slow to go...

[inventor]

hey guys, have any of you ever had your segway take longer than a few seconds to get up to speed off the charger? pushing you off the back, slowing down and then speeding up with a mind of its own for several minutes. if this is a battery thing i'm not surprised as i have had issues lately with batteries. these are two year old L-ions on a gen1 machine and somedays this problem is worse than others. just want to ask if maybe there is an old post i could review.

[bentbiker]

hey guys, have any of you ever had your segway take longer than a few seconds to get up to speed off the charger? pushing you off the back, slowing down and then speeding up with a mind of its own for several minutes. if this is a battery thing i'm not surprised as i have had issues lately with batteries. these are two year old L-ions on a gen1 machine and somedays this problem is worse than others. just want to ask if maybe there is an old post i could review.

Are you going downhill immediately after charging?

[KSagal]

I have noticed a wierd thing that may fall somewhat in this catagory...

Since 2003 I have been parking my seg(s) in the cellar, and I have a garage under the house, so I go in thru there.

I park both segs in the cellar, and they are pretty much always plugged in.

Lately, I have noticed a change to something that I have been doing right along...

Typically, I just remove the cord, close the cover, start the seg, get on, and ride across the cellar. Since Fibber McGee's closet designer also worked in my cellar, I go slow. I get to the garage, thru the door and glide up the ramp to the street.

In the past few months, my i2 gives me a flat face, and even once an unhappy face climbing that 40 feet of ramp. Total climb is only 5 feet.

In the past few months, my LiIon e-167 gives me a flat face, or an unhappy face, and one time, gave me a shutdown at the top of the ramp at the street. Kind of caught me by surprise, the orange blinking and the shake and all. I stepped off. Scratched my head. Re-started. Never dealt with that again.

I am sure my machines have conditioned me to take it slow climbing that first ramp. I have always thought it was a function of a cold machine, a slightly overcharged battery, and a large current draw to accellerate up that hill. By going slower than normal it does not show up... (Sometimes flat face but not more)

None of this can be duplicated after the seg has been on for more than 5 minutes or so, or I have traveled 1/10 mile or so...

I do recall that my old batteries gave me some reduced performance when cold, but do not recall the machine displaying it.

[hellphish]

I've found that the best way to get around this is to start your segway, put it in balance mode, and let it idle against a wall for a few minutes while you put on your shoes, pack your bag... whatever. I only see the symptoms you describe if I unplug it, turn it on, then start going right away.

[GlidingAlan]

Are you going downhill immediately after charging?


I see where bentbiker was going with this... I've had a similar thing happen. I live on a hill and when I unplug my Segway, hop on and go immediatly down hill it acts just as you describe. I believe because it was just charged, the regenerative braking has nowhere to store the energy and causes this. I recall this from back in 2002 when I did the train-the-trainer at the Segway facility. Attached is a page from the instructor manual...

[bystander]

Hey guys, have any of you ever had your Segway take longer than a few seconds to get up to speed off the charger? pushing you off the back, slowing down and then speeding up with a mind of its own for several minutes. if this is a battery thing I'm not surprised as I have had issues lately with batteries. these are two year old L-ions on a gen1 machine and some day this problem is worse than others. just want to ask if maybe there is an old post I could review.Your question has been pretty much answered, but here are some observations you may be interested in:

When the PT is first powered up, the controllers give the power system a bit of a stress test. This lasts a few seconds. If you happen to be riding it at the time (during the test), the test is harder to pass and may require further testing, making the flat face behavior (reduced performance mode) last quite a bit longer.

The workaround, as others have mentioned, is to let the PT just sit there for a few moments before mounting. I don't know the exact time interval, probably something like 5 to 10 seconds. 15 seconds would probably be longer than necessary.

As the batteries age, they may take a little longer to pass the test sometimes, so this behavior, while always present and built into the PT, may not be so noticeable when the machine/batteries is/are new.

