The Heybridge Railway, 1889 to 1913
- Thread starter Richard Gawler
- Start date
simond
Western Thunderer
No. If they catch fire, 100% immerse them in water. It won’t get worse, there is no metallic lithium in a li-po or li-ion battery.
Preferably deep water, but coffee, orange juice, whatever, anything very wet is better than nothing.
This is particularly true if you’re on a plane and your phone starts smoking. (Usually as a result of it being crushed in the seat recliner mech)
simond
Western Thunderer
Coincidentally, this morning
www.theguardian.com
Do electric cars pose a greater fire risk than petrol or diesel vehicles?
The first in a series exploring the myths and realities surrounding EVs
www.theguardian.com
Richard Gawler
Western Thunderer
I get the impression, most battery fires start during charging. If the charging process relies on control connections between charger and battery to provide a degree of functional safety, and most electrical faults still involve connector failures, then building the charger into the battery sounds good.
This sounds like a memorable flight.
This sounds like a memorable flight.
Last edited:
simond
Western Thunderer
Happily I wasn’t on a flight where it happened, though I’ve had a few “exciting” moments over the years. It is part of cabin crew training.
Battery fires are pretty much always caused by damage, either the gross physical kind of a car accident or a powered chair crushing & rupturing the cell casing, leading to a short circuit and uncontrolled discharge, or a rather more subtle internal damage issue.
Poor charging management (overcurrent, and/or over-, or under- temperature charging) can lead to dendritic growth of lithium “fingers” from the surface of one electrode. In normal operation this does not happen, the lithium ions normally ”intercalate” within the spongy structure of the electrodes, but if they start to plate the outside, and grow outwards, they can and will penetrate the polymer insulating layer between anode & cathode, and will eventually short the cell internally.
This then causes a massive current (the internal DC resistance is of the same order as a lead acid wet cell) and lots of heat, which leads to dissociation of the materials (most of which are flammable) from which the cell is made. The heat sets off adjacent cells, and so it goes on. There’s still a lot of stored energy in a discharged lithium battery, much more than the bit of vapour, and the steel or plastic from which an equivalent petrol tank might be made.
Hard-wired charge management is normal in the auto industry, eBikes, and mobile phones, etc., the key parameters are the voltage, from which the state of charge is inferred, and the cell temperature, and the charging current can be modulated based on these parameters. But it takes up space, and costs money, so lower end products may simply supply a low current and not bother to monitor cell temperature.
They are the ones to charge in a biscuit tin on the patio, whilst you’re watching!
Submersion doesn’t stop the chemical reactions, but it cools the assembly, hopefully prevents thermal runaway & ignition of adjacent cells, and stops the fire spreading to other flammable materials nearby.
Battery fires are pretty much always caused by damage, either the gross physical kind of a car accident or a powered chair crushing & rupturing the cell casing, leading to a short circuit and uncontrolled discharge, or a rather more subtle internal damage issue.
Poor charging management (overcurrent, and/or over-, or under- temperature charging) can lead to dendritic growth of lithium “fingers” from the surface of one electrode. In normal operation this does not happen, the lithium ions normally ”intercalate” within the spongy structure of the electrodes, but if they start to plate the outside, and grow outwards, they can and will penetrate the polymer insulating layer between anode & cathode, and will eventually short the cell internally.
This then causes a massive current (the internal DC resistance is of the same order as a lead acid wet cell) and lots of heat, which leads to dissociation of the materials (most of which are flammable) from which the cell is made. The heat sets off adjacent cells, and so it goes on. There’s still a lot of stored energy in a discharged lithium battery, much more than the bit of vapour, and the steel or plastic from which an equivalent petrol tank might be made.
Hard-wired charge management is normal in the auto industry, eBikes, and mobile phones, etc., the key parameters are the voltage, from which the state of charge is inferred, and the cell temperature, and the charging current can be modulated based on these parameters. But it takes up space, and costs money, so lower end products may simply supply a low current and not bother to monitor cell temperature.
