What The Hell Just Happened?

[Rehab1]

1 hour ago, Chris Westland said:

 

Sorry to hear about this @Rehab1... that just looks seriously painful.  

I personally think Deans Ultra Plug connectors are more reliable than the XT60s or EC3s and they ideally handle a lot more DC current because they have a lot more contiguous copper.  Another choice, perhaps, is Andersen Power Poles which offer current handling up to 350A (standard is 30A).

 

 

Thanks! It no longer hurts but I'm sure some of my kiddos and their parents will be asking all kinds of questions tomorrow. 

Yes I looked at the Deans as well. Great connector! I figured the arming switch I just purchased along with it's 12 awg wires connected to the XT60s would end up being beefier than any wire currently housed within the ACM including the motor wires. I feel comfortable with that application and I  also did not want to change out the other factory connectors. I'm am getting tired of this project!

[esaj]

4 hours ago, Rehab1 said:

I like that idea but being the polarized connectors are temporary how do you connect those leads without damaging the battery lead insulation?

I'm not sure I understand what you mean by "temporary", the idea is to connect the smaller connector with resistor, and then the larger one that's actually meant to deliver the high currents. The smaller connector is left in place even afterwards, it won't do pretty much anything (very small current runs through it while use, as long as the resistor isn't of some very small value and the connections are properly shielded from contact (heatshrink or such), the current running through that part is negligible.

 

2 hours ago, Chris Westland said:

I personally think Deans Ultra Plug connectors are more reliable than the XT60s or EC3s and they ideally handle a lot more DC current because they have a lot more contiguous copper.  Another choice, perhaps, is Andersen Power Poles which offer current handling up to 350A (standard is 30A).

I've had some Anderson Powerpoles, usually the problem is that the very high current versions are also VERY LARGE:

The PP15/45 is what I have here (marked as "Anderson Powerpole housings"), or maybe it's actually the PP75? Either way, the PP120/PP180 -housings are HUGE

 

 

Quote

EC3 should handle continuous 10A draws at 84V (800W ~ 1 HP)(3.5mm bullets) and I wouldn't trust 3mm XT60s for anything more.  Deans Ultra Plugs can probably handle 30A draws; they also have less resistance (according to their website) including the solder joints, than an equivalent length piece of 12 gauge wire.

But all of this is predicated on the quality of a particular manufacturers product; and on the quality of the solder joint produced by the EUC manufacturer (or owner if you are into DIY)  

I've had a few sets of XT60's, Deans (the "basic" T-plug, I don't know if the "Ultra Plug" is different) and EC3's/EC5's (and others) from Aliexpress... The quality certainly varies. Bullet-connectors with "cups" like XT60's/EC's/MT's are easier for me to solder than the "flat" plugs of Deans (they're not that bad either, but probably I should follow the advice someone else gave me once and drill holes to the flat parts for the wires). Just purely by the contact surface-area, I'd go with EC5's, but as they're also the largest, they may cause problems with the space (depending how tight the space where the mainboard / batteries are).

 

Quote

The downfall of any connector can be the solder connection.  Deans Ultra Plugs are hard to solder and have a tendency to drop connection to the V Regulator when twisted or sometimes just touched.  It is (IMHO) harder to get a good soldering junction on the Deans because there is more metal, and thus they require more heat.  With these small connectors you always run the struggle of keeping the heat high enough to get a good solder joint, but low enough not to melt the plastic.  

It helps to connect the other side of the connector while soldering (ie. both the male and female together), and having a "helping hand" alligator clipping the other end of the pin you're soldering, but still it can get tricky with "enough but not too much heat".

 

20 minutes ago, Rehab1 said:

Thanks! It no longer hurts but I'm sure some of my kiddos and their parents will be asking all kinds of questions tomorrow. 

Yes I looked at the Deans as well. Great connector! I figured the arming switch I just purchased along with it's 12 awg wires connected to the XT60s would end up being beefier than any wire currently housed within the ACM including the motor wires. I feel comfortable with that application and I  also did not want to change out the other factory connectors. I'm am getting tired of this project!