This symptom is often more obvious immediately after removing power from the charger. This is because the thing the stress test is testing for is how the batteries react to a substantial power drain. When the battery is partially discharged, it reacts in a linear fashion (which makes for a short test), but when fully topped-off, it's a non-linear result (due to the typical discharge curve of a battery) and the controller is programmed to err on the side of safety. A non-linear stress test result is also the symptom of a failing battery, so the test can take a little longer when the battery is all the way charged, just to make sure things are alright.

My i-167 v14.2 occasionally does this (reduce performance mode) when I forget to give it a few seconds to itself after power up. When it happens (flat face appears), I bring the PT to a stop, wait about 3-5 seconds, then start moving slowly, observing blinky. Almost every time, (by pausing like that), everything is good to go and I can continue at full speed. I find that if I do not stop, that the condition lasts quite a while longer.

Note: my "startup path" is flat, I don't have to deal with downhills right after a full charge. If you do, you definitely need to follow the advice of letting the PT sit for a minute or two (or several minutes) to bring the battery down enough to allow full speed on the hill.

[dynk]

I have noticed anytime I just unplug the machine and step up on it, it tends to push forward from me (as if I'm going to fall forward). It only lasts for 10 seconds max, then it is fine. I figured it needed a few seconds to acclimate itself to not being plugged in.

[polo_pro]

First let me say, I like bystander's answer. That fits what I've experienced.

I also want to add some other experiences for comparison.

I use to commute daily. I'd pull my segway off the charger and go down a 20 foot hill. Invariably, I'd ALWAYS find the handlebars in my gut and practically falling off the back of the machine. In an earlier reply, someone described this.

This differs from what the OP said. I've experienced this a fair number of times. And I'll add that I almost exclusively glide on flat ground now. It's a gentler pushback right after unplugging and mounting up. Sort of saying, "I'm not fully ready yet." A minute later, my segway is performing exactly as I expect with no unusual (but gentle) push back.

Now I'll also comment that I use hellphish's technique alot. I first started using it on my long distance glide down PCH. On some mornings, I'd start out giving a few demos before I hooked up the loaded trailer. Invariably, I'd get really great range from that first set of batteries!

I think if you don't interrupt this self test that bystander describes (just as hellphish says), then the builds up a good pictureof what the batteries are capable of. Then with a more accurate picture, it can more carefully draw down the batteries so that you get better range.

I bet a tour operation that has to do fast battery changes (with demanding customers waiting in the wings) might be able to coorborate this hypothesis of SLIGHTLY reduced range due to immediate use.

[Nospin]

The second week I had my X2 I took it out for a climb up a rough forest road, 3 miles with average grade 11% and lots of rocks and water washes. Everything worked fine, 6 mph avg speed up and 1/3 batt used. But coming back downgrade I noticed the (! + frownyface)alarm was showing. No other effect, no LSF feedback except that if I just slowed down to < 2.5 mph the alarm went back to smiley face but then the display alarm would come back on again if I went over about 3 mph. I cycled it through this condition several times and then stopped and toggled to the SP1 (turtle). There was no alarm then even at the 6 mph SP1 max. After I got back home it seemed to be OK and I haven't been able to repeat it again. I spoke to David at INC but he hasn't gotten back to me yet. Maybe just a new system break-in anomoly?

[bystander]

I think folks will find that the PT is more limited downhill than uphill. Why? Because of the regenerative charging.

It may not be obvious to all, but the workload on the powertrain is heavily influenced by operating up and down slopes. The PT is limited to dissipate about 1 hp on average. When going uphill, power is used to move forward, but the incline extracts additional power as if you were deadlifting the weight. When one encounters a steep enough uphill slope, more power is being used to lift the weight of the rider + device, than is being used to move. When the combined power averages over 1hp, you get reduced performance mode.

Downhill, you're getting power back, of course, but there's a limit as to how much can be taken safely. The limit is less than 1hp average, because the regenerative charging circuits only have so much capacity. How much? Well, a PT plugged in uses about 120-140 watts. Engineers from Segway, Inc. have indicated that the regenerative charging rate is about triple that of the AC charging rate. This puts us into the 360-420 watt range or about 1/2 hp.

That would explain why folks can go up a slope a little faster than back down without going into reduced performance. Depends on the slope of course, and the weight of the rider + device. Heavier gen2 and xt models are at a slight disadvantage unless the rider is very light.

