Jump to content

Strange behaviour?


Frank Skilbeck
 Share

Recommended Posts

OK, not my model but at the field today, one of the clubmates models lost power just after take off, fortunately he managed to get it down safely. We all assumed that the battery had gone faulty and the ESC LVC had cut in, but on testing the battery was showing 3.7v per cell when the ESC started to reduce the power (around 300 watts on his watt meter), plug in another same type battery and it will quite happily let that battery drop to 3.3v per cell (power around 500 watts) and the LVC doesn't cut in! Put the "faulty" battery in another model and get lots of power and LVC didn't cut in (battery around 3.4v per cell). Note none of the batteries even got warm when we were doing the testing and model had previously flown that morning and did a good 7-8 min flight with no issues, so we seem to have a problem with one particular battery only on that ESC, any ideas we are baffled.

 

 

Link to comment
Share on other sites

It could be a slight mismatch between the specific ESC/Battery connectors giving a higher resistance than other pairings. Fully recharge the battery, run the motor at WOT when the power starts to flag feel the connectors to check if they are getting warm/hot. 

Edited by PatMc
Link to comment
Share on other sites

Something similar happened to me with an ESC miscounting a 3 cell HV battery as a 4 cell & then cutting off very quickly. Unplugging & reconnecting restored normal operation as the initial voyage had dropped slightly by then. Try fully charging the suspect battery & checking the cell count beeps on connection.

Link to comment
Share on other sites

3 hours ago, PatMc said:

It could be a slight mismatch between the specific ESC/Battery connectors giving a higher resistance than other pairings. Fully recharge the battery, run the motor at WOT when the power starts to flag feel the connectors to check if they are getting warm/hot. 

 

We did wonder that, but there we couldn't feel any temperature increase in the connector, an XT60. It also did the same when we put a power meter in series between the ESC and battery.

Edited by Frank Skilbeck
Link to comment
Share on other sites

2 minutes ago, John Lee said:

Something similar happened to me with an ESC miscounting a 3 cell HV battery as a 4 cell & then cutting off very quickly. Unplugging & reconnecting restored normal operation as the initial voyage had dropped slightly by then. Try fully charging the suspect battery & checking the cell count beeps on connection.

 

We checked that, it's a 4s battery and we got 4 beeps on connecting. Very strange as pilot has several batteries of exactly the same make/capacity etc and they are all charged on the same charger, yet it's just this pairing. Put the battery back on when it's only showing 80% charged and it does the same, put a different one at 95%+ and it's fine.

Link to comment
Share on other sites

It sounds like one or more of the cells in that pack has got a high internal resistance. Drawing about 20A (your 300W) is causing the voltage on one or more cells to drop sufficiently for the ESC to reduce the power as it sees the total pack voltage reduce. Also the dropping voltage will itself produce a power reduction that appears similar to that when the ESC low battery protection takes place. If you can monitor the individual cells by plugging a monitor into the balance lead whilst running up the motor you will be able to see which cells the problem is occuring in. Obviously take all sensible precautions to restrain the model and keep clear of and behind the prop.

 

The second model in which the battery appears to be ok. Does that have a lower current draw? (Smaller prop, different motor KV etc.)

 

I assume the 3.7V per cell you mention is being measured off load? If the fault occurred just after take off, on a freshly charged battery, the cell voltage should have been closer to 4.2V per cell. Are you certain that the pack was recharged after the previous flight? Did you measure the voltage of all four cells?

 

If I understand you correctly, you are saying that straight off charge this pack is only reading 80%, yet is charged in the same way as other identical packs. If that's the case then it's definitely faulty.

 

Brian.

Link to comment
Share on other sites

Brian, no, we first thought it was a faulty cell, but on that battery the ESC is reducing the power when the voltage is 3.7v per cell and all the cells were good, (as measured with a battery monitor plugged into the balance lead) on another battery the ESC didn't reduce the power even though the pack was running at 3.3v per cell. The 80% charged reading was after the aborted flight and ground test run and all cells were within 0.01v of each other.

Edited by Frank Skilbeck
Link to comment
Share on other sites

Hi Frank,

 

I'm still not understanding why the battery voltage shortly after the aborted take off was only 3.7V per cell. It should have been closer to 4.2V. 3.7V per cell equates to around 15% remaining capacity. You wouldn't attempt a flight with that.

 

If that pack really is acting differently to all of the other identical ones, then perhaps there is a bad connection in the main connector on that pack. Although the voltage on load is about 4 x 3.7V at the battery itself (measured on the balance lead) it may be lower at the ESC due to a high resistance in either the connector on the battery leads, or (less likely) where the thick wires are soldered onto the battery. The latter is less likely given that the battery appears to function correctly on another model. Perhaps the connector is a slightly better fit on the second model?