They are the ones to charge in a biscuit tin on the patio, whilst you’re watching!
Submersion doesn’t stop the chemical reactions, but it cools the assembly, hopefully prevents thermal runaway & ignition of adjacent cells, and stops the fire spreading to other flammable materials nearby.
Richard Gawler
Western Thunderer
Okay - enough of the engineering and scientific stuff, time to move on to the voodoo 
I have given the loco a quick trial with the motor wires coiled over the receiver, the aerial inside the boiler and smokebox, and (removed for the photo) the saddle tank placed over the top. Needless to say everything works perfectly.
The aerial is the black wire at the end of the pink arrow.
I see three external locations for the aerial. I favour (A), along the underside of the tank, but possibly (B) underneath between the frames. (C) would be vertically in front of the firebox, but a bit close to the motor brushes.
Or quite possibly, the aerial will work best sticking out from the receiver in a straight line but encased in brass model, i.e. leave it exactly where it is. Short of nailing a dead chicken to the ceiling, I wonder which will be best?
I have given the loco a quick trial with the motor wires coiled over the receiver, the aerial inside the boiler and smokebox, and (removed for the photo) the saddle tank placed over the top. Needless to say everything works perfectly.
The aerial is the black wire at the end of the pink arrow.
I see three external locations for the aerial. I favour (A), along the underside of the tank, but possibly (B) underneath between the frames. (C) would be vertically in front of the firebox, but a bit close to the motor brushes.
Or quite possibly, the aerial will work best sticking out from the receiver in a straight line but encased in brass model, i.e. leave it exactly where it is. Short of nailing a dead chicken to the ceiling, I wonder which will be best?
Richard Gawler
Western Thunderer
I think the chimney would be ideal if I dismantle it and put some plastic tube in place of the brass bar. The trouble is I am rather proud of my effort here!
I think I will go for (A) because it gives an option to lay the aerial on top of the running plate above the cylinder where I could hide it with some plastic tools or whatever. It is also the easiest to build.
Still, it would be sensible to leave the other wires long enough to let me put the circuit board well forward, under the foot of the chimney.
I think I will go for (A) because it gives an option to lay the aerial on top of the running plate above the cylinder where I could hide it with some plastic tools or whatever. It is also the easiest to build.
Still, it would be sensible to leave the other wires long enough to let me put the circuit board well forward, under the foot of the chimney.
Last edited:
David Waite
Western Thunderer
Could you disguise the aerial as a vacuum hose on the buffer beam.
David
David
Richard Gawler
Western Thunderer
Well, the aerial wire is 30 mm long and the instructions explain, it should be straight and not shortened or lengthened. So it is the perfect length to make a brake standard and hose, but this would mean folding it back on itself.
I think the chimney, which is 28 mm long, could become the aerial. Unfortunately it is now a fixed part of the model, I fixed it to discourage it from developing a lean.
I think the chimney, which is 28 mm long, could become the aerial. Unfortunately it is now a fixed part of the model, I fixed it to discourage it from developing a lean.
Last edited:
Richard Gawler
Western Thunderer
Perfect! I am going for option (B) aerial between the cylinders. If this doesn't give enough range, look out for a shunter's rope nearby 
Last edited:
Richard Gawler
Western Thunderer
The batteries finally arrived yesterday. This ought to teach me a lesson about writing here on impulse straight after clicking "buy it now".
They are beautifully finished (is this the right way to describe a battery?) and presented in a moulded box arranged to hold the two side by side. They came with a glossy business card and this gives contact details and a list of the Poover range of batteries. And to top it off, a Digital Battery Series User Manual, two sides of A4 presented in English. The PP3 buying experience has changed.
The paperwork is clearly trying very hard to project these batteries as a safe product. I really do like the heading Matters needing attention where we learn about the need to keep the batteries away from children, fire and water; and what to do if you touch the liquid inside the battery.
