$1 battery connector challenge

[jondoh]

I am selling connector terminals which I guarantee superior to deans or anderson power poles for AWG #14 wire.

For the first 3 people who send me paypal payment of $1 (please send pm for paypal email address), I will send the connectors. I will be selling this pigtail kit for $7. I only ask for prompt, posted feedback on the connectors. Please also indicate wire gauge for battery and controller. Blue terminals are for AWG #14 and #16. Yellow terminals are for AWG #10 and #12.

IMG_5718.JPG (50.9 KiB) Viewed 1253 times

Wire color to change to red with male quick disconnects on the pigtail and female quick disconnects terminals to be installed by customer (opposite of what's in photo).

This connector has a lot of good things going for it:
1. Connector housings are polarized. You can connect only one way.
2. The connector housings latch to prevent accidental disconnection.
3. The contact pins lock securely into the housing. No need to worry about pins backing out of the housing. When this happens, it's something you can't immediately detect visually.
4. Durability. I have been using this connector for bicycles for years and while they're officially rated for about 50 mating and unmating cycles, I have gone up to a few hundred cycles without ill effect.
5. Contact pins have crimps for both conductors and insulation for strain relief.
6. High contact pressure.
7. Low mating force.

IMG_5714.JPG

(34.26 KiB) Downloaded 594 times

IMG_5715.JPG

(39.02 KiB) Downloaded 594 times

IMG_5716.JPG

(43.46 KiB) Downloaded 594 times

[Jeremy Harris]

Any idea what the current rating of these is, please?

They look mechanically a lot better than the Deans, but the Deans are a hard act to beat for current handling capability (Deans are rated at 70A, I believe, but many have run them way over that without problems).

If they can be crimped to fat cables (10g) with an easy to get hold of crimp tool, so removing the need for those pigtails then that would be even better and I'd certainly be interested in some.

Jeremy

[jondoh]

I'm using AWG#14 wire. With only two circuits, this connector is officially rated at 15 amps-- but that's a DC current running 24 hours a day, 7 days a week. I've used this connector on my phoenix racer for years with a 40 amp controller with no problems at all. The wires may be a little warmer after a spirited ride on a 40 amp controller and I wouldn't want to use it for a larger controller. I say the connector is just right for controllers around 40 amps and under.

The issue with connectors is heat. Heat isn't good for electronics. Heat causes the copper alloy contacts to fail more quickly. Since ebikes are run at the higher currents for short bursts of acceleration, there isn't much of a chance for heat to build up (usually). Perhaps if you are riding in desert heat or up mountains all the time, and you have a very very large battery, there might be an issue using this connector with a 30 or 40 amp controller.

There are contact pins for this connector which can accept AWG#10 wire. The official current rating would increase to 20 amps. The problem is that the crimp tool is expensive. I had my crimps done by an authorized assembly house. Crimping an open barrel contact can be tricky when performed with hand tools. You find a fair number of problem crimps when using open barrel contacts with hand tools which is why a lot of people both crimp and then solder connections. A properly crimped terminal has approximately the same tensile strength of any section of wire. A pigtail may introduce another connection point, but i've found the closed barrel, solderless terminals easy to crimp by hand and quite reliable. Too bad they cannot be mated more than a few times reliably.

If you are uncomfortable with quick disconnects, I also have end splice terminals for AWG #10 for a permanent, secure connection.

The logic behind the high current capability of dean/anderson connector is:
1. very thick closed barrel terminal
2. a separate spring member that applies high contact pressure

It's an effective design when properly assembled. Drawbacks to the design are
1. no housing latch
2. high mating forces-- which might contribute to pins backing out of the housing
3. no insulation crimp means less strain relief on wire

If you are in a situation where you are connecting and disconnecting your battery a lot, this is the better connector if only for the housing latch. The housing locks the pins inside very securely. The connector is very easy to mate and unmate. I believe this connector (JST VL series connector) is a far more elegant solution.

[Jeremy Harris]

Thanks very much for the info, I guessed that the crimp tool would be spendy, but one can but hope otherwise!

As you say, they look OK for modest power levels, contact resistance being the major restriction for use on mega power installations (it's connector resistance, primarily contact resistance, that causes connector local heating). I'm sure they're better than Andersons in some ways, my experience with Powerpoles is that they aren't as reliable as many think when run at currents around 30A or so, so this connector might be a better bet as a replacement.

If someone was to come up with a decent cable strain relief and housing for the Deans Ultras then I think they'd be ideal, as it stands they are electrically very good but have some significant mechanical deficiencies, although to be fair we're using them in applications for which they were never intended.

Jeremy

[jondoh]

So I've had one positive review so far.

http://endless-sphere.com/forums/viewtopic.php?f=2&t=23758

I'd like to say this challenge is still open to a few more people who are sick of their deans/andersons!