 

One would expect this to produce some localised heating at the point of the poor connection but perhaps the run time isn't sufficient for this to become obvious.

 

I had an instance a while ago on one pack where one of the connector pins had lost tension and got hot. Hot enough to unsolder the connector in fact. It wasn't obvious until the connector was dismantled as with both the positive and negative connections being made at the same time, the low friction of one of the pair was masked by the other pair (and the plastic housings). These were EC5 connectors but it could happen to other types.

 

If you are able to safely measure the battery voltage (on load) on the ESC side of the connector that would give a clue. If not then perhaps just replace the connector on the battery.

 

Brian.

Link to comment
Share on other sites

Brian, the pack was 100% before the aborted flight, with  resting voltage of 4.2v per cell, model took off great and then reduced the power. We thought it had a failed cell, so did a ground test measuring voltage on the balance lead and found the ESC was cutting the power when the cell voltages were 3.7v per cell at full throttle with no weak cell, so we thought ah ESC problem so we plugged in another battery (same type but fully charged) we got full power with no drop off but cell voltage under load was now 3.3v per cell (and on the inline watt meter around twice the power). 

 

Took the battery that had done the aborted flight, now showing 80% capacity on another model and got full power with an individual cell voltages of 3.4v per cell at max throttle so the battery is good on this ESC, ESC on first plane fine with another battery, but put the two together and LVC cuts in when cell voltage drops to 3.7v per cell under load.

 

Maybe it could be just those two connectors when together, but we saw the same with the inline watt meter, and we couldn't detect any heating in the connector. We didn't have the means to measure the voltage drop across the connectors at the field.

 

BTW it's a 4s system so if we are seeing a 4 x 0.5v, 2v drop across the connector at approx 20 amps, thats 40 watts lost in the connector, as you say enough to melt solder, but the connector didn't get warm.

Edited by Frank Skilbeck
Link to comment
Share on other sites

Hi Frank,

 

Well it does sound as if there is a voltage drop between the battery and the ESC. If the ESC gave four beeps on battery connection it recognised the pack as a 4s pack alright so to go into reduced power mode whilst the battery monitor showed 3.7V per cell it must have been seeing a lot less than this. I can't see where else the voltage could be being lost apart from somewhere down the 'thick' battery wires between the battery and the ESC (and via the wattmeter when it was connected). As the ESC wiring and the wattmeter are common to both good and bad conditions (ie different batteries) it does point to the battery connector, although as you say you would expect the connector to start to warm up.

 

Perhaps what is happening is that (assuming a poor connector on the battery lead), the higher resistance both drops the voltage delivered to the ESC and limits the current so the actual power lost across that connector is not as much as 40W. Also the motor will begin to slow down due to the reducing voltage at the ESC (which will perhaps be going into speed reducing mode as well).

 

The wattmeter should also provide a voltage (and current) indication, what this voltage reads will give a clue to what is happening. If it's significantly lower than the sum of the four cell voltages being shown on the monitor at full throttle then the only place the voltage can be being lost is the battery connector.

 

What type of connectors are being used?

 

Brian.

Link to comment
Share on other sites

Connectors are XT60, all factory fitted.

 

If it was a poor connector on the battery lead then the low voltage would show up on the inline watt meter or on the 2nd model, but it didn't. The only way the connector could drop the voltage is by having a high resistance, and if that's the case it has to dissipate the power lost across the connector, there's no other way.

Link to comment
Share on other sites

43 minutes ago, Frank Skilbeck said:

Connectors are XT60, all factory fitted.

 

If it was a poor connector on the battery lead then the low voltage would show up on the inline watt meter or on the 2nd model, but it didn't. The only way the connector could drop the voltage is by having a high resistance, and if that's the case it has to dissipate the power lost across the connector, there's no other way.


Does your charger measure IR? That is where I would start, I still suspect you may have one or more weak cells and the pack voltage is collapsing under load.

  • Like 1
Link to comment
Share on other sites

Hi Frank,

 

Yes I agree with your latest comments but the fact remains that at the time you witnessed the issue, for the ESC to truly have been going into voltage reducing mode the ESC could only have been 'seeing' 3.3V or even less per cell as you previoiusly suggested. Yet at the other end of the system (on the battery) the voltage was 3.7V per cell. That voltage difference had to have been being dropped somewhere down the line and there is only the battery leads and connector pairs in that circuit.

 

Perhaps slight tolerance variations in the different XT60s are giving slightly different contact pressures. Without further testing and verifying data we won't know but taking a voltage reading (as well as power) from the wattmeter is what I would do next.

 

As Matty B has just suggested, the most likely cause is still the battery pack itself as was suggested a few posts back. I know you assured us that all of the cells had balance voltages to within 0.01V but I find this a little hard to believe if that was truly measured with the motor running as you have since suggested. I'm assuming that whatever battery monitor you have been using displays all four cells simultaneously? If it only displays one at a time (like a lot of those little black rectangular ones do) are the four cell voltages definitely being read correctly? Not the same cell four times? Sorry to question this but someting doesn't add up here.