For example, on the following charts, my PT with me aboard weigh about 275 lbs. If I travel along a slope of 7.8° (a.k.a. 13.6%) at 10 mph, the elevation change is causing a power draw of 1 hp. That's over and above the power drawn used to turn the wheels. So chances are, going 10 mph up a 7.8° slope is an un-attainable limit, at least for more than a few seconds.

If I turn around and go down that same 7.8° slope, it's likely that I will get reduced performance mode if I try to stick to 10 mph. Because I know the downhill limit is closer to 1/2 hp (using the 2nd chart, 7.8° falls between the 4 & 6 mph columns), I'm better off not attempting to travel faster than about 5 mph. Of course, I can go a little faster than the 1/2 hp imposed limit, as power is used to turn the wheels, and regenerative charging is not all that efficient, a bit of heat is produced. This lets us exceed the "limit" by about 20-30% (guessing). Probably could do 6-7 mph or so on that down slope. Gen2s may be more strict about exceeding the downhill limit than gen1s.

How fast (almost) you can go uphill for a sustained period of time:
http://forums.segwaychat.com/gallery/albums/userpics/13270/full.jpg

How fast you can go downhill for a sustained period of time:
http://forums.segwaychat.com/gallery/albums/userpics/13270/half_hp.jpg

Interactive javascript chart attached. Unzip the attachment, open the html file in a browser, and permit javascript to execute. You will be able to select the amount of hp in the combo box. Various chart cells have pop-up hints.

The second week I had my X2 I took it out for a climb up a rough forest road, 3 miles with average grade 11% and lots of rocks and water washes. Everything worked fine, 6 mph avg. speed up and 1/3 batt. used. But coming back downgrade I noticed the (! + frownyface)alarm was showing. No other effect, no LSF feedback except that if I just slowed down to < 2.5 mph the alarm went back to smiley face but then the display alarm would come back on again if I went over about 3 mph. I cycled it through this condition several times and then stopped and toggled to the SP1 (turtle). There was no alarm then even at the 6 mph SP1 max. After I got back home it seemed to be OK and I haven't been able to repeat it again. I spoke to David at INC but he hasn't gotten back to me yet. Maybe just a new system break-in anomaly?

P.S. In the anecdote that Nospin related, since he characterizes the road as "rough", the x2 may have overreacted to the strong vibrations for a sustained period of time. By taking the time to change from normal to turtle mode, it may have been enough for the severe shock perception to fade from the part of the program running in the x2 that reduces performance when encountering severe vibration.

[hellphish]

From what I gather reading comments from i2 owners, you get reduced speed downhill no matter what the charge level is. Can someone please confirm/deny?

As an anecdote, I have never been forcefully slowed down when going downhill on a Gen1 machine.

[cmonkey]

I2 units do have a reduced performance when going downhill.

and as far as reduced performance goes at startup... I'm surprised no one has mentioned this yet...

when you first power up (at least on the gen I units) you WILL get a flat face when you try to immediately take off and try for max speed.
This will occur for the the first 5 seconds of operation.

[bystander]

I2 units do have a reduced performance when going downhill.

and as far as reduced performance goes at startup... I'm surprised no one has mentioned this yet...

when you first power up (at least on the gen I units) you WILL get a flat face when you try to immediately take off and try for max speed.
This will occur for the the first 5 seconds of operation.A number of posts touched on that. See #3, 4, 6, & 7.

The other posts (one of mine included) are on the slightly off-topic tangent of reduced performance when going downhill after a full charge.

Since the OP hasn't made a follow up, we don't really know which kind of problem he is having. (or both).

[KSagal]

One of the reasons I offered what I did, is because my routine has not changed substantially, but the action of the machines has.

A secondary consideration is that I am only on level for the first 50 or 60 feet, then up hill with the overly charged batteries...

I have found that my i2 is more conservative on the down hills under most all conditions compared to the gen 1, but now with 14.2 in both machines, the gen 1 is more conservative than ever before as well.

[bystander]

One of the reasons I offered what I did, is because my routine has not changed substantially, but the action of the machines has.