[jondoh]

Jeremy said:

contact resistance being the major restriction for use on mega power installations (it's connector resistance, primarily contact resistance, that causes connector local heating

Ah yes. Having worked for a connector company, lower contact resistance is the key to long term quality connections.

Even though this connector was designed for power levels up to 20 amps dc continuous, it was tested and qualified to signal level loads. Contact resistance for signal level connections are specified in the milliohm range-- often 20 milliohms or less. This is not a test you can do with just any old DVM either. You have to have a special lab grade 4 point probe milliohm meter. The JST VL connector passes this test. Even though 20 milliohm is overkill for a power connection (surface oxides and corrosion that may build up on the contact surface gets blown out at higher current levels) it does say a lot about the quality of the connection and performance over the long term.

That said, a dean or anderson connector should have very good contact resistance-- at least initially. I've wondered if the contact pressures were so high as to hasten the wear of the tin plating. Tin plating can wear off after repeated on the JST connector too-- but the design is precise and elegant so that a proper amount of force is applied and applied consistently. The contact pins have multiple contact points (three, i believe) to increase the contact area.

The only problem I ever had with a dean connector was the wire coming off the terminal-- and this was after a year and a half of constant use. It's not a bad connector but I maintain the VL connector is better.

[liveforphysics]

20mOhm of resistance is considered OK in a test for a power connector?

My entire pack resistance is under 20mOhm (my new nano-tech pack will be under 6mOhm!). I prefer connectors that are best measured in uOhm's rather than mOhm's. And yes, I have a uOhm meter (LCR bridge) to test them.

HobbyCity offers the XT160 connectors for under $1 each, insulated, don't come unplugged unless you yank them, and you can mate 6-8awg wire to them fairly easily, 10awg seems almost too small. They handle 100amps continuously with ease.


These might be fantastic for folks running lower power levels, but if you have high current needs, $1 can get you something superior IMHO.

[Gordo]

Luke;
I guess I have wired too many boats used in salt water. Crimps just don't work for long without solder. They are only intended for a mechanical connection as far as I am concerned. I have a box full of different styles of crimpers from $50 to $150, everything from BNC to SS messenger wire. It is great to have a nice strong water resistant connector on one end but I fail to see the advantage when the other end uses a crimped automotive bayonet connector? I much prefer soldering the bare crimped bayonet and then using shrink tube over it than trying to get a reliable crimp through plastic. This also goes for off road machinery or even highway trucks that have to run on salted roads. The cost of one ticket plus a trip to the shop makes soldering well worth the time, the first time, EVERY TIME. JWID

[spinningmagnets]

I agree on the soldering ends and heat-shrinking. I've seen plenty of bayonet connectors in old cars that have corrosion, even though they're aluminum.

I like this style of connector for medium-power builds by someone who lives where its not too rainy. Add a dab of silicone to where the wire goes in, and they are reasonably "weather resistant" though not weather-proof. Few would purposefully ride into the rain, but many sometimes unexpectedly get caught in a drizzle on the way home (Oregon?).

I got a set from Jondoh to review and I also compared them to a more expensive connector that has better weather resistance and higher amps (for a higher price).

http://endless-sphere.com/forums/viewtopic.php?f=2&t=23758

[liveforphysics]

Luke;
I guess I have wired too many boats used in salt water. Crimps just don't work for long without solder. They are only intended for a mechanical connection as far as I am concerned. I have a box full of different styles of crimpers from $50 to $150, everything from BNC to SS messenger wire. It is great to have a nice strong water resistant connector on one end but I fail to see the advantage when the other end uses a crimped automotive bayonet connector? I much prefer soldering the bare crimped bayonet and then using shrink tube over it than trying to get a reliable crimp through plastic. This also goes for off road machinery or even highway trucks that have to run on salted roads. The cost of one ticket plus a trip to the shop makes soldering well worth the time, the first time, EVERY TIME. JWID

Gordo, I whole heartedly agree my friend. My $1 connector challenge suggestion is the XT160 connector, which is done by solder-only. I don't have a crimp connection in my entire bike (or my racecars, etc). They are much like a deans ultra, but at little easier to work with.

http://hobbycity.com/hobbyking/store/uh ... oduct=9572

[jondoh]

XT160

XT60.jpg

(51.04 KiB) Downloaded 236 times

Really? No housing lock, solder only contact pins and you think this is better? They look rather small in the picture but I guess they can handle 60 amps ok.

What the XT160 has going for it is that it has GOLD PLATING. Gold can cover up a lot of deficiencies in precision and quality... until it wears off.

What I see as a deficiency in the design of the XT160 contact pins is that they're round. Round contact pins are a throwback to the earliest connector. I was told that, back in the day, the military made connectors by machining them. The easiest connector pin to machine would be a round one. These first connectors were literally and figuratively bullet proof since the machining process was extremely precise. Mating forces were very low if you polished and gold plated them. Problem is that machining pins is ridiculously expensive.