 

Forgetting for a moment about the application, this is just a simple DC electrical circuit and follows ohms law. The same current is flowing in every part of the circuit so all of the voltages must add up to zero. The current is determined by the overall circuit resistance, and that is the summation of the individual resistances of each element (ESC, connectors, wires and the battery's internal resistance). I know the ESC isn't a resistor but with the capacitors on its input it should act reasonably enough like one.

 

Brian.

  • Like 1
Link to comment
Share on other sites

Hi,

It is perfectly possible for the input to the ESC to be 3.7V on a voltmeter and yet the ESC detect a voltage of less than 3.3V. The voltmeter will only give an average voltage over time and will not show maximum and minimum voltages. The load coming from the motor and electronics is not constant and there are often some very high transient currents and this can cause instantaneous drops in the voltage at the input to the ESC. These dips will be very quick and not show on a voltmeter but if the ESC software happens to be measuring the voltage at that point in time (and we have no idea about the actual software algorithms here), then it will detect an under voltage. 

 

The extent of the voltage dips will be dependant on many things including the condition of the battery and all the wiring and connectors. The input capacitors of the ESC are designed to reduce these transients and so they will also affect the voltage 'seen' by the ESC. It is therefore possible, even likely, that different ESC's will actually trip at different levels in a real time environment and if there is a fault or just difference with the input capacitors, then this difference may be significant. If an ESC that trips at a higher level is coupled with a battery that has high voltage dips, then you may see this problem, even if simple measurements with a voltmeter do not show this. To really know what is going on would require coupling up to a decent high speed oscilloscope.

 

I am not saying that this IS the reason for the strange behaviour, just offerering this as a possible explanation.

 

Might be worth, replacing the connector on the suspect battery with a new one and seeing if this helps?

 

Simon

Link to comment
Share on other sites

7 hours ago, MattyB said:


Does your charger measure IR? That is where I would start, I still suspect you may have one or more weak cells and the pack voltage is collapsing under load.

That's what we initially thought but we checked each cell under load and they all read around 3.7v after the ESC LVC had automatically throttled back, on a different model (i.e. different ESC) the battery gave loads of power and ESC allowed cell voltages to drop to 3.4v under load without any LVC.

 

All the measurements are at the field, but on two other batteries the ESC performs as expected and allows the batteries to go down to 3.3v per cell under load, the "faulty" battery in another model different motor/prop/ESC and at WOT cell voltages drop to 3.4v per cell with no sign of the LVC cutting in and plenty of power, note the battery was physically a bit big for the 2nd model so it wasn't possible to flight test it. The battery was also cold to touch after running at WOT on either model.

 

The individual cell voltages were measured with a battery checker, but we cycled through the cells pretty quickly so didn't see much of a change between the cells, certainly not a failed cell and the underload total pack voltage measured by the inline watt meter was pretty much in line with the individual the cell voltages. Note some battery checkers do an initial measurement and then hold that reading so you can't use them for this sort of check, but the one we used for the tests didn't and gave a live reading.

 

Thoughts are that it is some mismatch on the connector on one particular combination, or there is a short term dip and then the ESC restricts the minimum voltage to 3.7v per cell.

Link to comment
Share on other sites

Hi Frank,

 

As long as the same battery pack and the same model (ESC and motor) combination always produces the same problem then it should be relatively easy to track down the cause as at least it is repeatable. The worst faults are the intermittent and umpredictable ones.

 

It's the cell voltages at full power just before the ESC automatically throttles back that you need to know, plus the voltage as close to the ESC as possible (which is probably on your wattmeter). Whatever it is that is causing the issue will to some extent recover as soon as the ESC throttles back and the current therefore drops. Having to cycle through the cells does mean that you could miss one quickly dropping just before the ESC slows things down, but you would probably catch it over a few attempts.

 

One question; after the ESC has started to cut the motor speed, and presumably at that point you throttle back anyway, what happens if you wait a few seconds and try again? Or a few minutes without recharging the battery?

 

Comparisons with the second model aren't necessarily valid unless it has a pretty much identical electrical setup. Different prop size, motor KV and ESC type and settings between the two would also have an effect.

 

I don't suppose you are anywhere near Bedfordshire are you?

 

Brian.

Link to comment
Share on other sites

1 minute ago, RottenRow said:

 

 

It's the cell voltages at full power just before the ESC automatically throttles back that you need to know, plus the voltage as close to the ESC as possible (which is probably on your wattmeter).

If it also happens with the wattmeter in circuit a battery/ESC connector mismatch isn't the problem, unless there's an unbelievable coincidence.

Link to comment
Share on other sites

  • 5 weeks later...

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...