Another option you might consider is something like these XT90's with built-in spark arrestors, this is just a quick link to HobbyKing I picked from Google, there may be XT60's also available, they're somewhat smaller in case the space is tight:  https://hobbyking.com/en_us/xt90-s-anti-spark-connector-2pairs-bag.html?___store=en_us

No need to fiddle around with buttons or such.

" IMPORTANT: "These are not cheap copied XT90 plugs. These original Nylon XT90 plugs, manufactured by AMASS, can handle over 90A current for extended periods without exceeding 80DegC thanks to their better contact surface area. The plug is also less likely to deform or melt in comparison to copied non-Nylon XT90 plugs. "

From what I know, Amass is an "actual" higher quality brand name (sold for example by TME), not a clone.

[steve454]

20 hours ago, Rehab1 said:

EMP....Ugh...did my government just perform a test in preparation for bringing down another North Korean missle? 

 

Or maybe it's the other way around?  (the other government was doing a test)

[steve454]

1 hour ago, Rehab1 said:

Thanks! It no longer hurts but I'm sure some of my kiddos and their parents will be asking all kinds of questions tomorrow. 

The first rule of fight club is we don't talk about fight club.

[Carlos E Rodriguez]

35 minutes ago, steve454 said:

The first rule of fight club is we don't talk about fight club.

I was making breakfast in bed for my wife and I burnt my hand. LOL you could try use that.

[steve454]

13 minutes ago, Carlos E Rodriguez said:

I was making breakfast in bed for my wife and I burnt my hand. LOL you could try use that.

Ha Ha, that's a good idea to say that, very positive and makes two people look good at the same time! 

[Rehab1]

11 hours ago, esaj said:

I'm not sure I understand what you mean by "temporary", the idea is to connect the smaller connector with resistor, and then the larger one that's actually meant to deliver the high currents. The smaller connector is left in place even afterwards, it won't do pretty much anything (very small current runs through it while use, as long as the resistor isn't of some very small value and the connections are properly shielded from contact (heatshrink or such), the current running through that part is negligible.

AH...permanent connection! Now I understand! 

I am now convinced that the sequence of how I reconnected the batteries was the cause. On my first attempt both battery packs were connected together with no issues. When I connected the main board to the packs there was an arc. My interpretation at that instant was that my sequence for connecting the batteries to the board was improper when in actuality it was the proper method. If I would have discharged the main board capacitors first before making the final pigtail connection the little arc would never had occurred! 

Please correct me if I wrong but I strongly feel my second attempt when reconnecting the batteries to the main board that resulted in a huge arc was caused when one battery was connected to the board with the capacitors fully charged. When I reconnected the second battery the huge arc resulted when from the huge flow of electrons as both batteries were trying to immediately balance themselves.  

The mini electrical explosion resulted as the male and female connectors came together. The gases from both the arc and the plastic connectors needed to go somewhere. Given the small compressed space between the two connectors and the rapidly expanding gases inside when the arc occurred acted like a mini bomb!  Thoughts? 

 

11 hours ago, esaj said:

Another option you might consider is something like these XT90's with built-in spark arrestors, this is just a quick link to HobbyKing I picked from Google, there may be XT60's also available, they're somewhat smaller in case the space is tight:  https://hobbyking.com/en_us/xt90-s-anti-spark-connector-2pairs-bag.html?___store=en_us

Thanks! I like these! I found some on Amazon (no wonder they are king of internet  commerce!) With all of these connectors coming my way I will soon be able to create a nice connector display board as well!

 

[meepmeepmayer]

1 hour ago, Rehab1 said:

I am now convinced that the sequence of how I reconnected the batteries was the cause. On my first attempt both battery packs were connected together with no issues. When I connected the main board to the packs there was an arc. My interpretation at that instant was that my sequence for connecting the batteries to the board was improper when in actuality it was the proper method. If I would have discharged the main board capacitors first before making the final pigtail connection the little arc would never had occurred! 