A secondary consideration is that I am only on level for the first 50 or 60 feet, then up hill with the overly charged batteries...

I have found that my i2 is more conservative on the down hills under most all conditions compared to the gen 1, but now with 14.2 in both machines, the gen 1 is more conservative than ever before as well.Yes, the internal resistance of the batteries has risen slightly as they age. And the capacity has gone down a few percentage points.

And the latest versions of firmware are more conscious of the dangers of overcharging.

Although this behavior seems to be a performance hit, by avoiding the overcharging on downhills, it's increasing the cycle life count of the packs.

I wonder if anyone will appreciate the fact that their batteries last a bit longer (weeks, months, even a year?) as a compensating factor for the flat face when trying to go downhill too fast.

[inventor]

yes, this idle time when allowed on start up does cure the sluggish starts, and there is no slow down when NOT coming directly off the charger:) i do live on a hill so from now on i will consider going up for the first glide of the day rather than over charging my already fully charged segway going down the street... thank you

i would like to know if this sensor you mention that is responsible for testing your batteries is located within the batteries or on board the segways electronics? and if you are so inclinded to say as you all are in this forum, does this sensor testing calibrate the battery gage each time because mine runs out the last bar long before the batts are drained these days.

just trying to decide if new L-ions will solve both my problems of slow warm up and mis-calculating blinky or if i have some other ghost in the machine.

i would love to disable this battery testing feature temporarily to troubleshoot... but where?

[bystander]

Yes, this idle time when allowed on start up does cure the sluggish starts, and there is no slow down when NOT coming directly off the charger:) I do live on a hill so from now on I will consider going up for the first glide of the day rather than over charging my already fully charged segway going down the street... thank you

I would like to know if this sensor you mention that is responsible for testing your batteries is located within the batteries or on board the segways electronics? And if you are so inclined to say as you all are in this forum, does this sensor testing calibrate the battery gage each time because mine runs out the last bar long before the batts are drained these days.

Just trying to decide if new L-ions will solve both my problems of slow warm up and mis-calculating blinky or if I have some other ghost in the machine.

I would love to disable this battery testing feature temporarily to troubleshoot... but where?There isn't a discrete sensor per se. The processors run a set amount of current through the motor's coils and measure how much the voltage drops at different levels of current. This process gives the processors a good idea of each pack's internal resistance, which in turn determines the performance envelope of the whole thing. By the way, others have reported that if one pack is signifigantly stronger than the other one, operation can be impaired.

Because everything is so tied-in together, your best bet if you wish to troubleshoot is to find another person with a set of batteries and swap them, noting any performance changes. By exchanging packs in pairs and one-at-a-time, you might be able to isolate a weak pack. But that may take quite some time to try all the combinations.

Gauge calibration on NiMH and Saphion packs is accomplished differently in each case. NiMH gauging is essentially an expanded scale voltmeter (and it can suffer from the "surface-charge" inaccuracy). Saphion gauging is somewhat of a "coulomb-counter", that is, the processors track how much power goes in and out over time and apply the "guesstimate" to the display.

Under certain rare conditions, the Saphion gauge can become inaccurate, but only if you never take it out for a long trip (nearly discharging it to minimum indication).

If you want to recalibrate the gauge (note that Saphions "recalibrate" and NiMHs "recondition"), you have to leave the PT in riderless balance mode and rest it against the wall until if falls down (might take 30 hours or more if you start from a full charge), then plug it in and leave it plugged in for 12 hours to charge it. You don't have to start from a full charge, you can ride it around normally until it's low, then let it rest against the wall. That way it should take much less than 30 hours to fall down.

By performing the gauge recalibration on Saphion packs, you are sacrificing one of the lifetime recharge cycles. It's only loosing 1/1000th of it's life, even so, it's something you don't want do too often. Doing the "recalibration" repeatedly back-to-back on Saphion is not recommended, but is recommended on NiMH, up to 5 times.

New Saphions would probably "solve" the things you mention because, compared to your old packs, the new packs would have lower internal resistance and would have up to 20% more capacity. But there probably still is plenty of life in your present packs.

