GGoodrum
1 MW
steveo said:
See above for the capacitor replacement. The 3.0V TC54 isn't stacked at Mouser. You have to get these from Digikey.
-- Gary
"Zephyr" - Finally, the "v4" Fechter/Goodrum/Hecker BMS...
- Thread starter GGoodrum
- Start date
See above for the capacitor replacement. The 3.0V TC54 isn't stacked at Mouser. You have to get these from Digikey.
-- Gary
What is the best way to test the LVC on a live ready to go setup?
I always like to occasionally check and make sure the LVC is correctly pulling the throttle down in case i have a bad connector or wire or something.
The old BMS had pads you could solder a switch to that turned on one of the LED's inside the LVC optos.
I wonder if we can just short the LVC + and - together with a switch? Or is there a simple way to turn on an opto LED? Shorting pin 1 and 3 of the TC54-N ?
I always like to occasionally check and make sure the LVC is correctly pulling the throttle down in case i have a bad connector or wire or something.
The old BMS had pads you could solder a switch to that turned on one of the LED's inside the LVC optos.
I wonder if we can just short the LVC + and - together with a switch? Or is there a simple way to turn on an opto LED? Shorting pin 1 and 3 of the TC54-N ?
Yes, either of those methods should work. Don't short the TC54 if you are using the "C" flavor.
The LVC+ and - pads would probably be the easiest. You could even wire up a swtich to them for testing.
One more way is to use a 1.5v alkaline battery and hook it up across a cell (without the main battery connected). This should trigger LVC.
The LVC+ and - pads would probably be the easiest. You could even wire up a swtich to them for testing.
One more way is to use a 1.5v alkaline battery and hook it up across a cell (without the main battery connected). This should trigger LVC.
steveo
100 kW
-can anyone specify what that aluminum bar from mcmaster is for?
-is it needed in a 24s bms?
also wanted to ask if one of the bms channel wire accidently disconnects while evey other channel is connected, will this damage the bms or bms channel?
thanks
-steveo
-is it needed in a 24s bms?
also wanted to ask if one of the bms channel wire accidently disconnects while evey other channel is connected, will this damage the bms or bms channel?
thanks
-steveo
The bar is to conduct the heat from the resistors to the case. It goes between the bottom of the board and the case. If you mount the resistors on the bottom of the board so they directly contact the case, you don't need the bar.
If any tap wire goes open, nothing should be damaged. In fact, it should be able to operate nearly normally as now a pair of cells will be going to a pair of cell circuits. If one cell goes low enough to worry about, it should still trigger the LVC. During charge, a similar thing will happen, but if the pair of cells are very unmatched, you could get some overvoltage on one.
If a tap wire gets connected to the wrong cell, all bets are off and something will most likely get blown.
If any tap wire goes open, nothing should be damaged. In fact, it should be able to operate nearly normally as now a pair of cells will be going to a pair of cell circuits. If one cell goes low enough to worry about, it should still trigger the LVC. During charge, a similar thing will happen, but if the pair of cells are very unmatched, you could get some overvoltage on one.
If a tap wire gets connected to the wrong cell, all bets are off and something will most likely get blown.
Gregb
100 W
How critical is the paralleled 6R2 pair? My supplier claims I ordered 14 not 16 (who knows why) and he wants $15.00 to deliver 2 more..... The closest I can get is a 3R3 5W wirewound which is inside the tolerance of 5% or 6r8 5w WW which is .05 outside. Are they critical to stop the oscillations?
thanks
Greg
thanks
Greg
I don't think it will be too critical. I tested with wirewounds and no oscillations.
Hifechter said:
If you have to go wire wound then go for the ones that have a built in heatsink with holes to attach to a bigger heatsink. From what fechter said to me in a pm a 3r3 would be the better option you will end up with just under 1A balancing but if you use the 10W that are not too hard to find in that type you should be ok.
If I remember right the higher the rated wattage the lower the heat produced.
Geoff
Gregb
100 W
Thanks for that but while they may be a bit cooler to the touch they produce the same heat from the same current. P=I squared R......
