F Series driver recommendation

Prawn Connery

Prawn Connery

Well-Known Member
Gotta cough to get off, mate! I usually have a shit before I hit the bong, as I tend to sit there all day if I'm stoned . . .
 
  • Like
Reactions: 3GT
Prawn Connery

Prawn Connery

Well-Known Member
3GT said:
A version has an internal potentiometer for dimming via flatblade screwdriver

B version has external dimming leads to connect a pot to? I went B version for all my drivers :) $2 100k ohm pot from jaycar soldered on and you're in business.
Click to expand...
What's the advantage? Difference between a screwdriver and a knob, or do you also have a digital readout connected to the pot?
 
3

3GT

Well-Known Member
Prawn Connery said:
I hit Arrow for the SI-B8R521560WW, but they were on back order and quoted minimum 150 purchase.
Click to expand...
Yeah they're not as good for stocking everything! I bought elg-200s off them for $50au ea, they didn't have the 240s in stock.

Hmm B version; easy to dim (turn knob), 0-5v / arduino dimming capable?

A version; 50-100% dimming via small screwdriver.

I haven't even connected my pots to the drivers yet, they're just ran @700ma or @350ma for the panels wired parallel. As you know though getting these lights as close as possible gives us the least losses so fine dimming adjustments may be a regular task? IMO buy B version + 100k pot and Mount that fucker somewhere handy/sexy looking.

I also have an arduino with soil sensors, water level sensors, temp/humidity sensors, co2 sensor, solenoid activated water valves etc that I need to get around to playing with/setting up.. $200 and I can automate every damn crazy idea I could concoct
 
alesh

alesh

Well-Known Member
Prawn Connery said:
What's the advantage? Difference between a screwdriver and a knob, or do you also have a digital readout connected to the pot?
Click to expand...
A (internal screwdriver pot) can be dimmed only to 50%; B can go down to 10%. The B version can be controlled externally via ie an Arduino (or any MCU).
e: too late :bigjoint:
 
Prawn Connery

Prawn Connery

Well-Known Member
When you say "internal" is it easily accessible, or do you have to unscrew the cover first?

As this is only for veg/cloning, I'm not too precious about fine-tuning it. I would rather just set and forget, raising and lowering the lights as my plants veg prior to going into flower.

My flowering chamber is going to be a different story . . . but the purpose of building this clone/veg light is to learn along the way so I can decide what final format my flowering chamber will be. That will be in the region of 450-500w.
 
Prawn Connery

Prawn Connery

Well-Known Member
3GT said:
I also have an arduino with soil sensors, water level sensors, temp/humidity sensors, co2 sensor, solenoid activated water valves etc that I need to get around to playing with/setting up.. $200 and I can automate every damn crazy idea I could concoct
Click to expand...
Mate, if there's one thing I've learned about growing over the years, it's that the fewer things that can go wrong, the more reliable the system, the more consistent the yields. Reliability trumps power/efficiency in my books, because all you have to do is lose one crop to a pump failure or something else, and all that efficiency goes out the window.

The trick is to find the balance while keeping it fool-proof. But that's just my philosophy, and obviously doesn't apply to everyone. Some people have got more time to spend monitoring their system, so they can catch things before they affect plant health whilst driving optimum efficiency. I'm more of a set-and-forget/check every now and then kinda guy.
 
nfhiggs

nfhiggs

Well-Known Member
Prawn Connery said:
Thanks for all the replies so far. You guys wouldn't believe the amount of info I've gleaned on these boards over the past few weeks. For an old-skool grower like myself wanting to get back in the game, it's fascinating!

I've been doing a bit more reading, so I think I'm starting to get this . . . but please correct me if I'm wrong!

So I've ruled out running 2s 2p with the 2.1a CC driver because, from what I understand, parallel circuits with constant current drivers are susceptible to thermal runaway. What happens is the current is split between two sets of boards. Without a current mirror or over-current protector, there is the risk one board set will draw more amps than the other. This in turn lowers the voltage, requiring more amp draw. The extra amperage is taken from the other board set, which raises its voltage, requiring less amp draw. Eventually, one board set is hogging the amps and burns out (thermal runaway) and when it does, all the power goes to the other board set, which also burns out. Unless the total amps in the circuit are less than 50% of the maximum amperage of each board, in which case there may still be an imbalance between each parallel board set.

