dramatic as yours, more like :20-:30 seconds / hour difference.

Saris said to try downloading to PA instead of WKO+, to see if there is a difference between the 2.

(But in the meantime, I got a Joule, so I don't even know if it's an issue anymore, since I've been downloading to PA, since it wont go to WKO+ v2)

You can edit out the end of the ride where the PT kept recording time (running thru TA, sitting on the rack) and that will get it a bit closer.
But unless you can find a fix, you'll likely still have the issue.

---

float , hammer , and jog

emailed a pal @ saris today:

Thought about calling in and seeing what CS would say, but then I figured I’d cut it all out and go straight to the best contact I have for PT questions.

Have had an odd situation with my CPU+ LYC over the last couple weeks when I download certain workouts. I’ve had it occur 3 to 4 times, but know the specific details for only two of them.

What is basically happening is that when I download the ride – WKO+ and/or poweragent show the elapsed time as being longer than the displayed time (which matches the actual time for a ride).

For example – yesterday I did pigman. Rode 2:25 and change. Race split says 2:25, Display says 2:25. When I download, I get 2:29 – just slightly over 4 minutes longer. When I examine the downloaded file in a chart – it’s not 2:25 + empty data – it’s as if time is compressed and there are no gaps in the data. I suspect that it has to do with the CPU going to sleep – as I did a long ride recently that was 5:59 in reality, and the download was 6:07 – with two stops long enough to let the CPU sleep.

Here’s the use case for Pigman yesterday.

1> Rack bike in transition.
2> Wake hub/CPU up make sure they are talking.
3> Clear Computer by pressing Mode/Select together.
4> Manual Zero
5> Do other stuff/head to swim start – CPU goes to sleep.
6> Swim
7> Grab bike from rack – run to T1 exit
8> Mount, wake CPU up and ride.
9> Download and notice that downloaded elapsed time is 4 minutes longer than displayed time.

I’ve attached the PA export and the WKO+ file for reference.

The use case for the 6 hour ride is similar. Ride – stop and refill bottles. Ride more; refill bottles. Finish – display 5:59 – download 6:08.

Also – I just looked; I have a longer ride recently that had the CPU go to sleep because of stopping to meet folks and refuel, and the issue does not appear there – however that was using a 2.4 LYC that I was borrowing.

Here is how I've fixed that problem on 3 different PT's...so far.

Download/install latest version of Poweragent
open poweragent
put head unit on cradle & turn on
IN power agent go to Tools ->Firmware ->Check for Firmware updates.

It'll pull down the latest and greatest.

---

Brian Stover USAT LII
Accelerate3 Coaching
Insta

already at the latest and greatest :(

After numerous attempts to update the firmware on my Joule, I read here that I must update PA first, then update the Joule firmware. I downloaded PA, but during installation, it opened a window telling me about possibly overwriting data or data may be lost or some such crap, so I aborted the installation. Been meaning to drop a note to Saris to ask about it, but haven't yet. Have you run into that problem?

Greg

---

If you are a Canuck that engages in gratuitous bashing of the US, you are probably on my Iggy List. So, save your self a bunch of typing a response unless you also feel the need to gratuitously bash me. If so, have fun.
"Don't underestimate Joe's ability to f___ things up" - Barack Obama, 2020

No. I back up all my data files on an external HD every week. So in that case, once backed up, I'd let it ride.

I only have poweragent b/c it seemed my PT wouldn't download into wko without it.

---

Brian Stover USAT LII
Accelerate3 Coaching
Insta

Just curious, do you have the power agent that lists pedaling time, overall time and one other time (not in front of my home pc, can't remember what it was)

If so do any of those match up?

If you export to a CSV file what is listed there? The first column is time, that 8 minutes has to come from somewhere.

I've got PA 7.5.1.19. Just checked for updates and it says up-to-date.

I don't use it for myself or my athletes. I think it's a POS. Everything gets put into wko

---

Brian Stover USAT LII
Accelerate3 Coaching
Insta

Just curious...what does the avg mph say for your race? Was it correct compared to time/distance recorded on the head unit?

I did a race rehearsal this past wknd...56mi, 2:40:26...the avg should be 2.94mph, but the head unit was displaying 20.6 IIRC. I think the speed was calculating elapse time whereas time was moving time.

I guess I should check for firmware update...

The average speed when downloading is off - due to the extra time.

The displayed average on the CPU right now - shows the average that the downloaded time does.

I.e. THe head unit displays 2:25:12, and shows 57.23 miles - but displays an average of 22.9

Thanks for checking. That's the same issue I have. It's crazy that the dist/displayed time <> avg mph.