Keep in mind though, that since you are on v14.x firmware, that the processors are programmed to be more strict about the safety margin than they were back on earlier versions. So that may be the "ghost" you are mentioning. No getting around that.

[inventor]

a most excellent collection of information, and bystander you must have built these things. i treated the segway to a short riderless balance mode after the charge and headed up hill from my residence first thing this morning so as not to force a downhill charge until there was capacity to do so. blinky smiled the whole way so it's out of sight out of mind for me concerning the slugglish start up problem. i know my L-ions are over the hill but they do still go the distance ;).

bystander, are you saying that the next set of batteries i purchase will be 20% more range or power than the two year old set i own now or are you just guessing this is my lose of effeceincy over time?

[Sharkie]

a most excellent collection of information, and bystander you must have built these things. i treated the segway to a short riderless balance mode after the charge and headed up hill from my residence first thing this morning so as not to force a downhill charge until there was capacity to do so. blinky smiled the whole way so it's out of sight out of mind for me concerning the slugglish start up problem. i know my L-ions are over the hill but they do still go the distance ;).

I don't understand why you say they are over the hill. If they still go the distance, there is nothing wrong with them. Or, perhaps you're another one that believes the BS about a 3 year shelf life? Even regular old lithium ion batteries don't have that problem, I have some that are now 4 years old, and have the same capacity as when they were new. It constantly amazes me how facts are ignored, but one silly idea can perpetuate itself on this board.

Jim

[bystander]

bystander, are you saying that the next set of batteries I purchase will be 20% more range or power than the two year old set I own now or are you just guessing this is my lose of efficiency over time?Not "will be", but "may be" 20% more range. I can't tell you for sure how much your existing Saphions have degraded since they were new. You can look it as a guess, but it's more of an uncertainty figure.

The reason I came up with the 20% figure is that the baseline capacity of new Saphion packs is 800 watt-hours. And back at Segfest 2006, the representative from Valence (the subcontractor that supplies the Saphions) mentioned that, in order to make sure all the batteries they produce for Segway, Inc. have 800 wh, that they actually make some of them up to 20% more capacity (up to 960 wh).

The point of bringing this fact up is to point out that there is a 20% "fudge factor" when people try to figure out exactly how much range they can get. With a 20% possible difference in results, it can be difficult for people to make tests on their PT and expect other people to get exactly the same results.

So, lets say someone says they weigh 200 lbs and have a i-180 and they can go 20 miles over a flat route on a full charge at full speed, and someone else (same weight & model) tries that and finds they can only go 17 miles. They are both correct because their results agree within 15% and the variability due to the capacity of the battery is 20%.

The best way a PT user can tell whether their batteries are "over the hill" is to measure how far they can go on a single charge. Segway, Inc.'s website says you can go 24 miles. I find this is only true (for me) if one travels less than full speed.

To do a full range test, it is useful to run a known distance route, or install a bicycle odometer. If concerned about running out of power far away from home, one can circle the block a number of times, once blinky indicates little range is left.

[polo_pro]

To do a full range test, it is useful to run a known distance route, or install a bicycle odometer. If concerned about running out of power far away from home, one can circle the block a number of times, once blinky indicates little range is left.

A friend was seeing degraded performance with his Li-Ions. To debug the problem, I borrowed this pair of batteries and did as you suggested. I laid out a 17 mile circuit. I ran it three times using my segway. One time with one of his batteries and one of mine. A second time with the other one of his batteries and the other one of mine. And then because neither of those tests showed anything odd, the third time was with both of his batteries.

I did notice the odd behavior where the indicated battery level dropped quicker, but it made it the whole distance and got safety shutdown about the same time as on the other circuits. So with no serious problem shown, we did one last test using HIS segway and HIS batteries on the same circuit (but again with me doing the gliding). Same odd behavior with the battery level dropping quick, but it made it the whole distance. Almost every variable had been removed at that point (well except riding style and the way that the batteries were charged).

I guess the moral of the story is "don't assume your batteries are on the way out until you've gone through GREAT lengths to prove it."

ps - That being said, this friend still has exhausted his Li-Ion batteries by the end of a polo match. But then so did I this weekend. I still think he rides real aggressively both in day to day life and on the polo field!