Indeed the User Manual is reassuringly candid in places
simond
Western Thunderer
UN38.3 is indeed the transportation testing of lithium batteries. In 2013 or thereabouts, I set up a lab to do this in Poland. This is not it, but gives you a flavour. UN 38.3 Certification for Lithium Batteries. It must be compliant in order to be accepted for any form of transport, but it does not require third party testing, self declaration is acceptable.
MSDS is not a specification, it’s the abbreviation for “Materials Safety Data Sheet”, which suppliers are obliged to supply with “materials”, typically chemicals, rather than products like batteries, or cars. Pointless in this context.
ROHS (pedantically RoHS) is the EU Restrictions on Hazardous Substances in electrical goods, presumably still applicable in the UK post Brexit. It must be in order to be sold, but of course, it is again, a self-declared “pass”.
UN38.3 should give you some confidence that the product will not burst into flames just because you looked at it, but it does not confirm that it cannot be used unsafely, so caution in charging & discharging is the order of the day.
It’s probably absolutely fine, but the stakes are, shall we say, discouraging.
MSDS is not a specification, it’s the abbreviation for “Materials Safety Data Sheet”, which suppliers are obliged to supply with “materials”, typically chemicals, rather than products like batteries, or cars. Pointless in this context.
ROHS (pedantically RoHS) is the EU Restrictions on Hazardous Substances in electrical goods, presumably still applicable in the UK post Brexit. It must be in order to be sold, but of course, it is again, a self-declared “pass”.
UN38.3 should give you some confidence that the product will not burst into flames just because you looked at it, but it does not confirm that it cannot be used unsafely, so caution in charging & discharging is the order of the day.
It’s probably absolutely fine, but the stakes are, shall we say, discouraging.
Richard Gawler
Western Thunderer
When I was working with medical devices around 2012 to 2015 and demonstrating compliance with EN60601, I found that none of the manufacturers of lithium batteries obtained the certification mandated for a medical application of their batteries. I've forgotten the reference of this spec, but no-one atttempted it and we had to simply ask the certifying authority to accept an approval by UL.
Anyway - I am getting a feel for the capacity of the Poover PP3 batteries:
1) Charging time from as-supplied condition was about 210 minutes for one and 240 minutes for the other. (They have red/green LEDs to show their charging/charged state)
2) Today I placed the chassis of my Manning Wardle upside down and ran it from one of the batteries until it stopped. Run time was more than 210 minutes but less than 230 minutes. The motor draws about 130 mA driving the unloaded mechanism so the capacity obtained is about 130 x 3.5 = 450 mAh.
3) The measured capacity of 450 mAh is smaller than the figures on the battery itself (1,200 mAh), in the eBay listing (1,000 mAh) and in the User Manual (680 mAh) but this doesn't really trouble me. In some ways, it is more reassuring than the larger figures. It is two or three times better than a budget primary battery and it places endurance firmly in the realm of a full day of shunting at a show.
4) Battery terminal voltage measures either 8.98 volts or nothing.
I'm not overly concerned, but innocence is bliss; I don't know what failure modes are possible. For example, why a biscuit tin and not a plastic bag or indeed nothing at all? This is a PP3 battery after all, not something sized to power a bicycle or a micro scooter.
Anyway - I am getting a feel for the capacity of the Poover PP3 batteries:
1) Charging time from as-supplied condition was about 210 minutes for one and 240 minutes for the other. (They have red/green LEDs to show their charging/charged state)
2) Today I placed the chassis of my Manning Wardle upside down and ran it from one of the batteries until it stopped. Run time was more than 210 minutes but less than 230 minutes. The motor draws about 130 mA driving the unloaded mechanism so the capacity obtained is about 130 x 3.5 = 450 mAh.
