Personal computing discussed

Moderators: askfranklin, renee, emkubed, Captain Ned

 
DPete27
Grand Gerbil Poohbah
Topic Author
Posts: 3776
Joined: Wed Jan 26, 2011 12:50 pm
Location: Wisconsin, USA

LED Light for Snowblower

Thu Nov 14, 2019 10:18 am

Wanted to post this here since I know there's gerbils knowledgeable enough to confirm. Looking for some parts links to mouser or digikey.

I'm putting an LED light on my snowblower. Here's the light I bought. The snowblower generates electricity from a strator inside the flywheel. There's a single [positive] wire coming out of it. Negative is grounded to frame.

I found this website which details using:
  • a "full wave" bridge rectifier - Any requirements here? I see a lot of people using 50A/100V ones, but since most people are reporting 12-18V coming from the strator and are saying a maximum of about 27W can be generated, is that really necessary? I may see if I can locate a multi-meter to confirm the Voltage and Amps.
  • two - 2200uF Capacitors in parallel between the rectifier and the light. Why 2 in parallel? Is this the right capacity?
  • Fuse - is this really necessary?

Here's a video

The results seem a bit janky. Want to keep this as clean and legit as possible. Maybe there's a small, mountable(?) breadboard/PCB I could use to clean up the Capacitor connections?
Main: i5-3570K, ASRock Z77 Pro4-M, MSI RX480 8G, 500GB Crucial BX100, 2 TB Samsung EcoGreen F4, 16GB 1600MHz G.Skill @1.25V, EVGA 550-G2, Silverstone PS07B
HTPC: A8-5600K, MSI FM2-A75IA-E53, 4TB Seagate SSHD, 8GB 1866MHz G.Skill, Crosley D-25 Case Mod
 
roncat
Gerbil First Class
Posts: 148
Joined: Wed Dec 14, 2016 8:29 am

Re: LED Light for Snowblower

Thu Nov 14, 2019 11:35 am

DPete27 wrote:
Wanted to post this here since I know there's gerbils knowledgeable enough to confirm. Looking for some parts links to mouser or digikey.

I'm putting an LED light on my snowblower. Here's the light I bought. The snowblower generates electricity from a strator inside the flywheel. There's a single [positive] wire coming out of it. Negative is grounded to frame.

I found this website which details using:
  • a "full wave" bridge rectifier - Any requirements here? I see a lot of people using 50A/100V ones, but since most people are reporting 12-18V coming from the strator and are saying a maximum of about 27W can be generated, is that really necessary? I may see if I can locate a multi-meter to confirm the Voltage and Amps.
  • two - 2200uF Capacitors in parallel between the rectifier and the light. Why 2 in parallel? Is this the right capacity?
  • Fuse - is this really necessary?

Here's a video


The results seem a bit janky. Want to keep this as clean and legit as possible. Maybe there's a small, mountable(?) breadboard/PCB I could use to clean up the Capacitor connections?


I recommend a flashlight and duct tape, but anyway...

1) The 18vac is prolly rms, plus inrush, so the bridge size seems about right.
2) Don't see why you wouldn't attach the capacitor right to the bridge. Capacitors add in parallel, so 2 in parallel would give 4400uF. Or you could just get a single larger cap.
3) A fuse stops you from cussing. You will likely accidentally short your power source to ground somewhere in this process, thereby destroying the electrical system of your snowblower. I assume you will cuss at this point if you have to toss your snowblower in the garbage vs replacing a fuse.

Have fun...
 
just brew it!
Administrator
Posts: 54410
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: LED Light for Snowblower

Thu Nov 14, 2019 12:25 pm

DPete27 wrote:
a "full wave" bridge rectifier - Any requirements here? I see a lot of people using 50A/100V ones, but since most people are reporting 12-18V coming from the strator and are saying a maximum of about 27W can be generated, is that really necessary? I may see if I can locate a multi-meter to confirm the Voltage and Amps.

Keep in mind that when dealing with AC, the peak amplitude (which is what the diodes in the bridge need to be rated for) is higher than the (quoted) RMS voltage by roughly a factor of 1.4. A 50V bridge at lower current rating would likely be fine, but isn't going to save you a lot of money.

DPete27 wrote:
two - 2200uF Capacitors in parallel between the rectifier and the light. Why 2 in parallel? Is this the right capacity?