Please correct me if I wrong but I strongly feel my second attempt when reconnecting the batteries to the main board that resulted in a huge arc was caused when one battery was connected to the board with the capacitors fully charged. When I reconnected the second battery the huge arc resulted when from the huge flow of electrons as both batteries were trying to immediately balance themselves.  

The mini electrical explosion resulted as the male and female connectors came together. The gases from both the arc and the plastic connectors needed to go somewhere. Given the small compressed space between the two connectors and the rapidly expanding gases inside when the arc occurred acted like a mini bomb!  Thoughts?

I can't judge if that is what happened, but it should not be possible to have this happen when connecting anything in any order. Well, it happened anyways. Thank you a lot for doing the (involuntary) experiment and exploring what one should and should not do, one (explosive) step at a time Seems like catastrophic failure (like a cut out, or this) is the way how we learn about EUCs. Too bad there are no less harmful middle steps. I'll certainly look very hard how I connect my batteries from now on! First the two small inter-battery connectors, then the fat cable, then everything to the board.

Have you paid more for the ACM itself, or all the tools,new motor, parts etc already? You must be getting close to parity

[esaj]

2 hours ago, Rehab1 said:

AH...permanent connection! Now I understand! 

Yeah, at least that way I would use them... having an open connector dangling inside the case connected to live batteries and mainboard wouldn't seem like a good idea to me    Apparently on the one with button, you first leave out the "jumper" at the end of the Anderson-connectors, press the button to charge the mainboard capacitors and then put in the jumper:

Connect the ESC, remove arming jumper, then connect your battery.

Simply hold down push button for 2-3 sec to charge the capacitors with limited current.  This will eliminate the nasty arching when the arming plug is plugged into the Arming Switch.

 

 

Quote

I am now convinced that the sequence of how I reconnected the batteries was the cause. On my first attempt both battery packs were connected together with no issues. When I connected the main board to the packs there was an arc. My interpretation at that instant was that my sequence for connecting the batteries to the board was improper when in actuality it was the proper method. If I would have discharged the main board capacitors first before making the final pigtail connection the little arc would never had occurred! 

Close, but the other way around    There should be no sparks if the mainboard capacitors are already charged. If they are discharged, large current will flow to charge the capacitors, and heat up the connector in-between (unless the current is restricted by a larger resistor, like in the case of those no-spark -connectors). Couldn't find any actually clear water analogy pictures of this right now, so these will have to do:

Many basic principles of electricity can be explained through "water analogies", although they do not work in all situations. Usually in the analogies, voltage equals pressure (typically depicted to be caused by difference in the water level / height), current equals water flow and resistance is depicted as either pipes (smaller pipe = more resistance) or valves to restrict the amount of flow (current). Voltage sources like batteries are either shown as big reservoirs (like the water tank above), or more correctly, as pumps pumping the water from the lowest potential (bottom of the water-pipe -system) back to top. In the picture, the beaker is the capacitor (or multiple capacitors in parallel). If the valve is fully open (small resistance) between the water tank and the beaker in the above picture, the beaker will fill up fast. When it's more closed (higher resistance) the beaker will fill up slower. If it's fully closed it's like there's no connection (connectors are severed) and no water will float at all.

If the beaker's water-level (voltage) is already at the same level as the water tank (battery) when the valve is opened, no water will flow (no current), and then also no sparks should fly. The idea of the no-spark connectors is to first add a much smaller pipe (higher resistance) in parallel with the "main"-valve to fill the beaker in a slower manner. When the main-valve is opened to allow the big flows, there will be no high currents, because the beaker's already filled.

 

A straight-forward simulation of the two cases shows vastly different power dissipations:

The upper graphs show the power dissipation at the resistance (connector or resistor) in red, the lower graphs show the current flowing through it (in green, right hand-scale) and the voltage of the capacitor (in blue, left-hand scale).

