DIY 1000 Watt HPS Replacement

Maybe i’m wrong or maybe i’m right. You can keep researching it till you get the answer you want. Or build it and find out.

I can’t find anything about molex, but I can find information on PCB terminal blocks. The look almost identical to the push-in wire connections on the Solstrips.

It seems they’re rated to 400v, but it may vary by specific type and I may be looking at the completely wrong component.

Only in case of heating effect, temperature varies in accordance to change in voltage or current. According to Joule’s law, heat dissipated while conduction of current through a wire is directly proportional to current or voltage. So with increase in current/voltage, heat dissipation will increase, raising the temperature of wire. Again there are materials where change in temperature increases resistance, that in turn increases temperature as resistance is also directly related to heating(by Joules’s law). Also keeping resistance and current/voltage constant, the time duration for which wire conducts also influences heat dissipation. Hence there are several cases where temperature is influenced whether directly by current/voltage or indirectly.

Well I think that answers part of the ? Voltage does effect heat so that means there is a limit and i’m 99% sure what your suggesting is bad mmkay

1859065.pdf (2.4 MB)

Here’s a technical datasheet on the PCB terminal blocks. It seems there’s different ratings, from 60v to 630v. I remember Growmau5 said something about a 300v rating for QBs, which uses similar components, but it may have been in regards to wire, not connectors. There is a connector rated to 320v.

Yup your right you should do it :unamused: better?

JFC, I came asking for help. I’m being forthright with my ignorance here. Like I said, I DON’T want to fuck up over an oversight or miscalculation. I’m taking your advice and researching rather than just assuming it’s safe.

I care about the data. It’s part OCD, part curiosity. I’d rather have more information than less.

Two quick points.

Heat losses are actually a function of the current squared, but its linear with voltage. So if you double the voltage, assuming the current remains the same, the heat loss doubles. However, if you double the current, again with the voltage remaining constant, the heat loss is 4 times as much.

BUT - that all assumes circuits with constant, fixed resistance. LED’s do not operate that way. There is a non-linear relationship between voltage and current within an LED. In the case of the LM561C chips, its governed by this graph.

Still, its close enough for estimating to just assume your heat losses will vary as the square of the change in current.

However, you cannot use changes in voltage to guesstimate heat loss changes with LED’s, because their resistance is not constant, plus these drivers are ALL constant current drivers - no matter how they are labeled.

On that last point - these drivers are ALL CC drivers. Despite the fact that the labeling shows some of them as CVCC, and others as CCCV, they are really all CC drivers. You can change the voltage on either type, which will change the current flowing in the circuit - BUT - the current flow will be dictated by that graph above. Thats because they are actually CC drivers with a voltage control and will always operate in CC mode in once the voltage is stable. They cant operate any other way or the LED’s would go into thermal runaway.

If you try to adjust the voltage outside the range the LED’s can handle one of two things will happen. If you go too low, the power supply will flicker as it tries to decide between the voltage selected and what the LED’s will accept and the current adjustment setting. If you go too hi with the voltage adjustment, it is possible to over drive the LEDs, but ONLY if you have the current setting higher than the max rating for the chips. As long as you have the current set below the max rating for the LED’s, they cannot be over driven no matter where you turn the voltage knob.

Here is the graph showing how these drivers work and the constant current zone.

In short, you are not at risk with either type. Which you decide to use depends on how you wire the strips up and if you can get the combination of voltages/current you need for your number of strips from a given driver.

It also matters how comfortable you are with hi voltages vrs hi currents. Hi voltage setups can kill you from electrocution, but poorly wired hi current setups can start fires due to heating in undersized wiring. Either one needs to be done correctly and safely.

With hi voltage setups, you dont need to worry much about voltage drop in small wires. At low currents, there is little voltage drop. At hi currents, voltage drop does become an issue. So if your doing an all parallel setup, larger wires are the smart way to go. In an all series setup, smaller ga wires will work just fine as long as they are sized for the max current flow you expect.

3 Likes

I forgot to put in the formulas to calc heat loss.

Power lost = Volts x Current or P=I x E In this case E is the voltage drop across the load, which may or may not equal the supply voltage.

Volts = Current x Resistance or E = I x R

If you substitute for E in the first equation, you get P = I x I x R

So actually, voltage has nothing to do with heat losses. Its all resistance and current with current being the dominating factor.

3 Likes

@Intersect Hi dude :slight_smile: this Just poped out on my YouTube :slight_smile:

Check it out. It might be what you are looking for.

2 Likes

Hi, I would stick to 3W Laser LED’s preferably Osrams, those do a quite good job in my eyes in replacing HPS.

Bt you have to mix the Spectrum in a way Sunlight gets replaced properly. It coul have a little more of the 600-700NM spectrum for Bloom, my plants love this.

1 Like

This is an example for 3W laserLED’S for commercials:

Sanlight Austria does a pretty good job.