There isn't really a single "right" capacity here; it depends on the current draw of the light and how much you care about flicker. For a given light, higher capacity will result in less flicker. Broadly speaking, a pair of 2200s in parallel should be electrically equivalent to a single 4400 from the same product line. The internal resistance will likely be lower with the paired 2200s making things slightly more efficient, but I'd be very surprised if you would notice a difference in practice. Go with whatever ends up costing less or fits better into your chosen enclosure, and get something rated for at least 50V so you have some safety margin.

DPete27 wrote:
Fuse - is this really necessary?

What roncat said!

DPete27 wrote:
The results seem a bit janky. Want to keep this as clean and legit as possible. Maybe there's a small, mountable(?) breadboard/PCB I could use to clean up the Capacitor connections?

The crucial thing is to make sure your electrical connections won't get wet.

Something else you may need to consider - if you do get 18VAC out of the generator, once rectified and filtered that's going to be around 25VDC. This will sag some under load, but could still be substantially higher than the 12V the light is rated for. I didn't wade through the whole thread or watch the video - is there any sort of voltage/current limiting in the circuit (e.g. a series resistor)? Otherwise you may burn the light out fairly quickly.

At the end of the day, you may end up wishing you'd taken roncat's flashlight-and-duct-tape advice. :wink:
Nostalgia isn't what it used to be.
 
Waco
Grand Gerbil Poohbah
Posts: 3647
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: LED Light for Snowblower

Thu Nov 14, 2019 1:00 pm

just brew it! wrote:
Something else you may need to consider - if you do get 18VAC out of the generator, once rectified and filtered that's going to be around 25VDC.

For a full-wave rectifier wouldn't the DC output voltage be a little less than the 18 VRMS input? I don't recall the exact math off the top of my head, but I generally figure you'll get about 90% of the input RMS voltage when rectifying AC. I'd bet that 12v light can handle 16v pretty handily (assuming the 18v AC is constant, I'd bet it's more 16v or so at normal running RPM).
Desktop: X570 Gaming X | 3900X | 32 GB | Eisblock Radeon VII | Heatkiller R3 | 4K 40" | 1 TB SX8200 Pro + 2 TB 660p + 2 TB SATA SSD
NAS: 1950X | Designare EX | 32 GB ECC | 7x8 TB RAIDZ2 | 8x2 TB RAID10 | FreeNAS | ZFS | Dual LSI SAS
 
superjawes
Minister of Gerbil Affairs
Posts: 2420
Joined: Thu May 28, 2009 9:49 am

Re: LED Light for Snowblower

Thu Nov 14, 2019 1:25 pm

Duck tape and a flashlight? No nerds vouching for headlamps?

Like JBI said, parallel capacitors add up, so 2 x 2200uF = 4400uF (4 x 1100uF would also work if you're feeling cheeky). These are necessary if you want to combat flickering. Doing a full wave rectifier gets you most of the way from AC to DC, but the zero crossing points would still go down to zero and would "turn off" the LED. The capacitors will "resist" the fall down to zero, giving you a rippling pseudo-DC voltage and keeping the LED on. If you want to get even better (and more controlled intensity) you would add in a voltage regulator to eliminate the rippling, but that seems entirely optional if other people are getting "good enough" without it.

And yeah, you want fuses...unless a circuit breaker is possible. Either way, you want something to break the circuit before you burn anything. The right circuit breaker could also serve as an on/off switch.

Waco wrote:
just brew it! wrote:
Something else you may need to consider - if you do get 18VAC out of the generator, once rectified and filtered that's going to be around 25VDC.

For a full-wave rectifier wouldn't the DC output voltage be a little less than the 18 VRMS input? I don't recall the exact math off the top of my head, but I generally figure you'll get about 90% of the input RMS voltage when rectifying AC. I'd bet that 12v light can handle 16v pretty handily (assuming the 18v AC is constant, I'd bet it's more 16v or so at normal running RPM).
VAC_RMS = V(peak)/sqrt(2). A bridge rectifier is going to give an output based on the peak voltage, so JBI is correct that 18VAC (RMS) ~= 25VDC (before conditioning and ignoring diode forward voltages).

That 90% figure sounds more like the efficiency AC to DC power supply. /opinion
On second thought, let's not go to TechReport. Tis a silly place.
 
Heiwashin
Maximum Gerbil
Posts: 4815
Joined: Wed Dec 13, 2006 1:21 pm
Location: Denham Springs, LA

Re: LED Light for Snowblower

Thu Nov 14, 2019 1:38 pm

Looking for Knowledge wrote:
When drunk.....
I want to have sex, but find I am more likely to be shot down than when I am sober.
 