On the left, 80V battery is connector across a 5 milliohm ("a connector not yet fully connected") resistance. The discharged capacitor is charged at a current peaking closer to 900A, and the connector power dissipation spikes closer to 4kW (although only for a small fraction of a second, about 30µs = 0.03ms = 0.00003 seconds), which likely could cause at least some sparking.

On the right, the capacitor is charged over a resistor of 330 ohms (I just used that value because it was in your video ). The capacitor is now charged by current starting from around 250mA (0.25A), and the current drops fast as the capacitor charges up (less voltage difference over the resistor). It takes about 1/3th of a second (330ms or thereabouts) for the capacitor to charge, about 10000 times longer than with the 5 milliohm connection resistance. There's a short-lived power dissipation spike, but it's of much smaller power (20W at peak). Not sure how well a 0.25W resistor would take that, although they're supposed to be able to handle much higher power dissipations in short spikes vs the continuous 0.25W max rating.

 

Quote

Please correct me if I wrong but I strongly feel my second attempt when reconnecting the batteries to the main board that resulted in a huge arc was caused when one battery was connected to the board with the capacitors fully charged. When I reconnected the second battery the huge arc resulted when from the huge flow of electrons as both batteries were trying to immediately balance themselves.  

I'm still not sure what caused the voltage difference between the packs on the second attempt. But first connecting the batteries in parallel and then connecting to the mainboard would seem "better" to me...

 

Quote

The mini electrical explosion resulted as the male and female connectors came together. The gases from both the arc and the plastic connectors needed to go somewhere. Given the small compressed space between the two connectors and the rapidly expanding gases inside when the arc occurred acted like a mini bomb!  Thoughts? 

Could be, I've had the connectors spark many times, but never as violently as in your case. You could use anti-spark connectors between the packs also, but do note that they're meant for short-lived charging spikes, the resistor-size may not be picked with longer high balancing current flow in mind. That's to say, even with a anti-spark connector, avoid connecting the batteries in parallel if there's a large voltage difference between them. But in the last measurement you showed, they were pretty much spot-on at the same voltage (which just confuses me more as to why the connector then sparked so violently when they were connected  ).

 

Quote

 

Thanks! I like these! I found some on Amazon (no wonder they are king of internet  commerce!) With all of these connectors coming my way I will soon be able to create a nice connector display board as well!

The simplicity and no need for extra wires/buttons etc. on the "pre-made" connectors with built-in resistors seemed better to me, the "downside" is that they're not "plug'n'play", as you first need to solder them yourself (but you're going to have to change the fried connectors anyway ). If you feel unsure of doing that yourself, probably some RC-shop or such can solder the connectors for a small fee? At least I remember one local shop offering such service.

[meepmeepmayer]

Could you charge the board capacitors by turning the tire when the battery is disconnected? Any other way to charge them? I boot up the wheel after disconnecting the battery so the residual electricity is drained and I can touch the board (even just accidentally) without getting shocked, but I guess it may be a good idea to recharge the capacitors before reconnecting the battery so there's no spark at all.

[esaj]

15 minutes ago, meepmeepmayer said:

Could you charge the board capacitors by turning the tire when the battery is disconnected?

In theory, yeah, but you might need to spin the tire very fast to induce enough voltage to charge the caps all the way to the full battery level. Probably as fast as the "no-load speed" of the motor (ie. the speed where it cutouts when lifted), which usually is much higher than the maximum speed you can actually ride?

 

Quote

Any other way to charge them? I boot up the wheel after disconnecting the battery so the residual electricity is drained and I can touch the board (even just accidentally) without getting shocked, but I guess it may be a good idea to recharge the capacitors before reconnecting the battery so there's no spark at all.

Using the batteries themselves with a resistor to limit the current, as depicted above, is fairly straightforward, albeit a bit fiddly (without pre-made connectors). If you have a power supply that can go as high as the batteries, you could use that, either in constant current -mode to limit the charging current, ramping up the voltage slowly or with a resistor between the PSU output and the mainboard power-connector to limit the current. Still, the anti-spark -connectors are probably the easiest way to go in the end.