Stamping and forming contact pins lower the cost of manufacturing but the problem is that it's very difficult to stamp and form a ROUND pin for both male and female sides. This, in turn, leads to difficulties in controlling contact pressure and mating forces. Another problem with round pins is keeping them locked into the plastic housings. The classic stamped and formed round pins have locking lance on the pin itself. The VL design has the locking lance molded into the housing. The pins are relatively clean which is an advantage as the lance on the pins tend to tangle and get damaged. This leads to problems with pins backing out of housings-- and this is a problem you cannot easily see (unless you have xray vision).

so to sum up the VL connector advantages:
1. locked housings
2. SQUARE contact pins are locked in place very securely with housing lance
3. SQUARE contact pins make multiple contact within the pin
4. consistent contact pressure makes for easy and high quality connections

Sorry but point for point the VL kicks but on the XT160. I DARE you to compare.

[liveforphysics]

Sorry but point for point the VL kicks but on the XT160. I DARE you to compare.

Ok, I accept the dare.

First, these are NOT stamped and rolled pins at all, they are precision CNC machined pins and sockets.

Second, a square socket inherently only gets to use 1/2 of the available surface area for contact. This has 4 branches out on each connector to press against the female walls rather than just top or bottom.

Third, the connection itself (the part you want to have excellent surface clamping pressure) is firm enough with these connectors, RC planes pulling 10g manovers, RC cars bumping and skipping along the ground at 80mph, and planes and helicopters crashing into the ground at 100mph do NOT cause this connector to dislodge. This gives you an idea of the contact force at the CONNECTION POINT (not just an external plastic clip that doesn't help with the connection), which is at least 2-orders of magnitude above any folded square pin-socket type I've personally seen. From much experience and 10's of thousands of other folks using these connectors on RC vehicles, we know coming disconnected when not wanted to disconnect is simply not an issue, even at vibration/shock/g-loads above what would kill any human on an ebike.

If you desire easier mating force (because they require at least 10lbs of force or so to mate), you can squeeze the male tabs closer together with needle nose. If you want more force, of if the connector surface wore over time, or got bent into a smaller OD than the female side, you simply take a punch or an ice-pick, or nail, or whatever, and tap it's tip the middle of the male prongs with a little hammer, and you can re-spread the male side to be any way you want it, from needing pliars to connect and disconnect them, to something you can connect/disconnect with 1 hand.

And yet... if you make some dumb mistake, like forgetting to replace the fasteners on your battery (or whatever), or hook a wire on a tree branch or something, or crash, or a hundred other things, this connector doesn't rip your wires out and require repair, it simply disconnects protecting the wiring. A clip solution can't offer this feature.

This connectors machined pins are cast into the high temp nylon holder, so they are sealed from the back (so a dab of grease in the connector side of the housing makes a weather-proof connection), and no amount of stress on the wires from the back can effect the quality of the connection in the connector.


Lastly, this connector is a 5mm ID bore on the female, 5mm OD on the 4 tabs of the male pin. The wire connection side is a 5mm ID pocket to solder. I'm guessing the square pin has maybe 1/4 this cross sectional area at the best, and no ability to connect 6-8awg wires?

Also, as many in the hobby world will confirm, the gold on gold contact surface on these seems to last even after loads of connect/disconnect cycles in my RC heli's and other things, enabling a consistent and corrosion-free connection.



Now it's your turn.

[Jeremy Harris]

I'm with you on this one, LFP. The connectors MUST have a contact resistance that is at least an order of magnitude less than the other sources of resistance in the circuit, preferably a lot less, as even one tenth of the resistance in the connector means that it's going to be 10% of the total resistive losses in the system. Typically high performance controllers, running low Rdson FETs, will have an effective resistance of around 2 mohm, maybe less. Motors vary a lot, but typically will be around 50 to 100 mohms, down to maybe 5 mohms for a high power, low turn count outrunner.

Contact resistance is dependent, in the main, to the area of the connector that's making contact and the pressure with which that contact is maintained, which is where well-designed cylindrical pins and sockets have a big edge.

The issue over gold plate is a bit of a red herring, because gold isn't that good a conductor. The gold flash is there for corrosion protection, something that badly affects tin plated terminals, or silver ones, some to that, and because gold has the property of effectively cold welding to itself without sticking, making for intimate contact between the two metal surfaces. As long as enough stays around to do this then the connector will still work fine - all it needs is a layer an atom or two think to do this.

The newer style of bullet connector, with the split male end, has a high insertion force, very high contact force and a high contact area, which makes it all around a pretty good design. It's no accident that many other very high current connectors that are used industrially have a very similar construction, with a solid tube for the female part and a split rod for the male part - its a design that works well.