DPete27
Grand Gerbil Poohbah
Topic Author
Posts: 3776
Joined: Wed Jan 26, 2011 12:50 pm
Location: Wisconsin, USA

Re: LED Light for Snowblower

Thu Nov 14, 2019 2:03 pm

I have a headlamp. But this is much cooler! Also not interested in battery operation, same reason.

Lots of great advice so far. Tell me about this voltage regulator..! Could this be a challenge since the strator output voltage has some variability?

I like the single larger capacitor from a wiring standpoint. I can't imagine heat dissipation is going to be an issue outside in the winter. Any reasonable(safe) limitations if bigger is always better? I notice smaller capacitors have lower ripple. Is that why you'd put two smaller in parallel as opposed to one larger one?

Still not sure on the fuse requirement though. If the system is pushing through an (assumedly low resistance) LED anyway, isn't it already basically shorted? Can it really be that destructive if the wire(s) fall off somewhere and ground in front of the rectifier even? Not sure what I'd fry on the snowblower really. The only other electrical element is the spark plug which is fed by a separate strator outside the flywheel.

As you can tell (as stated in OP) I'm not that knowledgeable in this stuff, so I kinda just need a shopping list to go buy.
Last edited by DPete27 on Thu Nov 14, 2019 2:49 pm, edited 3 times in total.
Main: i5-3570K, ASRock Z77 Pro4-M, MSI RX480 8G, 500GB Crucial BX100, 2 TB Samsung EcoGreen F4, 16GB 1600MHz G.Skill @1.25V, EVGA 550-G2, Silverstone PS07B
HTPC: A8-5600K, MSI FM2-A75IA-E53, 4TB Seagate SSHD, 8GB 1866MHz G.Skill, Crosley D-25 Case Mod
 
roncat
Gerbil First Class
Posts: 148
Joined: Wed Dec 14, 2016 8:29 am

Re: LED Light for Snowblower

Thu Nov 14, 2019 2:27 pm

superjawes wrote:
Duck tape and a flashlight? No nerds vouching for headlamps?


That headlamp definitely has a high nerd factor. If the flashlight is not redneck enough, back your car into the garage and turn the headlights on. Turn the car off of course, don't be a deadneck.
 
Waco
Grand Gerbil Poohbah
Posts: 3647
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: LED Light for Snowblower

Thu Nov 14, 2019 2:43 pm

superjawes wrote:
VAC_RMS = V(peak)/sqrt(2). A bridge rectifier is going to give an output based on the peak voltage, so JBI is correct that 18VAC (RMS) ~= 25VDC (before conditioning and ignoring diode forward voltages).

That 90% figure sounds more like the efficiency AC to DC power supply. /opinion

It's been a while, but after consulting Wikipedia, I'm pretty sure it's closer to my 16V figure -

Vdc = Vavg = (2*Vpeak)/pi

Am I just being stupid?
Desktop: X570 Gaming X | 3900X | 32 GB | Eisblock Radeon VII | Heatkiller R3 | 4K 40" | 1 TB SX8200 Pro + 2 TB 660p + 2 TB SATA SSD
NAS: 1950X | Designare EX | 32 GB ECC | 7x8 TB RAIDZ2 | 8x2 TB RAID10 | FreeNAS | ZFS | Dual LSI SAS
 
superjawes
Minister of Gerbil Affairs
Posts: 2420
Joined: Thu May 28, 2009 9:49 am

Re: LED Light for Snowblower

Thu Nov 14, 2019 3:12 pm

Waco wrote:
superjawes wrote:
VAC_RMS = V(peak)/sqrt(2). A bridge rectifier is going to give an output based on the peak voltage, so JBI is correct that 18VAC (RMS) ~= 25VDC (before conditioning and ignoring diode forward voltages).

That 90% figure sounds more like the efficiency AC to DC power supply. /opinion

It's been a while, but after consulting Wikipedia, I'm pretty sure it's closer to my 16V figure -

Vdc = Vavg = (2*Vpeak)/pi

Am I just being stupid?
Kinda? :lol: j/k

Check it here.