 

[meepmeepmayer]

Thank you! The ACM motor switches off at 65 km/h for safety, so I don't think the tire thing will work. What happens to the produced electricity from slower turning, where does it go? Heat?

You could make a high resistance middle piece that you can plug in between battery and board for this, right? Maybe an idea to make/have such a piece. No need to modify the wheel then.

[esaj]

25 minutes ago, meepmeepmayer said:

Thank you! The ACM motor switches off at 65 km/h for safety, so I don't think the tire thing will work. What happens to the produced electricity from slower turning, where does it go? Heat?

My best guess is that it turns into heat in the internal resistances of the motor/wiring/etc. I wrote a bit about the more finer details how the motor charges a cap here:

 

 

Quote

You could make a high resistance middle piece that you can plug in between battery and board for this, right? Maybe an idea to make/have such a piece. No need to modify the wheel then.

That's pretty much what the anti-spark connector is, although there are numerous different models available (or you can make one yourself, probably will make one for my own use):

The last picture shows a XT90 where the resistor is built into the connector itself. You see a longer metal-sleeve on the right side female connector hole, there's no cut-through image of it, but I'd imagine it being something like this:

Yeah, I can't draw...   The idea is that the male-pin will first contact the red part, at which point the connection is made over the internal resistor "R", so it's charging the caps while you're pushing the connectors together. With "suitably small" value (the connector side says 56 ohms), the caps will have already charged with the lower current to the full voltage by the time the pin is deep enough to make the "full" contact with the bluish part.

 

 

[Carlos E Rodriguez]

1 hour ago, esaj said:

Yeah, at least that way I would use them... having an open connector dangling inside the case connected to live batteries and mainboard wouldn't seem like a good idea to me    Apparently on the one with button, you first leave out the "jumper" at the end of the Anderson-connectors, press the button to charge the mainboard capacitors and then put in the jumper:

Connect the ESC, remove arming jumper, then connect your battery.

Simply hold down push button for 2-3 sec to charge the capacitors with limited current.  This will eliminate the nasty arching when the arming plug is plugged into the Arming Switch.

 

 

Close, but the other way around    There should be no sparks if the mainboard capacitors are already charged. If they are discharged, large current will flow to charge the capacitors, and heat up the connector in-between (unless the current is restricted by a larger resistor, like in the case of those no-spark -connectors). Couldn't find any actually clear water analogy pictures of this right now, so these will have to do:

Many basic principles of electricity can be explained through "water analogies", although they do not work in all situations. Usually in the analogies, voltage equals pressure (typically depicted to be caused by difference in the water level / height), current equals water flow and resistance is depicted as either pipes (smaller pipe = more resistance) or valves to restrict the amount of flow (current). Voltage sources like batteries are either shown as big reservoirs (like the water tank above), or more correctly, as pumps pumping the water from the lowest potential (bottom of the water-pipe -system) back to top. In the picture, the beaker is the capacitor (or multiple capacitors in parallel). If the valve is fully open (small resistance) between the water tank and the beaker in the above picture, the beaker will fill up fast. When it's more closed (higher resistance) the beaker will fill up slower. If it's fully closed it's like there's no connection (connectors are severed) and no water will float at all.

If the beaker's water-level (voltage) is already at the same level as the water tank (battery) when the valve is opened, no water will flow (no current), and then also no sparks should fly. The idea of the no-spark connectors is to first add a much smaller pipe (higher resistance) in parallel with the "main"-valve to fill the beaker in a slower manner. When the main-valve is opened to allow the big flows, there will be no high currents, because the beaker's already filled.

 

A straight-forward simulation of the two cases shows vastly different power dissipations:

The upper graphs show the power dissipation at the resistance (connector or resistor) in red, the lower graphs show the current flowing through it (in green, right hand-scale) and the voltage of the capacitor (in blue, left-hand scale).