3) The measured capacity of 450 mAh is smaller than the figures on the battery itself (1,200 mAh), in the eBay listing (1,000 mAh) and in the User Manual (680 mAh) but this doesn't really trouble me. In some ways, it is more reassuring than the larger figures. It is two or three times better than a budget primary battery and it places endurance firmly in the realm of a full day of shunting at a show.
4) Battery terminal voltage measures either 8.98 volts or nothing.
I'm not overly concerned, but innocence is bliss; I don't know what failure modes are possible. For example, why a biscuit tin and not a plastic bag or indeed nothing at all? This is a PP3 battery after all, not something sized to power a bicycle or a micro scooter.
Last edited:
Richard Gawler
Western Thunderer
I can think about what is normal operation, and what is foreseeable misuse. The obvious misuse is a dead short across the terminals and, in a compact model locomotive where there isn't room for a mating snap connector, I need to insert some sort of mitigation e.g. insulation between terminals and brass body, and a resettable fuse in one of the external leads.
With the battery installed inside a scale model, many methods of potential misuse, such as crushing, immersion in water or placing in a fire, are surely too improbable to pursue.
Richard Gawler
Western Thunderer
I have noticed, the quantity of 'likes' has reduced since I started writing about power and control! So for the record, here is one more post to explain my custom configuration settings for the Micron MR601a receiver.
Motor start power 12% ( menu 1, 1, 4, 1, 2 )
With the 12 volt Mashima motor being driven from a 9 volt supply, the default value of zero starting power means small throttle settings don't make the motor do anything. This setting lets control of motor speed extend over the whole movement of the throttle.
PWM frequency 120 Hz ( menu 1, 1, 7, 8 )
Choice of pulse width modulation frequency depends on motor characteristics. The default setting of 1 kHz makes my Mashima motor produce a constant and prominent whine. Increasing the frequency to 8 kHz silences the whine but greatly reduces motor torque. So I have settled for 120 Hz, which is the lowest frequency available and the closest to the frequency produced by a conventional analogue controller running from 50 Hz AC power.
Motor soft start / stop 0.25 s acceleration, 0.25 s deceleration ( menu 1, 1, 8, 2, 2 )
Putting a brief delay into the throttle response will, at least according to the instructions for programming, reduce motor heating using the low PWM frequency. I can look for the effect of this when I get the model on a roundy-roundy test track, but I am not expecting motor heating to happen on a short layout at home.
I am not enjoying this part of the build very much. I want to have radio control if I possibly can, but the setting-up gets stressful when "nothing happens", especially when motor control seems to have been lost. So far, judicuous power-cycling has resolved these losses. Configuration involves five main menus each with up to five sub-levels, with navigation using two push buttons and one LED. The scheme is easier to work than it sounds, but Micron have mentioned a graphics-based programming tool on their web site, and I think this will make my life more relaxing.
Motor start power 12% ( menu 1, 1, 4, 1, 2 )
With the 12 volt Mashima motor being driven from a 9 volt supply, the default value of zero starting power means small throttle settings don't make the motor do anything. This setting lets control of motor speed extend over the whole movement of the throttle.
PWM frequency 120 Hz ( menu 1, 1, 7, 8 )
Choice of pulse width modulation frequency depends on motor characteristics. The default setting of 1 kHz makes my Mashima motor produce a constant and prominent whine. Increasing the frequency to 8 kHz silences the whine but greatly reduces motor torque. So I have settled for 120 Hz, which is the lowest frequency available and the closest to the frequency produced by a conventional analogue controller running from 50 Hz AC power.
Motor soft start / stop 0.25 s acceleration, 0.25 s deceleration ( menu 1, 1, 8, 2, 2 )
Putting a brief delay into the throttle response will, at least according to the instructions for programming, reduce motor heating using the low PWM frequency. I can look for the effect of this when I get the model on a roundy-roundy test track, but I am not expecting motor heating to happen on a short layout at home.