The "Average" voltage of an AC source is zero because of that alternating bit. Equivalent peaks at positive and negative voltage cancel out, and that's why we use RMS instead of "Average" voltage for Vdc comparisons. When you have the wave's peak value, you would just divide that by the square root of 2 to get RMS, but we're using a full-wave rectifier, so we need to go backwards and get the peak voltage, so Vpeak = Vrms*sqrt(2). Plugging in numbers, [18 VAC (RMS)]*[1.414] ~= 25.5V (peak). Then the capacitors and/or regulators go to work smoothing that out to get closer and closer to a DC value.
On second thought, let's not go to TechReport. Tis a silly place.
 
just brew it!
Administrator
Posts: 54410
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: LED Light for Snowblower

Thu Nov 14, 2019 3:46 pm

Waco wrote:
superjawes wrote:
VAC_RMS = V(peak)/sqrt(2). A bridge rectifier is going to give an output based on the peak voltage, so JBI is correct that 18VAC (RMS) ~= 25VDC (before conditioning and ignoring diode forward voltages).

That 90% figure sounds more like the efficiency AC to DC power supply. /opinion

It's been a while, but after consulting Wikipedia, I'm pretty sure it's closer to my 16V figure -

Vdc = Vavg = (2*Vpeak)/pi

Am I just being stupid?

You're looking at the average voltage. With the filter caps you'll see closer to the peak voltage, since the caps charge up. Unless the load is high enough that the caps discharge appreciably between peaks, the DC voltage will be higher than the RMS AC voltage. And if the caps are discharging that much, there's not much point in having them in the first place (i.e. you need larger caps if you want to avoid flicker).

An interesting consequence of this is that the issue could potentially be addressed by leaving the caps out entirely, provided the thing spins fast enough that the flicker isn't objectionable.
Nostalgia isn't what it used to be.
 
Waco
Grand Gerbil Poohbah
Posts: 3647
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: LED Light for Snowblower

Thu Nov 14, 2019 4:48 pm

Durf. Thanks. I knew I was missing something; I should drink more coffee before posting.

Two lights in series would mitigate any voltage-related concerns. I'd worry that nearly 26 volts peak would damage the LEDs. I'd guess it probably runs in the upper 6k RPM range if it's a small two stroke, but at lower speeds, it would flicker like crazy without capacitors. 6000 RPM is roughly 100 Hz.

superjawes wrote:
The "Average" voltage of an AC source is zero because of that alternating bit.

Not after rectifying. :P I get where I went wrong though.
Desktop: X570 Gaming X | 3900X | 32 GB | Eisblock Radeon VII | Heatkiller R3 | 4K 40" | 1 TB SX8200 Pro + 2 TB 660p + 2 TB SATA SSD
NAS: 1950X | Designare EX | 32 GB ECC | 7x8 TB RAIDZ2 | 8x2 TB RAID10 | FreeNAS | ZFS | Dual LSI SAS
 
just brew it!
Administrator
Posts: 54410
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: LED Light for Snowblower

Thu Nov 14, 2019 5:25 pm

Waco wrote:
Two lights in series would mitigate any voltage-related concerns. I'd worry that nearly 26 volts peak would damage the LEDs.

Yeah, it's all gonna depend on how much voltage he actually gets.

Waco wrote:
I'd guess it probably runs in the upper 6k RPM range if it's a small two stroke, but at lower speeds, it would flicker like crazy without capacitors. 6000 RPM is roughly 100 Hz.

Since it is a full-wave rectifier the flicker will actually be at 2x the AC frequency, which helps.
Nostalgia isn't what it used to be.
 
Wirko
Gerbil Team Leader
Posts: 278
Joined: Fri Jun 15, 2007 4:38 am
Location: Central Europe

Re: LED Light for Snowblower

Thu Nov 14, 2019 5:32 pm

DPete27 wrote:
I have a headlamp. But this is much cooler! Also not interested in battery operation, same reason.


You can have it all and be more cooler than everyone else in the middle of Wisconsin winter. Don't use batteries, run two wires from the snow blower to your headlamp instead!

More seriously, you may find out it's best to use a headlamp in addition to the light on the snow blower. You're looking ahead most of the time but you also often need to look sideways, and a fixed light doesn't cast much light there.

Where are you going to put the electronic components? A small waterproof junction box should be best for that.

Edit: A breadboard is for prototyping in an electronics lab; in your case, I think a screw terminal block is a better option to reliably connect the components, some of which might be quite large.