On the left, 80V battery is connector across a 5 milliohm ("a connector not yet fully connected") resistance. The discharged capacitor is charged at a current peaking closer to 900A, and the connector power dissipation spikes closer to 4kW (although only for a small fraction of a second, about 30µs = 0.03ms = 0.00003 seconds), which likely could cause at least some sparking.

On the right, the capacitor is charged over a resistor of 330 ohms (I just used that value because it was in your video ). The capacitor is now charged by current starting from around 250mA (0.25A), and the current drops fast as the capacitor charges up (less voltage difference over the resistor). It takes about 1/3th of a second (330ms or thereabouts) for the capacitor to charge, about 10000 times longer than with the 5 milliohm connection resistance. There's a short-lived power dissipation spike, but it's of much smaller power (20W at peak). Not sure how well a 0.25W resistor would take that, although they're supposed to be able to handle much higher power dissipations in short spikes vs the continuous 0.25W max rating.

 

I'm still not sure what caused the voltage difference between the packs on the second attempt. But first connecting the batteries in parallel and then connecting to the mainboard would seem "better" to me...

 

Could be, I've had the connectors spark many times, but never as violently as in your case. You could use anti-spark connectors between the packs also, but do note that they're meant for short-lived charging spikes, the resistor-size may not be picked with longer high balancing current flow in mind. That's to say, even with a anti-spark connector, avoid connecting the batteries in parallel if there's a large voltage difference between them. But in the last measurement you showed, they were pretty much spot-on at the same voltage (which just confuses me more as to why the connector then sparked so violently when they were connected  ).

 

The simplicity and no need for extra wires/buttons etc. on the "pre-made" connectors with built-in resistors seemed better to me, the "downside" is that they're not "plug'n'play", as you first need to solder them yourself (but you're going to have to change the fried connectors anyway ). If you feel unsure of doing that yourself, probably some RC-shop or such can solder the connectors for a small fee? At least I remember one local shop offering such service.

 

 Its interesting and puzzling that the battery to battery interface step caused this event. I can only think that assuming the batteries had enough delta that when the connectors reach the near engagement enough ionization occurred during the normal pop-spark to complete a bridge between the positive and negative pins. It is plausible because the connectors form a closed chamber allowing the ionized air to stay concentrated. @Rehab1indicated that his connector engagement was not deliverate and that would allow for a prolonged mini arc to cascade into a full short plasma and then arc flash of molten and evaporated copper and residues. Copper expands to 67,000 times the solid volume. 

 Also during battery to battery engagement, they are not sharing the common ground plane. Not sure if that would have contributed for a higher voltage potential. Granted it does not have an earth ground but its not at the same potential either in addition to the internal charge of the battery.

 

[esaj]

15 minutes ago, Carlos E Rodriguez said:

 

 Its interesting and puzzling that the battery to battery interface step caused this event. I can only think that assuming the batteries had enough delta that when the connectors reach the near engagement enough ionization occurred during the normal pop-spark to complete a bridge between the positive and negative pins. It is plausible because the connectors form a closed chamber allowing the ionized air to stay concentrated. @Rehab1indicated that his connector engagement was not deliverate and that would allow for a prolonged mini arc to cascade into a full short plasma and then arc flash of molten and evaporated copper and residues. Copper expands to 67,000 times the solid volume. 

Yeah, could be, I don't know enough about plasma and the related physics.

 

15 minutes ago, Carlos E Rodriguez said:

 Also during battery to battery engagement, they are not sharing the common ground plane. Not sure if that would have contributed for a higher voltage potential. Granted it does not have an earth ground but its not at the same potential either in addition to the internal charge of the battery.

AFAIK, they are sharing a common ground through the negative terminals connected together when placed in parallel. Sure, high current does cause some (negligible in relation to the total voltages) voltage drop over the "ground" connection, parasitic inductances and capacitances can cause some minor effects etc. but it takes a lot of guesswork to figure out any meaningful values for them, and in the end, they probably have very little if any effect in this kind of circuit.

