Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Parallel DC Boost Converters?
(08-06-2019, 06:49 PM)completelycharged Wrote: The DC Boost converters you have listed will not work in parallel and a lot of non isolated boost converters which have current limiting will also not work.

My post on a poer supply option details the boost converters here :

The reason is that the return negative line is used to detect current with a small shunt resistance and when the negative lines are in parallel the current is distributed near evenly between all of the units, so one unit is sending out say 10A and only returning 3.3A (in the case of 3 in parallel) while the other units are sending out nothing and have 3.3A current sense active, which then prevents them from turning on.

I had played around with the 1200W and 1800W units a lot (blew FET's and repaired them) and they do work quite well with reasonable efficiency. The 1800W units have a 40A input current limit and think it's a 20A output limit. They also work as step down units, although they are not intended to be and a short failure would pass through the battery voltage (normally FET's will go open circuit).

That said I have only used them to step down from 50V to 48V while charging a pack from a power supply (and then stepping up from 50V to 54V as the pack charges).

why would 1 unit send out 10 and the others nothing? if im using 2 and they are both set not to limit current and output at say 13v, they should be outputting an equal amount of amps shouldnt they?also not sure what you mean about turning them on, as long as a current is flowing through them they will be on? 

for more info, my 2 bucks should never output more then a total of 20A and my battery is set up to do up to 30A.

@Korishan you might be right about putting diodes on the pos output of the bucks, i noticed as soon as i connected the bucks outputs in parallel, the other buck (which was not connected to the battery) had its power LED light up, so this shows that output from one buck is feeding into the output of the other. 

I'll look into adding those diodes as you suggested.

One thing i found odd, i set both my bucks out output at 11.75v but once i put a 200w load on my inverter after a little while the voltage coming out of the bucks still dropped from 11.75v to 10.6-8v. i thought the whole point of these things was that they were supposed to regulate voltage...?
Unit A + --> 10A
       - <-- 3.3A

Unit B + --> 0A
       - <-- 3.3A

Unit C + --> 0A
       - <-- 3.3A

What I found was that one unit would start up and the others would remain at a zero current flow. If I swiched off the one which was flowing one of the others would then increase to take over. That is what I found out the hard way testing them, expecting them to share.  I have not looked at the circuit in detail to figure out the exact mechanism, however as far as I can tell it is due to the negative return current sharing. Real shame as they would have been absolutely perfect for scaling to many kW if they worked.

If someone can get them working in parallel that would be awesome.
If you can't quantify how much they cost, it's a deal, I'll buy 5 of them for 3 lumps of rocking horse ......
right, so I've been thinking about this and doing some more research and came up with the following ideas:

firstly i will have to make my battery a bit bigger, i have a new car and a bigger space to store it so... more cells = less v load dip.

secondly, i looked into going down the 24v route. it seems 24v inverters have a better input range than 12v models. 12v models all seem to bottom out at about 10v, so when the cells reach about 3.33v the inverter will cut off, but 24v inverters bottom out at 21v which in a 7s battery works out at 3v per cell.

having more cells in series also reduces the current draw per cell, so the under-load voltage drop would be reduced even further, if i put a 300w load on my current 3s24p (96 cells) setup that would be about 1.04A per cell, but if it was converted to 7s14p (98 cells) it would work out about 0.89A per cell. again i reckon at minimum i should be able to make a 7s17p in the new container which at 300w would be 0.74A per cell so even better.

this all seems like going slightly bigger and onto 24v would give me access to much more of the capacity in the pack, though i know due the the load voltage drop I'll still probably lose 10-15% off the top. i was wondering if going to 10s/36v and using one of those 1200w DC bucks to buck down to say 28v would allow me to use even more of the cells capacity, at those voltages the current should never really exceed 10-12a which is well within the spec of the bucks.

what do you guys think?
Depending on your cells you might not want to go down to 3.3V. You might be bound to 3.6V to 4.0V depending on pack chemistry and loading.

Would really recommend you read these threads before settling on a voltage... especially if your thinking of close to 1A per cell

And then this one, particularly what happens to the different chemistries and additional loading on weak cells at a low charge state

I learnt a few really interesting bits from Wolfs testing regarding mixing cells in a pack

24V is better than 12V.. 36V is marginal to problematic without compromising (buck/boost)... 48V better but really only if you have a large load and large pack

Separate thought (just remembered) and did you know, how much, cor blimey..

Car engine is say 28% efficient
Typical car alternator say 60% efficient
Petrol has about 9.7kWh per litre
Assuming a petrol price (UK) of 128 pence (13.2p/kWh)

Your electricity price in the car costs 78p/kWh due to the poor conversion efficiency

78p/kWh and at current (rough) exchange rate, that's $1 per kWh.

Battery pack for the car to run the lights, etc let alone the camper ! Making me think of using a pack for the car in winter, when the headlight lights, fan, etc are all normally on and then charge at home. Buck down from 24V pack to 13.25V (just above float).

I did not realise it was this high, so the car is now the last place I will charge from.
If you can't quantify how much they cost, it's a deal, I'll buy 5 of them for 3 lumps of rocking horse ......

Forum Jump:

Users browsing this thread: 1 Guest(s)