-Tim

For those that care - I'm currently exchanging mails with a saris CS rep + dev regarding this issue.

Don't know if anything will be discovered - but I'm OCD so I won't give up until they fix it or they send me something really nice new toy and tell me to go away.

Another follow-up:

Did an easy ride on the trainer on saturday. Started a timing when I started pedaling after clearing/zeroing the computer.

Finishing my ride and let the wheel spin down. PT computer stopped ticking @ 1:00:18 - stopped wristwatch at the same.

Downloaded as 1:01:11.


*The 18 and the 11 could have been swapped.

Also rode 3:54 yesterday - downloaded as 4:00.

Wow... this is interesting. The problem I've been noticing is that the *distance* is too long. I've done a bunch of TTs on a course that is ~10.15 miles, but the Powertap shows 10.32 miles. And yes the wheel size is correct and exactly the same value for both measurements.

I've had this same issue. Keep us posted as to what you find out from Saris.

Thanks.

Curious....is the downloaded time the same in PA and WKO+?

The PA download time and WKO+ download time are close enough that they are the same.

I noticed the same thing. Riding a 700x23 tire with the CPU set to 2096. I rode a course on friday that my SRM and prior to that my Cateye Astrale 8 measured as 48.3 miles - came up as 49.8 on the LYC.

I adjusted the tire calibration number for my ride yesterday as described here:
http://www.sheldonbrown.com/...on.html#measuredmile

Have to check the distance of the course to see if it worked, but based on landmarks I know - it seemed "better"

I do know I was following one of the Joule threads - I think they were talkign that it appeared to be off about 2-3% - which is what I saw with my LYC

Very weird... there are $5 computers that get speed, time, and distance correct.

Just checked one of the files (10.15 mi TT) where I had an iBike and the PT running together. The PT shows ~1.4% longer time, and ~1.9% greater distance. I'm using the power and speed numbers to determine the starting and stopping points, so there could be a couple of seconds variation between them... but not 20 seconds. Wheel diameter (Conti SS 20mm) is set to 2060mm for both and the rollout confirms that this is correct.

Very weird... there are $5 computers that get speed, time, and distance correct.

Hey - yep.

Most frustrating part to me is that my SRM was/is flawless - and the only reason i switched to powertap was for a new bike.

Now I have quarks to deal with on the new bike and quarks to deal with on the new power meter.

I love my life.

My PT computer dropped the ball severely at timberman.

The power didn't seem right for what I was doing while I was riding, power seemed pretty low, distance was fine. Rode a low 2:38:xx on... 145 avg power (I'm 165lb or so).

Downloaded it a few days later and according to the download I rode.... 16 miles. It's been fine up till that race, since I've transitioned to marathon training since the race, I haven't investigated it much.

---

---------------------------------------------------------
All we have to decide is what to do with the time that is given to us. ~Gandalf

I have had to set my wheel circumference to 2092 to get it to stop overstating the ride distance, on both the LYC and the Joule.

Also, fwiw, since I now have to download my Joule to PA, and then export that to WKO+ (I have v2, doesn't support direct Joule download), the distance and time overstating issue has gone away. I was advised to try this for the LYC, but I'm not using the LYC anymore - perhaps one of you kids could try it out.

---

float , hammer , and jog

THe distance being overstated isn't happening on download (atleast not for me) - it's prior to download.

Mine was too. Same thing.

The "time expansion effect" is what happened after downloading to WKO. The LYC head unit display agrees almost perfectly w/ my watch/GPS/HRM in terms of time elapsed, assuming I didn't have any stops.

---

float , hammer , and jog

Poweragent downloads it the same way. So does WKO+2.

Interestingly when you look at the avergae speed as displayed on the computer - it gets calculated with the expanded time - not the displayed time.

I.e. 20 miles in 60 minutes = 19.6 mph average displayed depending on what the expansion coefficient is.

I have seen this too. Correct avg speed based on true elapsed time/distance is more than what PT head unit shows while on the bike (i.e. PT unit is off on the low side).

---

-------------------------
http://trirochester.com

I've seen this as well. Always wondered if it was due to extra distance from passing / not travelling in exactly in a direct line from start to end. Now methinks it's something with the PT.

---

-------------------------
http://trirochester.com

In my case I had two different computers with the same tire circumference loaded, and the distances and times were substantially different. Every other computer I've used gives ~10.15 miles for this course... that's a Cateye, Sigma, and two iBikes.

Seems like people are having different issues, which makes this all the more mysterious.

ya know - I don't mind hounding saris on this, but the more people call in and can give explicit details on what they are observing - the more data the developers will have to figure out what's wrong with their firmware.

Gotcha thanks. Never got around to try putting another computer on my bike with the PT (which as you showed answers that question).