True. If all the heat is eventually getting dissipated on the case, the case temp will be the same either way. A bigger resistor will tend to run cooler due to larger surface area. I'd be more worried about making it fit in the case. Those TO126 transistor looking things are extremely compact and rated for 15W with the right heat sink.Gregb said:
http://www.caddock.com/Online_catalog/Mrktg_Lit/MP9000_Series.pdf
Gregb
100 W
wow. didn't even know they existed. Unfortunately neither do my normal suppliers................... :?
hiGregb said:
Mine don't know what they are as well, I found ones in digikey value 3r3 that would fit the bill only problem is minimum order of 360 but come to think of it that would only be enough for 15 of the 24 channel zephyrs all that would be needed is a bit of leg manipulating to make them work and a load of USA getting together to place an order.
Geoff
Gregb said:
They're expensive as hell too. I wouldn't really recommend them but they are nice looking.
Zenid
100 W
Why would you need cell-tap wires to be thicker than the main + and - wires needed for the full charge current? Other people are telling me that the cell tap wires only need to be teensy to handle the 'bleedover' current when the cells are fully charged...fechter said:
Is this something to do with this particular BMS design?
[EDIT] I was getting confused here about the AWG wire-grading system, disregard
Hi
Thin wires that are long have resistance so give a false voltage value by using thick wires you can have long wires and not have a voltage drop that is noticeable. The other thing about using thick wires as apposed to the standard thin 22 gauge that you find on LiPo balance wires is you can pump more current through them without the wires getting hot.
Geoff
Thin wires that are long have resistance so give a false voltage value by using thick wires you can have long wires and not have a voltage drop that is noticeable. The other thing about using thick wires as apposed to the standard thin 22 gauge that you find on LiPo balance wires is you can pump more current through them without the wires getting hot.
Geoff
heathyoung
100 kW
Gregb said:
In Aus - Farnell have them. Or element 14 whatever they call themselves.
Gregb
100 W
Yes just found a mob called AJ distributors in Adelaide do too. Will sell any quantities minimum post $15.00.
Zenid
100 W
Thanks, Geoff, though I'm not sure where to get suitable cell tap wires from. The holes in the end-piece for the case for these wires are about 2mm, and the 2mm (total) diameter wires I've looked at (for example http://www.maplin.co.uk/miniature-extra-flexible-wire-6202) seem to be have way too thin guage cores - 0.24mm2 in this case, which must be about 24 AWG. Do you happen to know of a suitable UK supplier to get some thicker core wires from?geoff57 said:
Gregb
100 W
If you are really worried about it, just use the thicker wire external to the unit (ie that you can get through the hole) by stripping the insulation inside and replace with heat shrink.....
Hi
Zenid if you are buying through maplins then go for the 24/02 wire, I had a look at that in comparison with some 18 gauge that I got from Gary with some LVC HVC boards, the size is so close as to not make any difference. It is PVC coated and nice and flexible the wire with PVC will fit through the holes in the zephyr end plate. You get 10m for less than £3.
Geoff
Zenid if you are buying through maplins then go for the 24/02 wire, I had a look at that in comparison with some 18 gauge that I got from Gary with some LVC HVC boards, the size is so close as to not make any difference. It is PVC coated and nice and flexible the wire with PVC will fit through the holes in the zephyr end plate. You get 10m for less than £3.
Geoff
Gregb
100 W
Page 6 of the instructions refers to 14 pin chips. That should be 8 pin and I think it would be more helpful to refer to them as the opto isolators or by part no or cct designation and maybe include their designation on future iterations of the board. On my instruction sheet photo they look like one big black mass.
Zenid
100 W
Before I solder on all my 47uF caps, on can you confirm that the ones in the picture are still shown the wrong way round, and that the white strip should be ON THE RIGHT and NOT the left as shown?fechter said:
Just want to make sure I have this straight. Thanks.
Gregb
100 W
I have noticed some changes between my board 4.4 and the photos and the layout drawings. In particular the I kohm trimmer. The alternate parts number choice is too big and I notice the transistor next to it has been repositioned in the photo. I can find no mention in the text or on the drawings of the "2 pin header". Is it for the alarm, hvc or lvc??
Gregb
100 W
you are right. according to the schematic they are the other way round. why weren't we told by email or something? I now have to unsolder all mine......not impressed.Zenid said:
Zenid
100 W
Well yes. I was hoping either Goodrum of Fechter would post an alert to clarify this and try to prevent more people falling into this trap. I tend to refer to photographs if I have any doubts, so it would be nice if a photo of a correctly wired one could be posted.Gregb said:
I'll be taking pictures of mine as I progress, and they're more than welcome to use one of mine if they want a shot of one in a half-assembled state...