I am not enjoying this part of the build very much. I want to have radio control if I possibly can, but the setting-up gets stressful when "nothing happens", especially when motor control seems to have been lost. So far, judicuous power-cycling has resolved these losses. Configuration involves five main menus each with up to five sub-levels, with navigation using two push buttons and one LED. The scheme is easier to work than it sounds, but Micron have mentioned a graphics-based programming tool on their web site, and I think this will make my life more relaxing.
Richard Gawler
Western Thunderer
27. Battery Bracket
My battery bracket is the final structural part of the build, and it will hold the saddle tank as well as the battery. This bracket has been a long time coming because I thought it would commit me to using a PP3 battery. The way it has turned out, I could still fit two AAA batteries instead, or indeed no battery at all.
I put the boiler back into the lathe and took a bit more material off each end of the opening. I will waste about 1 mm of space inside the tank immediately behind the smokebox (on the left here), but I need some metal in the boiler to support this end of the tank. I need the metal at the other end (beyond the terminals) because there is a gap between rear of tank and front of firebox.
The free space between a PP3 battery and the inside of the saddle tank is about 0.6 mm, so this has forced me to use 0.3 mm brass for the bracket to let everything fit together.
I have fitted the bracket as well as I can. Doing this has revealed an offset between the two top edges of the boiler, so I need some packing under this long side of the battery if the saddle tank is going to sit upright. I say packing because 0.8 mm is a bit much to call a shim
I have room for some insulation between battery terminals and brass model. The battery is specified for a thousand charge/discharge cycles so I will be soldering the wires onto the terminals.
The bracket is short enough to reveal the charging socket on the battery.
This all of the aerial I can persuade to come outside. I expect it will stay vertical like this for the next round of range trials. It is best left away from metalwork so at worst it resembles some sort of imaginary drain pipe.
The radio receiver is inside the smokebox and extends back into the space under the battery bracket. It is as far away from the motor brushes as I can make it.
The model is going to end up lighter than I expected because a lithium PP3 weighs barely half as much as an alkaline one
. Perhaps about 280 grams in all, plus crew. There is scope to add ballast in the smokebox and bunkers but I doubt I could muster another 20 grams and this hardly seems worthwhile.
My battery bracket is the final structural part of the build, and it will hold the saddle tank as well as the battery. This bracket has been a long time coming because I thought it would commit me to using a PP3 battery. The way it has turned out, I could still fit two AAA batteries instead, or indeed no battery at all.
I put the boiler back into the lathe and took a bit more material off each end of the opening. I will waste about 1 mm of space inside the tank immediately behind the smokebox (on the left here), but I need some metal in the boiler to support this end of the tank. I need the metal at the other end (beyond the terminals) because there is a gap between rear of tank and front of firebox.
The free space between a PP3 battery and the inside of the saddle tank is about 0.6 mm, so this has forced me to use 0.3 mm brass for the bracket to let everything fit together.
I have fitted the bracket as well as I can. Doing this has revealed an offset between the two top edges of the boiler, so I need some packing under this long side of the battery if the saddle tank is going to sit upright. I say packing because 0.8 mm is a bit much to call a shim
I have room for some insulation between battery terminals and brass model. The battery is specified for a thousand charge/discharge cycles so I will be soldering the wires onto the terminals.
The bracket is short enough to reveal the charging socket on the battery.
This all of the aerial I can persuade to come outside. I expect it will stay vertical like this for the next round of range trials. It is best left away from metalwork so at worst it resembles some sort of imaginary drain pipe.
The radio receiver is inside the smokebox and extends back into the space under the battery bracket. It is as far away from the motor brushes as I can make it.
The model is going to end up lighter than I expected because a lithium PP3 weighs barely half as much as an alkaline one
. Perhaps about 280 grams in all, plus crew. There is scope to add ballast in the smokebox and bunkers but I doubt I could muster another 20 grams and this hardly seems worthwhile.