Breadboard: https://www.ebay.com/itm/400-Points-Solderless-Breadboard-Protoboard-PCB-Test-Tafel/303104250905
Screw terminal blocks: https://www.ebay.com/itm/25A-6-Position-Dual-Row-Screw-Terminal-Barrier-Block-P5L7/173475958684 and https://www.ebay.com/itm/3x-12-Way-Barrier-Screw-Terminal-Block-Wire-Connection-Connector-Strip-3A-6A-10A/192425098700 , and there are other variants with solder terminals or solder + screw terminals.
Last edited by Wirko on Thu Nov 14, 2019 5:54 pm, edited 1 time in total.
 
Waco
Grand Gerbil Poohbah
Posts: 3647
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: LED Light for Snowblower

Thu Nov 14, 2019 5:49 pm

just brew it! wrote:
Since it is a full-wave rectifier the flicker will actually be at 2x the AC frequency, which helps.

Good point. It sounds like a typical running RPM for most snowblowers is 3500-4500 RPM, so it'd be somewhere north of 100 Hz, and that's likely not enough to bother anyone. http://www.bio-licht.org/02_resources/i ... s-1789.pdf
Desktop: X570 Gaming X | 3900X | 32 GB | Eisblock Radeon VII | Heatkiller R3 | 4K 40" | 1 TB SX8200 Pro + 2 TB 660p + 2 TB SATA SSD
NAS: 1950X | Designare EX | 32 GB ECC | 7x8 TB RAIDZ2 | 8x2 TB RAID10 | FreeNAS | ZFS | Dual LSI SAS
 
SecretSquirrel
Minister of Gerbil Affairs
Posts: 2722
Joined: Tue Jan 01, 2002 7:00 pm
Location: North DFW suburb...
Contact:

Re: LED Light for Snowblower

Thu Nov 14, 2019 7:21 pm

First questioin: does the snowblower already have lights? I live in Texas, so I don't know these things.

If so, what is the voltage rating on the replacement bubs? 12V or 24V. That will give you some clue.

Where does the "positive" wire go? I put it in quotes because if the voltage coming out is AC, then there isn't a positive and negative line. Depending on the design of a small gas engine, this line may actually be the ignition output, so it it just goes to the spark plug, then don't bother -- there would be 10-20kV running through it.

The make/model of the snowblower would be a highly useful thing here.

--SS
 
JustAnEngineer
Gerbil God
Posts: 19588
Joined: Sat Jan 26, 2002 7:00 pm
Location: The Heart of Dixie

Re: LED Light for Snowblower

Thu Nov 14, 2019 7:37 pm

The guys at our facility in Niagara Falls told me what they do when they retire. They sell the house and everything in it. Then they load up the snow blower in the back of the pickup truck and drive south. When they’re stopped somewhere and a local points at the back of the truck and asks “what the heck is that???” they know that they’ve driven far enough. :lol:
· R5-3600X, Liquid Freezer II 280, RoG Strix X570-E, 32GiB, RX5500XT, 2TB SX8200Pro +2TB MX500 +NAS, Define 7 Compact, Focus PX-850, S3220DGF +32UD99
· i7-9700K, NH-D15, Z390M Pro4, 32GiB, HD7770, ½TB 960Pro +5TB HD, Define Mini-C, SS-660XP2, U2410
 
DPete27
Grand Gerbil Poohbah
Topic Author
Posts: 3776
Joined: Wed Jan 26, 2011 12:50 pm
Location: Wisconsin, USA

Re: LED Light for Snowblower

Thu Nov 14, 2019 9:59 pm

The lead I'm connecting to is designed for an AC headlight, yes. It's an "optional" accessory that comes/ came with the higher end models, but for whatever reason, they still put the strator in on the lower end ones.
Unfortunately, nobody sells the actual headlight assemblies, or they're crazy expensive. Hence why people DIY.
Main: i5-3570K, ASRock Z77 Pro4-M, MSI RX480 8G, 500GB Crucial BX100, 2 TB Samsung EcoGreen F4, 16GB 1600MHz G.Skill @1.25V, EVGA 550-G2, Silverstone PS07B
HTPC: A8-5600K, MSI FM2-A75IA-E53, 4TB Seagate SSHD, 8GB 1866MHz G.Skill, Crosley D-25 Case Mod
 
just brew it!
Administrator
Posts: 54410
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: LED Light for Snowblower

Thu Nov 14, 2019 10:44 pm

DPete27 wrote:
The lead I'm connecting to is designed for an AC headlight, yes. It's an "optional" accessory that comes/ came with the higher end models, but for whatever reason, they still put the strator in on the lower end ones.
Unfortunately, nobody sells the actual headlight assemblies, or they're crazy expensive. Hence why people DIY.