[hellphish]

^^^^ Truth


The only annoying thing about having a wonky battery gauge is that the segway limits its own performance based on what the guage reads. So even though the batteries last a while, being limited to "only 2 bars left" levels of performance is sort of a bummer.

[bystander]

Re: bummer.

Or why the PT reduces performance when only a couple of bars are left.

Not trying to justify the PT's programming, just my explanation of how I think it works.

A PT spends most of it's operating time at low to medium power levels (1/5-1/2 hp). It doesn't "know" for sure how many watts are available at very high levels, such as those over 1hp. It can operate in the high wattage region briefly, such as recovering from a bump, or low traction detection event.

The key to predicting it's own behavior at higher current levels is to determine each battery pack's internal resistance. Without this ability for it to predict this, the PT would operate in flat face mode at all times.

The PT measures the voltage of the batteries while operating. It also measures the voltage drop on the batteries while various amounts of current run through the motor windings. This allows it to calculate the battery's internal resistance. Internal resistance varies with the current drawn, temperature, state of charge and age of the battery. Internal resistance is very important if you want to predict how many watts of power you can get out of the battery at a given power level.

In general, internal resistance is low at low currents, and higher at high currents. For example, If one calculates power remaining in a battery by loading it with a low current drain, one cannot simply scale that answer up to a much higher wattage figure because it becomes wildly inaccurate. The inaccuracy is due to the fact that a battery's internal resistance is generally higher and non-linear at higher current drains.

In a PT, there are two batteries, and two motors. Each motor has two independent sets of windings, call then side A & side B. The side A windings are both connected (through the front controller board) to the front battery pack. Likewise, the side B windings connect to the rear battery pack.

For example, lets say the user is running up a slight slope and 100 watts are being consumed per winding in each motor, two motors, two windings each, for a total of 400 watts from the pair of batteries.

Now let's say there is a catastrophic failure in the power train, and one winding in one of the motors opens. In order for the PT to stay balanced, the current in the non-open winding has to be increased to 200 watts, with the extra power drawn from the battery it is connected to. This occurs nearly instantaneously and lasts until the safety shutdown ends. It can last for several seconds, depending how quickly the user can safely dismount.

For several seconds, the PT finds itself delivering 300 watts from one pack and only 100 watts from the other. Can it do it? What if the person is running up a steeper hill and is drawing 750 watts (1hp) total from the set of packs? In that case the PT would be temporarily be drawing 3/4ths of that from one battery (or about 562 watts) and 1/4 (188 w) from the other. Can it do that?

To avoid faceplants under these conditions, the PT puts itself into reduced performance mode based on what it calculates for the values of the battery pack's internal resistance. If the PT determines that it can't get 560 watts out of the weaker of the two packs for 15 seconds at any given time, it's going to go flat face when the user demands too much torque from it. It tries to keep itself from getting into a position it knows it would fail to recover from. This means the user sees sluggish acceleration / deceleration and slower speeds when climbing inclines.

A pack can have plenty of power in it, but the internal resistance is the "bottleneck" the power must flow through to be effectively delivered to the motors in any given instant. Given enough time, the power can be drawn. But there isn't always enough time.

In the above scenario, 560 watts would be needed for 15 seconds to recover (aka 8400 watt-seconds, or 2.3 watt-hours), and the internal resistance calculations may indicate that it will take 25 seconds to obtain the 2.3 wh from the batteries (due to the presence of that infernal internal resistance). It doesn't matter that the pack still may have 200 watt-hours left in it, the important part is that the system could need 2.3 wh for 15 seconds and calculates that it won't be able to get it.

So even though the pack has good capacity, if the PT determines that the internal resistance is high, it will set limits on itself as a safety feature.

[inventor]

i knew i should have kept the original software version and not have my segway over compensate or be too sensitive to such performance demands. anyone offering to uninstall the lastest version and rollback the programing?

[bystander]

I knew I should have kept the original software version and not have my segway over compensate or be too sensitive to such performance demands. anyone offering to uninstall the latest version and rollback the programming?I'm not sure if you understand. The latest software lengthens the cycle life of your battery by protecting it from overcharging.