That leaves 4 parallel boards with CV or 4 in series with CC . . .

Running all boards in parallel with a constant voltage driver means one or more boards may still be drawing more amps than the others, but there is no risk of thermal runaway. But it may still create imbalances in each board, as well as individual leds. In addition, as each board heats up, it requires more amps for the same set voltage, and efficiency is lost through the ballast resistors.

(Correct me if I'm wrong, but the spec sheets I've read state forward voltage drops with higher temperature - not the other way around.)

The main advantages to CV that have been pointed out are that you can add boards without exceeding driver voltage, as well as the ability to identify, isolate and/or bypass individual boards if there are problems.

The last point is worth noting: I guess with a CV driver you could test each individual board without exceeding amperage - you can't do that with a CC driver. Or can you (by winding down the amperage)?

I have no experience with these drivers, so I don't know.

What I do know is I'm not planning on adding any more boards to this set-up, nor playing with the output between cycles, as it is purely for vegging/cloning - it will be a dedicated light source - and the 4 boards I've purchased are more than enough for my needs, and will likely need to be driven at about 55-60% for the desired output. That's plenty of room to grow (pun intended!).

That just leaves the constant current driver.

What's the difference between the A and B drivers?

And finally, I've posted a spec sheet link below. On page 21 you can see the relationship between current and voltage. I've worked out that at 1.08a, each board will drop 46v (2.75v/120ma per diode).

http://www.samsung.com/global/business/business-images/led/file/product/lighting/201511/Data_Sheet_LM561C_Rev.3.2.pdf

This seems to be my sweet spot. So my question is, if I had a HLG-240H-1400, and I dialed it down to 1.08a, would the resulting series voltage of 184v exceed the driver maximum output voltage of 179v - or is there a buffer, or should I be conservative and simply run the HLG-240-1050?

EDIT: OK, I think I've answered this one ^. Apparently the HLG-240H-1400 will hold the total voltage at 179v (maximum) and lower the current - which means at its maximum voltage output, it becomes a constant voltage driver. To keep voltage below 179v, I'd need to drive each diode at about 2.8v/75ma (44.8v per board) for total output of 0.675a.

Is this right?
Click to expand...
I've found that the current draw difference between boards when running parallel in on a CV driver is inconsequential. Yes you may have some differences, but it simply does not amount to anything. For instance when I got my strips, I put each one an the driver for an hour to see its current draw - the variance between ten strips was under 10% - lowest was around 980 mA and the highest was about 1050 mA.

Ballast resistor is not needed with strips
 
Prawn Connery

Prawn Connery

Well-Known Member
7% still seems like a lot. But I can see why you guys like CV drivers, as they appear to be a lot more flexible than CCs. I also read this on the Mean Well site:
The MEAN WELL HLG-H series LED drivers are in fact constant voltage LED drivers, which can be used in constant current mode.
Click to expand...
Is this true? Can the HLG-240H-48 effectively be run as a constant current driver in a parallel circuit? Because that would appear to be the best of both worlds . . . though I can't see how that works if the strips are not in series . . . Or do they mean they put out a constant voltage within certain current parameters?
 
nc208

nc208

Well-Known Member
Prawn Connery said:
7% still seems like a lot. But I can see why you guys like CV drivers, as they appear to be a lot more flexible than CCs. I also read this on the Mean Well site:

Is this true? Can the HLG-240H-48 effectively be run as a constant current driver in a parallel circuit? Because that would appear to be the best of both worlds . . . though I can't see how that works if the strips are not in series . . . Or do they mean they put out a constant voltage within certain current parameters?
Click to expand...
Yes, in post 6 I mentioned - Stick with the 1400 ma or look at using a 48v 5 A and wire in parallel for 1250ma per string which works out to 138.8 ma per diode.
the 240h48 is exactly that a 48v 5 A driver.
http://www.mouser.com/ds/2/260/r1424_3-1109735.pdf
http://ca.mouser.com/ProductDetail/Mean-Well/HLG-240H-48A/?qs=Pc4kzEvJlcS8QtF1/k4rXg==
 