Presumably the original accessory was an "old school" incandescent... in which case no rectifier would've been needed, and the thermal inertia of the filament would've smoothed out the flicker from the AC.

Was the light you selected something recommended on that forum you linked that people have tested with this mod on your (or similar) model of snowblower? I'm just trying to get a feel for how paranoid you need to be about potentially letting the magic smoke out of the light by over-volting it.

Ideally you'd measure how much voltage your rig is putting out, and measure the current draw of the light when it is connected to a known 12V source. Then we can figure out the value for a series resistor to keep things in spec, if your output voltage is appreciably above 12V.
Nostalgia isn't what it used to be.
 
roncat
Gerbil First Class
Posts: 148
Joined: Wed Dec 14, 2016 8:29 am

Re: LED Light for Snowblower

Fri Nov 15, 2019 8:04 am

DPete27 wrote:
I have a headlamp. But this is much cooler! Also not interested in battery operation, same reason.

Lots of great advice so far. Tell me about this voltage regulator..! Could this be a challenge since the strator output voltage has some variability?

I like the single larger capacitor from a wiring standpoint. I can't imagine heat dissipation is going to be an issue outside in the winter. Any reasonable(safe) limitations if bigger is always better? I notice smaller capacitors have lower ripple. Is that why you'd put two smaller in parallel as opposed to one larger one?

Still not sure on the fuse requirement though. If the system is pushing through an (assumedly low resistance) LED anyway, isn't it already basically shorted? Can it really be that destructive if the wire(s) fall off somewhere and ground in front of the rectifier even? Not sure what I'd fry on the snowblower really. The only other electrical element is the spark plug which is fed by a separate strator outside the flywheel.

As you can tell (as stated in OP) I'm not that knowledgeable in this stuff, so I kinda just need a shopping list to go buy.


Two 2200uF caps are prolly $1.50, one 4400uF is prolly $20. That's pretty much why. Also, at that size they are likely to be electrolytic, so you will know right away if they are installed backwards. Be very careful. If you are unsure if you installed them correctly, stay safely away at startup. They do not like to be biased the wrong way. I'm sure you can google some pretty impressive explosions when that happens.
 
DPete27
Grand Gerbil Poohbah
Topic Author
Posts: 3776
Joined: Wed Jan 26, 2011 12:50 pm
Location: Wisconsin, USA

Re: LED Light for Snowblower

Fri Nov 15, 2019 8:25 am

Yes, the stock headlights are typically incandescent/halogen so the lead just goes straight to the bulb.

The light I bought was $10. Not a huge deal if it gets smoked but I'd obviously like to avoid it if possible. So if there's other components i could/ should add to regulate voltage to a safe range for my light, I'll do that.
I bought the light because it's warm white color, cheap, and is a reasonable 800 lumens (as much light as your typical 60W equivalent home LED bulb)
Most people are using vehicle grade LED clusters like this but they're SUPER bright (because they're intended to light the road for driving) and have a high color temp.

How is capacitor ripple current selected? Does it refer to input tolerance, or output consistency? I noticed some 3300uF caps with a pretty low ripple, may just go with a single one of those if low ripple is desirable.

As far as the enclosure goes, the rectifier/capacitor/etc will be tucked up underneath the "console" plate between the handlebars. That will keep them well shielded from snow. I'll probably end up jimmying together a plastic enclose of some sort just to be safe, but I'm not worried about that.
Main: i5-3570K, ASRock Z77 Pro4-M, MSI RX480 8G, 500GB Crucial BX100, 2 TB Samsung EcoGreen F4, 16GB 1600MHz G.Skill @1.25V, EVGA 550-G2, Silverstone PS07B
HTPC: A8-5600K, MSI FM2-A75IA-E53, 4TB Seagate SSHD, 8GB 1866MHz G.Skill, Crosley D-25 Case Mod
 
SecretSquirrel
Minister of Gerbil Affairs
Posts: 2722
Joined: Tue Jan 01, 2002 7:00 pm
Location: North DFW suburb...
Contact:

Re: LED Light for Snowblower

Fri Nov 15, 2019 8:49 am

DPete27 wrote:
The lead I'm connecting to is designed for an AC headlight, yes. It's an "optional" accessory that comes/ came with the higher end models, but for whatever reason, they still put the strator in on the lower end ones.
Unfortunately, nobody sells the actual headlight assemblies, or they're crazy expensive. Hence why people DIY.