@Fisch That dimming controller is fantastic. Looks like I may need to swap out the analog potentiometer in my design for an Amalech, especially if they’re designed for MeanWell dimming AND can be used as a timer.

@anon32470837 thank you for your insight. That aligns with what I was already suspecting about current, voltage, and heat. I’d rather operate at a fixed current and avoid overcurrent problems, fires are no bueno. I’m less scared of voltage because I’m not going to be fiddling with the lamp when it’s plugged in and powered up, so risk of shock can be mitigated on my end. I’m thinking series wiring to get the full current to each strip. I suppose I’d prefer to run a CCCV over a CVCC because the amperage is fixed across the entire system.

Also, I wouldn’t tell to much about how genius I am, when I didn’t do anything in Life. Proof is in the Seed and Plant.

1 Like

@Illyssian I would love to eventually run supplemental 660nm diodes, but first I want to ensure the light is solidly built. Supplemental spectra just aren’t in my budget yet.

And I’m no genius. I’m just a student trying to learn something. If I’m wrong, feel free to correct me. I’d prefer to make my mistakes now, rather than break something expensive because I’m a noob.

Sorry, If I did offend, but your pic says your a genius so this was a starting point for me.

As for LED Lights, I did build some and I have to tell you, the new horticulture series of osram (german company), did a really great job. If you mix your lights by 450NM, 660NM and Far red 730+ there is not much which can go wrong!

Be well my Brother.

2 Likes

My pic was supposed to be a tongue-in-cheek Looney Tunes reference, because Wile E Coyote thinks he’s a genius with it all figured out, yet is constantly foiled by his own hubris and is an abject failure. Also the “have brain will travel” joke is pretty funny, as is the idea of Wile E Coyote having a business card at all.

Point is, I’m not gonna go buy an ACME 1000000W LED LAMP just to have it blow up in my face while I try to catch the elusive Budrunner.

I want to do things right. Preferably sans explosions, electrocutions, or burns.

1 Like

The marketing material and labeling of the MeanWell drivers is confusing.

BOTH types of drivers control/regulate the current as their primary function, so the current is always ‘fixed’ or limited by the adjustment pot. The current can never go above the limit set by the internal pot or external controller.
BOTH types of drivers can over drive the LED’s if the adjustment pots are set incorrectly and the drivers are over sized for the load.
BOTH types are first and foremost “constant current” or CC power supplies. The CV aspect is secondary, no mater what the label says as far as CCCV vrs CVCC.

BOTH work exactly the same way. The only difference is that the ‘CC’ types are hi voltage/low current while the ‘CV’ types are hi current/low voltage. They both function exactly the same way.

The way they work is that the voltage pot, and the current pot, establish a max value when set. The driver will do its best to keep the voltage and current at LESS THAN or EQUAL TO those set values. However, there are some subtleties, and confusing aspects to how they work.

Lets say you have the voltage pot set to 10 volts and the current pot set to 1 amps and you have 3ea 3 volt LED’s in series - just to have some numbers to play with.

The first thing to know is that the voltage will NOT be determined ONLY by the voltage setting. It will also be determined by the number of series LED’s in the circuit, but limited by the voltage setting. In other words, these drivers will always set the output voltage high enough to drive the LED.s - no matter what you have the voltage set to. See that last graph I posted above. It says “In the constant current region, the highest voltage at output of the driver depends on the configuration of the end systems.” What that means is that the minimum voltage is set by the number of series LED’s with the voltage adjustment pot being sort of a guide line that will not be exceeded.

The current adjustment works in a similar way. The adjustment pot sets a MAX value that will never be exceeded, but the minimum is always going to be enough to drive the LED’s - no matter how far down you turn either one of the adjustment pots.

So - lets go back to the example. If you start out with the voltage set to 10 volts and amps set to 1 amp, the driver - either type - will do its best to maintain 10 volts at 1 amp on the output no matter what the load does. Now, plug in your 3 led string. If no current is flowing, because the voltage is less than the minimum forward voltage for the LED, then the driver will increase voltage until the max current is reached OR the max voltage is reached - which ever comes first. BUT - the number of series LED’s also plays into that.

Grrrr. I have to run do chores. More later… hopefully I can come up with a better way to explain this…

1 Like

SolStrips use the Molex 1041880210 connector. It’s rated for 500v DC and 9 amps with 18 AWG solid wire. The spec is derated significantly with stranded or thinner wire.

Molex connector PS-104188-001.pdf (723.9 KB)

2 Likes

That’s incredibly helpful @anon32470837. I may need to read it a few more times before fully grasping what you’re saying, but essentially there’s not much difference between the drivers and it would be best to get a proper sized driver, like the 320h/480h-24 rather than an oversized one like the 480h-c1750. I wasn’t sure that the pots on the driver regulated current, but the spec sheet for the 320h-24a says it’s got both Vo and Io trimming, so now I can see how that makes the type A the more desirable option.

Does anyone know if the 3-in-1 dimming on the Type B drivers regulates drive current, too? Or is it just the built-in pots that can do that?

1 Like