The downside of both Deans Ultras and the XT60s is the lack of cable support an the reliance on a soldered joint for mechanical strength. Adding a bit of heatshrink partially alleviates this, but a design like this with better wire support would be considerably better. I'm using the plastic shields for the 6mm bullets that HK sell and find they do a reasonable job of supporting the wire, but it still really needs some form of proper strain relief.

Jeremy

[jondoh]

I'm preparing a response for LFPs latest post but first...

When I said 20 milliohms for signal circuits, this is the upper limit after LIFE TESTING (after being thrown in the lab oven, after vibration testing, etc). The idea is that if the signal is low level and low power, at 20 milliohms, the oxides between the contact surfaces may not allow the signal to pass or may introduce noise. A freshly mated contact pin usually has less than 10 milliohms of contact resistance for tin plated surfaces.

If you add up the resistance of the wire from battery to the motor, you'll most likely have something higher than 200 milliohms.

Not to offend anyone here, but it seems that a lot of people on these boards feel that whatever has the highest EXTREME specification is better, "you know, never mind that my controller puts out 30 amps max (of which the RMS value is something under 10 amps), i want the connector rated for 100 amps!" Fine. Take two wires and terminate the ends with a ring terminal and put a brass bolt, washer and nut through them to connect them together. Wrap the whole thing up in electrical tape and then dip it in tar and you'll have a connection that can handle 100 amps and can be used under the ocean. This connector costs pennies, is water proof, can handle high power. So what's wrong? It's not easy to connect and disconnect for one. It's not an elegant solution.

LFP mentioned the mating force is 10 lbs. This sounds very very high. Mating forces are too high for gold flash plating, i think. High mating forces are not a trivial matter. Connectors that are used for high volume consumer products (like the VL) MUST HAVE LOW MATING FORCES. If you use your fingers to mate a 10 lb connector thousands of times a day, you will be filing for workmans comp in about a week. Sure contact surface area is important, but it needs to be balanced with other factors.

I'm busy at the moment but I will detail why i think the VL is STILL the better overall connector solution. I do have ring terminals (AWG#14, blue) for 6mm screws if you are interested in the tar ball waterproof submarine connector solution... honestly, i do have ring terminals.

[Jeremy Harris]

If you add up the resistance of the wire from battery to the motor, you'll most likely have something higher than 200 milliohms.

Sorry, but that means that there's as much, or even more, more resistive loss in the wiring than there is in the motor. At a humble 25 amps (which is typical for a low power ebike) that 200 mohm wire loss is soaking up 125 watts, which is plain crazy! I mean, an ebike running on 36V at 25A (pretty typical) will be losing 14% of its total power in the wires?

Let's look at some numbers and see what sort of wire gauge gives us 200 mohms of resistance, for the typical 8ft or so of high power wire on a bike. First, here are some typical ebike power cable wire gauges, with the length needed to give around 200 mohms resistance:

10g wire has a resistance of around 1.18mohms per foot, so you need around 169ft of it on your bike to get 200 mohms.
12g wire has a resistance of around 1.87mohms per foot, so you need around 106ft of it on your bike to get 200 mohms.
14g wire has a resistance of around 2.97mohms per foot, so you need around 67ft of it on your bke to get 200 mohms.

In fact, to get 200 mohms with around 8ft of wire you need to wire the bike up with wire of around 23g. That's right, really thin hook up wire with a maximum safe current handling capability of around 2 to 3 amps.............................

Sorry, no offence intended, but this single statement on wire resistance undermines the credibility of the rest, because it's so far off the mark.

Jeremy

[jondoh]

Jeremy,
you are right. I was going off of memory testing connectors in the lab where we subtracted resistance of the length of a few inches of wire which was generally 10 milliohms or so-- but the wire gauges were smaller than AWG#14.

But if you are looking for a connector with a contact resistance that is one tenth of say 2 or 3 milliohm. I'm not sure you're going to find it in a normal connector.

[liveforphysics]

If you were on a production line, made to connect xt60 connectors thousands of times a day, you would use some custom tooled pliers.

For connecting and disconnecting a few times a day, even a few 10s of times a day, its simply a non-issue. Fail to see how high mating force impacts the ebike user in a meaningful way (10s of thousands of RC users sure don't mind.)

The gold stays on fine. How do we know? Because folks who fly RC helis and planes and cars with batts that charge in 6-10mins connect these things 10s of times a day on a regular basis (including myself) and I've yet to see the gold worn off on any of mine.


I have a precision uOhm bridge meter (4 wire Kelvin clips etc).
My entire wiring system including all connectors is under 8mOhm.
If you want some third tests on your connectors, ive got the LCR bridge, constant current sources, and Flir thermal imagers to check things out.

[jondoh]

pm me of where you want them sent