In that case, here is what I would probably do.

Run the AC through a bridge rectifier, filter the output with a sizeable capacitor, use this to feed a DC/DC converter set for 12V output. Use that to drive your light. Shove all of it into a waterproof project box.

As far as a fuse is concerned, in theory, one should be included. Not entirely sure its of a whole lot of use. Based on the expected current draw, I would put a 1.5A fuse between the AC output and the bridge rectifier. However, if you are correct that the blower is only capable of outputting around 30W, then its quite likely it can't output enough current to blow a 2A fuse. If you do use a fuse, you can do something like a waterproof inline fuse holder.

--SS
 
just brew it!
Administrator
Posts: 54410
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: LED Light for Snowblower

Fri Nov 15, 2019 9:32 am

I doubt the ripple current rating of the caps will matter in this context, I wouldn't worry about it.

If you'd like to avoid the added expense and complexity of a 12V step-down regulator, you will want to do 2 things:

1. Measure how much current the light draws when connected to a nominal 12V source (car battery or the 12V rail of a PC PSU). I suppose you could just trust the 10W spec on the Amazon listing, which would give you a current draw of 0.83A; but I'm not sure assuming the listing is correct would be a good idea...

2. Measure the voltage at the output of your rectifier circuit, with the caps installed. Make sure this is with the engine running at close to its maximum RPMs.

Then we can help you calculate what sort of series resistor you need to keep the voltage within the safe range for this light.
Nostalgia isn't what it used to be.
 
DPete27
Grand Gerbil Poohbah
Topic Author
Posts: 3776
Joined: Wed Jan 26, 2011 12:50 pm
Location: Wisconsin, USA

Re: LED Light for Snowblower

Fri Nov 15, 2019 10:02 am

I kinda figured a resistor was in the cards. My 1943 multimeter restoration isn't quite finished, so I'll try and locate a multimeter to test.

Here's what I'm looking at for parts order:
1 - 35V 4700uF Capacitor = $2.03
1 - 35A, 1000V Bridge Rectifier = $2.92

probably looking at something in the 10 Ohm resistance? So maybe something like this 10 Ohm 25W resistor = $1.09? Would be nice to save on shipping a single resistor after the inital order since shipping is crazy expensive (why?), but the only thing I can measure is the AC current and voltage from the strator until I get the rectifier ordered. Sounds like that's not going to be enough info to determine require resistance.
Main: i5-3570K, ASRock Z77 Pro4-M, MSI RX480 8G, 500GB Crucial BX100, 2 TB Samsung EcoGreen F4, 16GB 1600MHz G.Skill @1.25V, EVGA 550-G2, Silverstone PS07B
HTPC: A8-5600K, MSI FM2-A75IA-E53, 4TB Seagate SSHD, 8GB 1866MHz G.Skill, Crosley D-25 Case Mod
 
SecretSquirrel
Minister of Gerbil Affairs
Posts: 2722
Joined: Tue Jan 01, 2002 7:00 pm
Location: North DFW suburb...
Contact:

Re: LED Light for Snowblower

Fri Nov 15, 2019 10:22 am

just brew it! wrote:
I doubt the ripple current rating of the caps will matter in this context, I wouldn't worry about it.

If you'd like to avoid the added expense and complexity of a 12V step-down regulator, you will want to do 2 things:

1. Measure how much current the light draws when connected to a nominal 12V source (car battery or the 12V rail of a PC PSU). I suppose you could just trust the 10W spec on the Amazon listing, which would give you a current draw of 0.83A; but I'm not sure assuming the listing is correct would be a good idea...

2. Measure the voltage at the output of your rectifier circuit, with the caps installed. Make sure this is with the engine running at close to its maximum RPMs.

Then we can help you calculate what sort of series resistor you need to keep the voltage within the safe range for this light.


You really need to measure the smoothed output of the rectifier under a load equivalent of the light (approx .8A). Otherwise the voltage is going to be significantly higher than expected. Unloaded, it will approach peak AC voltage. Under load, it is going to be something less, depending on the size of the capacitors. Under load, as the capacitor size increases, the measured voltage will increase and the ripple will decrease. There will be a limit as the generator can only provide so much power to recharge the caps on top of running the light.