---

-------------------------
http://trirochester.com

I keep thinking more and more of switching back to SRM. But that means selling 3 PT wheels one which is a disc, the other an 808, 2 lyc and it seems like a BIG hassle.

On the other hand what good is having a PM if it can't get basic data correct?

---

Brian Stover USAT LII
Accelerate3 Coaching
Insta

Willing to cut a sweet deal on the disc?


I know where you're coming from - it's really, really annoying to me to look at my training logs and see the time of rides inflated by 1% to 2%. I'm actually about to just save off the "bad" files - export them to a CSV and apply a correction to the time stamps.

In some ways it might mess the data up more...


I've made arrangements to ship my LYC bad to saris for them to rip apart. It'll arrive in their hands with a ride: 3:54 that downloaded as 4:00, and calculates average speed based on 4:00. hopefully it will tell them something.

The replacement LYC - seems to still have the distance issue - i.e. distance is not accurate for a given rollout.

However - the replacement LYC seems to be much better time wise. When downloading into Poweragent it is as expected i.e. moving time matches displayed time.

When downloading into WKO+ the "moving" time is long, but seems to match up correctly for extra time at stops and such. Will see how it goes at IMWI and a sprint two weeks later.

FWIW - a 40 minute and change interval matched up time wise on interval display length and download time into WKO+ on a ride yesterday.

Will see how things go this week on rides - but the time display issue preliminarily thus far appears to be a "hardware" issue.

I'd consider it :-)

---

Brian Stover USAT LII
Accelerate3 Coaching
Insta

Just did a TT at Moriarty (40k) and both the distance and time were inflated by 1.8% on download. The time displayed on the unit during the race was correct. This is very weird...

I have the same problem and i'm pretty sure it's a WKO+ issue, not a powertap issue. I get same inconsistencies when I transfer data from my garmin or lyc (different data on the watch/lyc versus what I get in WKO+).

---

_____________________
Don't forget to attack!

My issue is occurring with Power Agent.

borrowed a buddies joule tonight.

Did a ride with the wheel sync'd to both of them.

Was a trainer ride about 45 minutes.

Wheel stopped moving @ about 45:05 on my stop watch. Joule stopped ticking @ 45:08 - cervo @ 45:18.

That seems consistant to me as I've observed the cervo has a fairly long "cool down" period.

Downloaded into power agent.

Joule - 45:08 moving time

Cervo - 45:44 moving time


f'd up.

This is a big problem, imho. I've been having it for at least 6 - 8 months now. I've called Saris at least 3 times now and have made no progress other than "try this, try that...." The problem was that I didn't have enough data at the time to determine exactly where the issue was. I wasn't sure if it was Wko+, although I had made zero upgrades to Wko+ when the problem started occurring, or some place else.

So, here's my data from Kona:

CPU -- Ride time = 5:20; Distance = 116.36
Wko+ -- Ride and total time = 5:34; Distance = 116.3 (this was actually a rare case where my mileage matched my cpu display but time was WAY off)
PowerAgent == Ride/moving time = 5:34; Distance = 114.8 (what was really strange here is that total time = 5:28 which was less than ride time)

Now when I exported my PowerAgent file to a csv and imported it into Wko+, I got the following:

Ride time = 5:25; Total time = 5:28; Distance = 114.8

I started my computer right out of T1 so mileage should be right around 112 miles (even if wheel circumference was off slightly). My official ride time was 5:24. I also wore a 310XT which I started about 1/4 mile after T1. It has my mileage at 111.8 and ride time was 5:24.

I have this problem on every single ride. As Scott stated, the longer the ride, the worse it gets. One thing I just remembered is that I sent my old YLC back to Saris last season along with my hub (due to a hub problem). I remembered they told me that they replaced my old YLC with a new one. I thought that was cool. Apparently that wasn't cool because that was about the time when I started having the problem.

Thanks, Chris

I've had the same problem since I got the ANT+ hub upgrade to my PT 2.4 about a year ago. Prior to that, my ride distances were exactly correct (based on comparison with GPS and mapmyride). Since the upgrade, the distances I get from the CPU are about 3% high - just about exactly the error you are reporting here. Ride times are wacked too, although I have tracked these less carefully.

I went through Saris support and they were even kind enough to send out a new head unit to see if that fixed the problem (it didn't). I was told that my only other option was to send in the hub for service. I have a feeling they won't find anything wrong and I will waste my time and money sending it in, only to get it back with the problem unresolved. This is clearly a software problem and it is time that Saris wakes up to that, comes up with a solutiion, sends out a patch to affected customers and restores confidence in their product.

