ggoodrum said:
I think a 4p parallel board for the main pack leads would be a good idea, I've actually done this before, but it is too much of a pain to solder in the bullets. The problem I had is that I sized the holes to fit the 4mm bullets that are on the Zippy and Turnigy 5Ah packs, but the "loose" bullet connectors that HC sells have a slightly smaller diameter, so they don't fit snug into the holes. the next size smaller hole was too small. I gave up at that point, but after seeing what Kim (AussieJester...) did, where he made a wooden jig to hold the bullets, I might revisit this.
Some of the higher-C Turnigy packs are now coming with 5.5mm bullets, so if you are going to do two sizes, I'd vote for 4mm and 5.5mm.
-- Gary
Gary... I agree with almost everything you have said, except the pain to solder the bullets part!
I've done similar multiple diameter hard plugs attached to PCB in the past and there are a few alternatives to solve the issue of snugness or adaptability to various bullet / deans type without making multiple parallel boards...
Solution #1 - This would be my third runner up choice but it works and is solid if the solder joints are done properly!
Works for all size and types of connectors, round plugs/ sockets, Anderson Power poles, etc but it's a bit more time consuming than the rest!
Simply have normal VIAs bored for with 10mil (min) thickness traces running from each discharge plug (positive and negative) to the next VIA for the next discharge pin... this done on both sides with .9 double clad would provide the (remember it's flat and doesn't have as high of an inductance?, well there is somthing to be said for flat trace vs round wire - Now this is the simplest (and most time consuming portion of this solution) using aligator arms... presolder a small 1/2" section of either stranded and pre-tinned or solid copper (preferrably teflon) min 10AWG into each gold plug / socket... then simply ensure the diameter of the vias or actually I guess they would be Thrus (it's been a while since I cared to design multi layer boards for higher current handling)...
Solution #2 - This would be my second runner up choice because although it's simple(ish) it would possibly require either the use of clamp arms or for automating (or atleast making assembly efficient) and honestly I would build a jig for each type of connector to be fit... have a thru with this minimum diameter predrilled but with sufficient trace material surrounding (5-10mil) so that if needed you can drill the holes to a larger diameter with a dremel press and a proper bit to still require a bit of a press fit... The other option (and a fully viable one given the diameter of the via/thru and the width and thickness of the traces (dual plane) would be a jig to hold X connectors of proper male female order in exactly dead center of the via as it exists and then just solder away at the base to join it with the surround (insead of placing the wire within, just solder around but the jig would ensure proper placement and centering of various connector types (by jig it could be as simple as a 2x4 block / with the inverse of what needs to be placed on the PCB and prefitted (loaded) with the negative couplings (the ones we want on the PCB for parallel discharge).
Solution #3 - This is actually my first choice because it's simple, keeps distribution of volume taken by paralleling package in a left to right and mostly 2 dimensional or at minimum very short profile... also it's perhaps the easiest to peform manual soldering while ensureing maximum current ability and minimum resistance.
Simply design the far edge of the PCB with a series of Traces which seem to dead end but are via paralleled just about 3mm from the edge of the PCB. The width would be 6mm with either 6mm spacing between traces (to prevent high voltage possible arching or cross trace damage in event of a short) - now just use an alligator clip (or a jig if you can make one up, wouldn't be too hard once PCB was designed... could even be an identical PCB but mirrored with the opposite plugs mated already) and solder at the edges of the connector and the back side... this will provide a low resistance and strong connection... also it will allow for the last 2mm of sheath (the red, blue whatever they are) to be cut off and then the protective polarized connector surrounds to be simply placed over the plugs and sockets and affixed using expoxy... If you look at an HV110 or HV140 they do this way and actually you can beef them up (or atleast lower the resistance) by adding connectors to each side of the PCB for the EagleTree and using 2 sets of say 10G AWG 200C 600v Teflon wire you can half the resistance and double the peak current handling while halfing the power loss to voltage sag over the same length of run...
That's my .02 on the Parallel termination blocks... I will do somthing on these because I am damn tired of breaking out the 10G... cutting an output line, then cutting x input lines... soldering all the positives and negatives to the individual collector connections and then finally doing as tight a wirewrap as I can with solid core copper to bind the X number of 10G lines together before I solder them, cover them with high temp and v electrical tape and then 100C shrinkwrap. Its a time consuming process and requires a very fine soldering hand because if you slop outside the cup end of these damn things they really don't want to go into their connector housings, one trick I figured out is to use a lighter to heat them up (hold em with needle nose) then just push real hard and quick and the heat will help the plug slide right past the catch... took me a week to figure out - I was using a hammer and aluminum tube based chisel type thing which would apply pressure evenly across 90% of the terminator sheild head for forcing it on.