"Common ground" or 0V is more a matter of an "agreement" anyway, consider:

Two 15V batteries in series, a "common ground" is taken from between them, that way you have +15V at the positive terminal of the upper battery and -15V at the negative terminal of the bottom battery (both in relation to the GND between them). BUT, if you think of the -15V terminal as 0V / "common ground", you have +15V at the middle and +30V at the top...

 

[Rehab1]

4 hours ago, meepmeepmayer said:

Thank you a lot for doing the (involuntary) experiment and exploring what one should and should not do, one (explosive) step at a time

A Neil Armstrong quote...well almost!

 

4 hours ago, meepmeepmayer said:

I'll certainly look very hard how I connect my batteries from now on! First the two small inter-battery connectors, then the fat cable, then everything to the board.

I think @esaj and @Carlos E Rodriguez might agree that taking the extra precaution and discharging the capacitors before making the final connection to the main board might be prudent to avoid

 

3 hours ago, esaj said:

Yeah, at least that way I would use them... having an open connector dangling inside the case connected to live batteries and mainboard wouldn't seem like a good idea to me    Apparently on the one with button, you first leave out the "jumper" at the end of the Anderson-connectors, press the button to charge the mainboard capacitors and then put in the jumper:

Connect the ESC, remove arming jumper, then connect your battery.

Simply hold down push button for 2-3 sec to charge the capacitors with limited current.  This will eliminate the nasty arching when the arming plug is plugged into the Arming Switch.

Very good! The arming switch appears to be well designed. If the arming switch is too large I may just use the spark arrestor connectors instead.

 

3 hours ago, esaj said:

Close, but the other way around    There should be no sparks if the mainboard capacitors are already charged. If they are discharged, large current will flow to charge the capacitors, and heat up the connector in-between (unless the current is restricted by a larger resistor, like in the case of those no-spark -connectors). Couldn't find any actually clear water analogy pictures of this right now, so these will have to do:

Darn...I thought I was correct!Your logic and the awesome diagrams make perfect sense! Thanks for taking the time to explain! Visuals always help!

3 hours ago, esaj said:

Many basic principles of electricity can be explained through "water analogies", although they

The analogy works in this scenario!

3 hours ago, esaj said:

Could be, I've had the connectors spark many times, but never as violently as in your case. You could use anti-spark connectors between the packs also, but do note that they're meant for short-lived charging spikes, the resistor-size may not be picked with longer high balancing current flow in mind. That's to say, even with a anti-spark connector, avoid connecting the batteries in parallel if there's a large voltage difference between them. But in the last measurement you showed, they were pretty much spot-on at the same voltage (which just confuses me more as to why the connector then sparked so violently when they were connected  ).

As much as I like science and unless someone else would like to try to recreate the scenario I will connect the battery packs first and then the main board with the spark arrestor.

[Rehab1]

2 hours ago, Carlos E Rodriguez said:

It is plausible because the connectors form a closed chamber allowing the ionized air to stay concentrated. @Rehab1indicated that his connector engagement was not deliverate and that would allow for a prolonged mini arc to cascade into a full short plasma and then arc flash of molten and evaporated copper and residues. Copper expands to 67,000 times the solid volume. 

Bingo! Closed chamber is what I was alluding to. Thanks! It brings back bad memories when I had a young, immature non- integrated cortex!   I built a few pipe bombs in the day (stupid!) .....40 years ago....and the closed chamber concept...well you get the idea!

[Rehab1]

On 5/1/2017 at 7:49 AM, meepmeepmayer said:

First the two small inter-battery connectors, then the fat cable, then everything to the board.

Perfect! I would still discharge the capacitors before the final connection to the board. I think @esaj would agree. Just use a 60 w light bulb with 2 leads attached. I was able to discharge them with a small LED light but I was surprise it did not blow.

[SuperSport]

On 5/1/2017 at 9:26 AM, Rehab1 said:

It brings back bad memories when I had a young, immature non- integrated cortex!   I built a few pipe bombs in the day (stupid!)