Prawn Connery

Prawn Connery

Well-Known Member
Quick question: the 1400ma CC driver peaks at 179v, but total voltage in the series at 138.8ma per diode is about 186v (just under 2.9v per diode x 16 for about 46.5v per strip). What will happen to the current in the series if voltage drops to 44.75v per strip (179v divided by 4 strips)? The 2.9v figure is a reference to the FC vs FV graph on page 21 of this data sheet: http://www.samsung.com/global/business/business-images/led/file/product/lighting/201511/Data_Sheet_LM561C_Rev.3.2.pdf

Am I interpreting this correctly? This is the bit I'm a bit confused about. If the driver can't meet the total voltage requirements of the series, then doesn't that mean it can't supply the full 1400ma per strip?
 
Prawn Connery

Prawn Connery

Well-Known Member
I read that, which is why I asked. When he added a 5ohm resistor - I think, the equivalent of adding about 7v to the circuit - current dropped from 1400ma to 1050ma.

So wouldn't I just be better off with the 1050ma CC driver?

I guess my confusion stems from wanting to know if you dim the current on the 1400ma driver, does that increase the voltage it can handle? Because as you mentioned earlier, 1250ma per string would be good. But it seems the only way to get that is to use a CV driver.
 
ledgardener

ledgardener

Well-Known Member
Prawn Connery said:
I read that, which is why I asked. When he added a 5ohm resistor - I think, the equivalent of adding about 7v to the circuit - current dropped from 1400ma to 1050ma.

So wouldn't I just be better off with the 1050ma CC driver?

I guess my confusion stems from wanting to know if you dim the current on the 1400ma driver, does that increase the voltage it can handle? Because as you mentioned earlier, 1250ma per string would be good. But it seems the only way to get that is to use a CV driver.
Click to expand...
If you dim the current on a driver, it doesn’t increase the max voltage output of the driver, but by dimming and putting less current through the diodes, the voltage drop across them decreases and may bring the total system voltage drop down into the constant current range of the driver. In this case, it would still not be enough to make a 240H work.

If you want to drive at 1250mA per strip (139mA per diode), using the Samsung calculator, we can determine that at this current, each diode will have 2.97V across it. Therefor, at this current, your strip voltage will be about 47.5V. Four in series will be 190V. You’ll need an HLG-320H-C1400 to run at this current (just dim it down a touch), as the 240H will not be capable.
 
Prawn Connery

Prawn Connery

Well-Known Member
Thanks mate - that's just what I wanted to read. Really enjoy your site and videos, BTW - I've learned a lot from them.

OK, the tribe has spoken - I just ordered the HLG-240H-48A

Did some more reading and discovered the H series can be run in constant current mode or constant voltage - so best of both worlds - and the 48A will let me run about 200w dimmed with a bit of room to grow. Went with the A model as once I adjust it, I'll probably leave it. Also went with the HLG over the ELG because there's only a few $ difference and the HLG is slightly more efficient (93.5 vs 93), has a 7-year vs 5-year warranty, better pot adjustment position, and has an earth wire (ELG none, apparently), according to this thread: https://www.rollitup.org/t/anybody-tried-the-elg-version-of-meanwells-rather-than-hlg.935844/

Thanks to everyone who took the time to reply.

nc208, I know you recommended this driver early on, but I got there in the end :oops:
 
haze010

haze010

Well-Known Member
nfhiggs said:
The 2" wide double row F-series in 560 mm length have 144 diodes each (16s x 9p). I think you are looking at the single row H series which have 48 diodes each. This is why the F series was so eagerly anticipated when they were announced - 50% higher diode count at a lower price point, plus double row option, which the H series did not offer.
Click to expand...
This is just incorrect. I Built two lights each with 8 h-series strips about 6 months ago before F series was even announced. I paid $9.25 per H series strip for the 2 foot ones with 48 dioides. The F series are 50% more diodes, but to say they are a lower price point is just plain incorrect.
 
You must log in or register to reply here.
Top