Yeah, I've had a very similar experience except they haven't asked me to send my hub in for service. I wouldn't bother. I have two hubs and it's happening on both. I've sent an e-mail to Jesse (Product Manager) but haven't received a response.

Thanks, Chris

Yup. I have a Joule and LYC on my aerobars and the LYC reads longer every time. I put another email in to Saris. This problem should be their number one issue to resolve.

For sure!

A bit more background on my problem (for the benefit of any Saris support people reading this as much as anything). I have been basically been struggling with this problem the whole past summer. I'm using the ANT+ upgraded PT 2.4 with the LYC and downloading to PA. THe LYC and PA are both upgraded to the most recent firmware/software. I have done numerous tests on the same test course (approximately 9km long) and find that the distance is approximately off (high) by about 3% every time - primarily comparing to my Garmin 310XT but I have also tested against mapmyride and my 2009 Honda Accord odometer. The distance they measure agrees, the distance measured by the LYC is high by 3%. The strange thing is that the speed measured by the LYC and the Garmin 310XT agree exactly, once I get to a steady state (the speed averaging on these units is different). This tells me my wheel circumference is very accurate and clearly not the problem.

The distance error also persists when I upload from the LYC to PA. I don't use WKO+, I upload my data to TrainingPeaks - there error persists there, as expected.

If the Joule fixes this problem, I'd consider replacing the LYC. However, I'd prefer that Saris come up with a software patch to resolve this issue.

I bought a joule last week specifically because it doesn't experience this issue at this time.

I now just realized that the cpu Saris replaced for me about a year ago didn't have the ANT+ upgrade but the one I received when they shipped my wheel back to me did have the upgrade. I'm seeing a common denominator here. What's hard to tell is if there is another variable that factors into the problem. For example, is it a combination of having the ANT+ upgrade with the latest firmware? I'm pretty sure I was running the latest firmware on my previous cpu but not positive.

Anyone seeing this problem who doesn't have the ANT+ upgrade?

Most importantly, why is Saris continuing to ignore us?

Thanks, Chris

Since it's a problem with the LYC and not the Joule, I'm guessing that the problem involves the LYC firmware upgrade that is required when you do the ANT+ upgrade. I'm positive my LYC firmware is up-to-date, as I updated it through PA but also I had the problem with the replacement LYC that Saris shipped to me.

I'm curious to which communication channel the Joule uses - is it the same one as the LYC or is it the ANT+? If it's the same as the LYC, then the problem has to be the LYC firmware - because they should in essence be receiving the same information from the HUB. If that's correct, then this is probably a couple of lines of code to fix.

I'm going to see if Saris can resolve this one over the winter - if not, I'm pretty much decided on buying the Garmin Edge 500 as my replacement head unit before the spring season. About half the price of the Joule, gives you GPS and is a nice, compact little unit. Let's see what happens.

Trevor

First off an apology- certainly did not mean to make anybody feel ignored or slighted by our silence on the forum and lack of info. for those who have called in.

But, we've heard you loud and clear and are actively pursuing this issue. As you all know from reviewing this thread the issue isn't easy to pin down but you've all contributed a valuable piece to puzzle. We think we have the root cause isolated to the LYC firmware but need to confirm via some testing. We will then make the necessary patches available via PowerAgent.

Sorry for the lack of detail here but we're still investigating and I don't want to jump to any conclusions and give anybody the wrong idea. More details soon.

---

Jesse Bartholomew

Jesse,

Fwiw, consider the apology accepted. I can't thank you enough for the update. Very encouraging!! Just so you know, I'm always willing to put in countless hours to help you guys resolve these kind of issues. If you need data, I'll send you data. If you need a specific type of data and I don't have it, I'll find it.

Thanks again!!

Chris

i also have some files that are way off if you need them. But they are in wko+. lmk if you want them.

---

Brian Stover USAT LII
Accelerate3 Coaching
Insta

Ditto to Lakerfan's last reply. Jesse, if you need data and I have it I'll be happy to send your way. Looking forward to my LYC back to being one of my best and most reliable friends. :_)

Yeah!

Just realized I neglected to update this thread when we released the firmware addressing this issue.

Version 7.63 for wireless and 6.63 for wired should fix all these issues and is available via PowerAgent.

---

Jesse Bartholomew

I have an old wired powertap with firmware version 4.07. Is this time/distance problem an issue for me?

Frankly, as long as the power is accurate I don't really care much if the distance is off a little.

It's more then that, it's the time, which will skew AP and NP.

So in my case the samples are 1.26S apart. Is the bug that the time period is something other than 1.26?
Perhaps my PT is too old for this problem. I rode with it and an SRM and don't recall the power samples being skewed in time.