One last thing... Gary, I had a though I wanted to test out and thought I'd ask your opinion before trying it...
1.) The MKS balancers seem to do a much better (more accurate) balance job than the display would indicate - some spec somewhere lists 4% variance on display but much better on the balance side of things (I guess they are not inter dependent upon one another...
2.) I have 5 of these in total (for 2 packs) and I have a relay which kills the Ground line (isolated 3PST) for each of a maximum of 3 I use right now at a time on each pack... I have added a small delay (R/C discharge timer) which gives me about 4-5 seconds and then triggers another 5v relay (rated at .25ma max) which simply presses the balance button on the MKS (only 3 of mine have this mod) but what it allows is when I engage the charging mode... the balancers automatically power up (off the charge port input) and within 5 seconds begin their balance cycle.
3.) In addition to the 5 full working (but more than 35mv variance between them all) I have 4 units with cracked screens which are basically useless to me without the screens (well I could use them blind on packs on the bike but)... I have been thinking given the 400ma of balance current but the 66% duty cycle so between 225 and 300 max of balance current at a time however as I read in your previous posting the discharge control and load board is seperated from the main MCU/UI interface... this is how you managed to tap into the 7 lines I assume you needed to trigger your "MKS on Roids" discharge / balancer enhancer.
So now the question... do you see a potential issue with daisy chaining additional discharge / dissappation layers from my 4 broken LCD units and combining them in parallel with the existing units or would that reduce the balance current (parallel) and require some method of series connecting ?
I know you have studied these in more depth than I...
I guess I have one final question... the pcb seems to have many ports and hand offs (not just the ISP programming port) which are unused... do you see any sign (or better since it would be at 2400baud, data) of an interlink bus intended to join multiple Master units to provide a common reference voltage and high / low cell information from other MKS units... that would be the ultimate combined with your booster circuit but I would remove the onboard resistor discharge network and repackage the MKS in your booster box (or make it possible if you didn't do it?)
Gary with regards to your kits/PCBs...
Please reserve for me:
4 x 6S HVC/LVC PCBs (and parts if you have purchased in qty and are offering as kit, I saw the site already but I assume if you can assemble them you have parts on hand and could sell them as kits)
(These can be adapted for use on 15S adapting the jst-hxt 3x5S to fill the first 2x6S and the final one with 3S right? or it is covered in the manual) also can these be adjusted for the HVC optocouple for Lipo chemistry to above or below 4.16 if a user so chose to - so I could set cutout at 4.15 (component value changes and hard set is fine as this will be for testing various cutout levels on packs) or even as high as 4.20v?
3 x Charge Controller PCBs (and again parts if you have them on hand - how many PCBs do you have on hand - pm if you prefer, jsut want to see the capacity on raw materials!)
3 x HK Balance Booster PCBs (again with parts if available from you without delay at reasonable prices, but with the above... if PCB and BOM + directions but without parts is the fastest way to get my hands on and built... please just send PCBs!
Also...
I know were supposed to keep the cross threading to a minimum to avoid confusion but the next time you have a version ready for beta testing / even alpha testing, Im always down with receiving the smallest possible module (I have nothing smaller than 3S but mostly 5 and 6S) and I would be happy to test it on some of my packs which aren't in active use anymore (puffing but still doing 5000mah + at 2C (highest testing current I can pull) and have everything but a reflow station ... never needed one though, just use a super fine tip on my weller!
Great work Gary and Fechter, glad to see it coming to fruition and ordering status / ability!
-Mike
), and has both balancing and discharge modes. The latter allows you to pick a set voltage point you want the cells balanced to, like 4.10V, which is what I use. This is safer than the balance mode, which simply brings all the high cells down to the level of the lowest one. Occasionally, I've seen some units "overshoot" the low cell, and so it now becomes the low cell, and so on. The discharge mode doesn't do this. The cells all stop being discharged, once at the set point. Anyway, the only problem I have with the BMs is that with only about 200-300mA of balance current, it can take awhile to do a 10 or 15Ah pack. After some investigation, I determined it is pretty easy to tap into the balance circuits on these units, as they are on a separate PCB, mounted to the bottom of the main PCB, and do an external "booster", which adds another 1A+ of balance/shunt current. This worked so well, I've decided to make this BM booster available as well. 