If, by some miracle, you were able to go back to 14.0, your batteries would wear out faster, and you would end up spending more on replacement batteries.

Unfortunately, technology hasn't progressed enough yet for the best of both worlds.

But it could happen someday, I suppose.

[ryan_walters]

So even though the batteries last a while, being limited to "only 2 bars left" levels of performance is sort of a bummer.

Whenever I've done full range tests on my i2, I only notice any difference in performance once I've gone down to 'zero' bars left. If I've still got 1 (out of 8) bars on the infokey, I get the same performance that I did at 100% batteries. Once at 0 bars, I will get lower and lower performance right up 'till safety shutdown.

[SEGsby]

Tis better to be safe, than kissing concrete. ;)

SEGbsby

^^^^ Truth


The only annoying thing about having a wonky battery gauge is that the segway limits its own performance based on what the guage reads. So even though the batteries last a while, being limited to "only 2 bars left" levels of performance is sort of a bummer.

[bystander]

Whenever I've done full range tests on my i2, I only notice any difference in performance once I've gone down to 'zero' bars left. If I've still got 1 (out of 8) bars on the infokey, I get the same performance that I did at 100% batteries. Once at 0 bars, I will get lower and lower performance right up 'till safety shutdown.hellphish's Saphions may not be evenly matched. They (apparently) have a long and unknown history. His results may be atypical.

When the two batteries are well matched, that is, they have nearly the same capacity, the blinky display is fairly consistent.

But when one pack is significantly stronger than the other, or the packs have been used enough to develop extra internal resistance, all bets are off about how "consistent" the blinky display is. (except that it is consistently inaccurate).

I believe Mr. Protocol and cmonkey both experienced (in-warranty) problems with their Saphions that gave odd blinky indications from time to time before it was determined that there was an actual battery failure.

And by the way, hellphish has a gen1, so two bars is 40%, not the 25% it would be on an gen2.

But thanks for the anecdote about i2 performance at low bars. Good to know what to expect for other i2 users.

[SEGsby]

Herbie had the same "cantankerous" attitude, until his battery was replaced under warrenty.

SEGsby

hellphish's Saphions may not be evenly matched. They (apparently) have a long and unknown history. His results may be atypical.

When the two batteries are well matched, that is, they have nearly the same capacity, the blinky display is fairly consistent.

But when one pack is significantly stronger than the other, or the packs have been used enough to develop extra internal resistance, all bets are off about how "consistent" the blinky display is. (except that it is consistently inaccurate).

I believe Mr. Protocol and cmonkey both experienced (in-warranty) problems with their Saphions that gave odd blinky indications from time to time before it was determined that there was an actual battery failure.

And by the way, hellphish has a gen1, so two bars is 40%, not the 25% it would be on an gen2.

But thanks for the anecdote about i2 performance at low bars. Good to know what to expect for other i2 users.

[polo_pro]

hellphish's Saphions may not be evenly matched. They (apparently) have a long and unknown history. His results may be atypical.

That's what I thought too when he first brought this problem to me. If you re-read my earlier post, you'll see the first two tests (where I matched each one of his batteries up with one of mine before doing the 17 mile circuit) were suppose to determine this. But in each test, I completed the circuit with shutdown within 100 yards of each other.

[hellphish]

Yes, my batteries are kinda spooky.

[bystander]

That's what I thought too when he first brought this problem to me. If you re-read my earlier post, you'll see the first two tests (where I matched each one of his batteries up with one of mine before doing the 17 mile circuit) were suppose to determine this. But in each test, I completed the circuit with shutdown within 100 yards of each other.
Yeah, what you tested was (mostly for) overall capacity, and only a little for possible higher internal resistance. It's important to make such a (capacity) test first, as it eliminates a couple of possibilities. But a low capacity symptom is not the only affliction that batteries can harbor.

Not sayin' hellphish's batteries are low capacity / low range. Your test demonstrated that the capacity is similar to "typical". I'm sayin' hellphish's batteries seem to be showing their age by exhibiting signs of higher internal resistance.