My brothers and I also did that when we were young.  I can't believe we all still have our digits, arms, legs, heads, etc...

[Rehab1]

My new XT60 connectors arrived and the male and female ends are now soldered in place. Still waiting on the spark arrestor connectors.

 

[esaj]

4 minutes ago, Rehab1 said:

Perfect! I would still discharge the capacitors before the final connection to the board. I think @esaj would agree. Just use a 60 w light bulb with 2 leads attached. I was able to discharge them with a small LED light but I was surprise it did not blow.

Still the other way around... there should be no sparks if the capacitors are charged to the same voltage as the battery packs, that's what the anti-spark connectors do, they charge the capacitors over a higher resistance before the actual low-resistance connection is made.   The discharging was made to make sure that your multimeter wouldn't get damaged when measuring the resistance.

[esaj]

20 minutes ago, Rehab1 said:

My new XT60 connectors arrived and the male and female end are now soldered in place. Still waiting on the spark arrestor connectors.

 

Nice work, I have a tendency to forget the heat shrinks far too often, and then have to desolder the wire and start over     Personally, I use a bit heavier gauge (as in, larger) soldering tip for the bigger connectors, and usually apply heat much longer... but the main thing is you get good connections. In case you sometimes work with "deeper" connectors (ie. where the "cup" the wire goes into is deeper), this trick works well ("preform/pre-filling"):

 

[Rehab1]

20 minutes ago, esaj said:

Still the other way around... there should be no sparks if the capacitors are charged to the same voltage as the battery packs, that's what the anti-spark connectors do, they charge the capacitors over a higher resistance before the actual low-resistance connection is made.   The discharging was made to make sure that your multimeter wouldn't get damaged when measuring the resistance.

I'm dead weight!! GOT IT!!

[Rehab1]

4 minutes ago, esaj said:

I use a bit heavier gauge (as in, larger) soldering tip for the bigger connectors,

Good idea! I thought about switching over to a larger tip but I did not want to burn my fingers in the process!

So my feeling is if my batteries are still at 84V there is no need to charge them at the moment. I just wish my new motor would get here. The Chinese have another Holiday!

[esaj]

On 5/2/2017 at 11:19 PM, Rehab1 said:

Good idea! I thought about switching over to a larger tip but I did not want to burn my fingers in the process!

A bit different kind of soldering... Just finished soldering two of these:

The picture's taken before scrubbing off all the excess flux   The wire's 16mm² silicone cable, about 5 AWG, the PCB's got 105µm copper-coating (3 times thicker than the usual 35µm / 1oz). The combination sucks in quite a lot of heat and it took quite some time, but 400C, extra flux and the largest conical tip (4C in the 900M-series) I have did get the job done...

In this case it also helped that I spread the conductors inside the cable on the board (need large surface area anyway, this thing's meant to deliver those nice kiloampere-level spikes at best...), so didn't need to heat up the entire cable in one go. No burned fingers, although I did touch the board accidentally for a split second while soldering... HOT!  

The 4C is a large conical tip:

Although I wouldn't go that large for the "normal" connectors, usually I use a flat 1.6-2.4-3.2D there, depending on the connector size, sometimes maybe one with a round and larger head, B in the below picture I think... best to judge by looking at the connector size and the shape of the surfaces you need to solder, but the "flat screwdriver head"-types (D's in 900M's) are usually my go-to -tips in almost any kind of work. But I can't say that I'm anything like a "professional" solderer

The picture's missing the 0.8D (0.8mm wide flat-blade), which I use almost always when doing SMDs. Some very closely spaced (low pin-pitch) IC's and such may require going to the "needle"-like tips ("I", actually, I don't have an IS-tip, probably should order a couple ). I don't know if your iron uses the 900M-series -tips, but probably about the same sizes/shapes are available in other series too.

Here's a better picture of the different tips, sizes and shapes:

Don't know what the temperatures mean..? Maybe that the diameter etc. measurements are taken with the tip at that temperature, as they may (very slightly) expand when heated to several hundred degrees...