To do a field test which checks for possible higher internal resistance, you need a good, long slope, and if using both PTs "side-by-side", the PT+rider+cargo weight for the two should be as close as possible. One wants to measure the time it takes to go a distance up and down a set distance at a set (or averaged) angle. Then you can calculate how much hp the PT is dissipating. What's important here is the time it takes to travel between the start and end points of the course, and the altitude difference between them.

One would make a low speed run, say around 4 mph, up for 50 yards, then down for 50 yards, still at around 4 mph. Measure the exact time it takes for the 50 yards each way, so you can calculate the exact average speed later. Then repeat the test, attempting to keep the speed at 6, 8, 10, 12 mph. Depending on how steep a slope you picked for the test, one or both the PTs will start showing flat face and the average speed, (determined by the measured time interval) will reach a limit.

Oh, and try to keep the PT at constant speed, you will want to put the PT in motion a few feet before the "start" line, and measure the time interval of the passing between the start and end points. The reason for the constant speed is to avoid power surges inside the PT that would make the collected data slightly less precise.

The length in distance of the test (50 yards, 200 yards, etc.) isn't so important, what's important is to use the exact same linear distance and the exact same vertical distance during the suite of tests to eliminate variables in the testing. If only testing two PTs on one day, in one place, it's not too important, but if you want to compare tests to other places, or other days, you need to get both the linear distance and altitude difference recorded with a low uncertainty value. For example, if you use a GPS for the altitude readings, some (more expensive ones) have a built in barometer, and much less uncertainty. If you run a short course, and the altitude difference is only 100 feet, and your GPS has a plus/minus factor of 25 feet, that 25 feet could be in all 3 dimensions, so the uncertainly is 100 +/- 25, or 75 to 125 feet. That's quite an error factor when trying to deterring the hp used to lift weight that distance. When using a short course like that, it would be better to use an inclinometer and some trigonometry, rather than a GPS only.

If the slope is well suited (flat, even angle, no interfering traffic), but not steep enough, make some runs with additional cargo weight, say 25 or 50 lbs. The measurement of the weight of the PT+rider+cargo should be a low uncertainly value for best results when performing the calculations later.

If possible, one ought to re-do the 4 mph attempt at the end of the testing round, when blinky is indicating less bars. If the result is significantly less than the prior 4 mph attempt, it can indicate to the rider exactly how less aggressively the PT will respond under "low bars" conditions.

I know hellphish already knows how sluggish his PT can be at times, I'm not suggesting these tests be performed by him. Just outlining the procedure that could be done if anybody really wants to get meaningful results.

Also, when using a combination of batteries, as the PT does, _if_ one battery happens to be stronger than the other, the weakness of the weaker one will be masked. It can be very challenging to interpret the results of testing in order to reveal just how weak or strong the batteries are.

The blinky gauge just isn't precise enough to deal with certain marginal battery situations.

[polo_pro]

It sounds like you measure higher internal resistance via difficult terrain or hard riding. Now I did have a significant mile long steep hill (5% grade I'd guess) in this 17 mile circuit. And I took blinky measurements every few miles at set points so I could make fair comparisons about how the battery was being drawn down.

By the way, check out http://www.trailjournals.com/entry.cfm?id=134344 if you want to see me getting around your problem of "how high did I climb". You'll see me testing out a pair of Li-Ions for a long distance glide. I'd suspect that if you had a ride back down from the top of the peak and a spare set of Li-Ions (used as a control), you could do a very good "internal resistance" test by duplicating what I did twice, eh?

But I guess I do have one other question. In some sense, would a strenuous polo match be a good measure? I'm thinking if hellphish played one match on his segway and a second match with a fresh set of Li-Ions. If the matches were played on the same day with the same team members, I'd think alot of the potential variable would average out. This assumes that hellphish spends the same amount of time playing goalie, and generally plays either up or back for the same amount of time in each match.

ps - Now when might we play two carefully timed polo matches back to back, eh? 8^) 8^) 8^) Of course, it wouldn't be against the same team, but you know what I mean. 8^) 8^) 8^)

[hellphish]

Yes, if anyone wants to loan me their machine for scientific polo testing, that would be great.