If you assume a 18V nominal, the you are going to need a 6ohm resistor in series with the lgiht. It will need to be rated at least 10W (twice its dissipation) or its going to get real hot. Not much cheaper than the DC/DC converter. If the nominal voltage is 15V, then you need a 3ohm resistor (you can use two of the 6ohms in parallel). One downside of the resistor is that it will potentially cause the light intensity to vary with the RPM of the engine. The rectified and smoothed output voltage will vary with engine RPM, and you have to size things for the max voltage. This means that at lower voltages, all the equations change. But, it's not guaranteed to vary the brightness. It will depend on how the LED light is constructed, how much the voltage varies, and how much the light varies it's draw over this range. The DC/DC converter removes all this variability. As long as the input voltage is above the minimum necessary to drive the converter (probably 13.5-14V for 12V output), it will output exactly 12V and the lamp will run at its 10W rating.

You've got a couple of options. You can just get the rectifier and capacitors and just directly connect the light and see what happens. I will probably work just fine. There is a risk you blow out the light. More probable is that you will just shorten its life span by some amount, however this is almost certainly going to be less of a reduction than the high vibration of the snow blower is going to cause. If you blow out the light, you'll have to replace the light plus buy a regulator or resistor. Or you can just buy the regulator or resistor up front. The former might cost you an extra $16 or so. The latter will definitely cost you the extra $5-$6 for the extra part.

--SS
 
just brew it!
Administrator
Posts: 54410
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: LED Light for Snowblower

Fri Nov 15, 2019 10:28 am

SecretSquirrel wrote:
You really need to measure the smoothed output of the rectifier under a load equivalent of the light (approx .8A). Otherwise the voltage is going to be significantly higher than expected. Unloaded, it will approach peak AC voltage. Under load, it is going to be something less, depending on the size of the capacitors. Under load, as the capacitor size increases, the measured voltage will increase and the ripple will decrease. There will be a limit as the generator can only provide so much power to recharge the caps on top of running the light.

I considered going into more detail, but figured in this case (generator designed to drive an incandescent being used to drive an LED instead) that the loaded voltage would likely be close enough to the peak unloaded voltage to not skew the calculations too much. And in any case, the error would be in the direction of over-estimating the voltage, which will result in slightly less light output, but will also avoid risking blowing out the light.
Nostalgia isn't what it used to be.
 
SecretSquirrel
Minister of Gerbil Affairs
Posts: 2722
Joined: Tue Jan 01, 2002 7:00 pm
Location: North DFW suburb...
Contact:

Re: LED Light for Snowblower

Fri Nov 15, 2019 10:38 am

DPete27 wrote:
I kinda figured a resistor was in the cards. My 1943 multimeter restoration isn't quite finished, so I'll try and locate a multimeter to test.

Here's what I'm looking at for parts order:
1 - 35V 4700uF Capacitor = $2.03
1 - 35A, 1000V Bridge Rectifier = $2.92

probably looking at something in the 10 Ohm resistance? So maybe something like this 10 Ohm 25W resistor = $1.09? Would be nice to save on shipping a single resistor after the inital order since shipping is crazy expensive (why?), but the only thing I can measure is the AC current and voltage from the strator until I get the rectifier ordered. Sounds like that's not going to be enough info to determine require resistance.


If you have Amazon Prime, the final cost is going to be cheaper, or if you aren't a Prime member and can stand the wait for free shipping. As you noted, shipping from Mouser on a small order is killer. Only when I just have to have it do I make a small order from Mouser. Even if I only need a couple of things, I'll generally odd to the order until it's around $50, just so the shipping isn't a killer. It's how I keep the parts bin stocked.

If you have an auto parts store nearby, you can go pick up a light bulb and connect it to the AC output and measure the current it draws, and the voltage across it. Most of them will list their wattage on the package. You want something that is 10-15W, and something that you can easily rig a connection to the terminals. Sylvania 921s, for example. They have folded wire terminals that you can straighten out for easy connection to a wire.

--SS
 
DPete27
Grand Gerbil Poohbah
Topic Author
Posts: 3776
Joined: Wed Jan 26, 2011 12:50 pm
Location: Wisconsin, USA

Re: LED Light for Snowblower

Fri Nov 15, 2019 12:19 pm

Sorry SS. Didn't see your earlier post about the DC/DC converter. I'd seen that component on browsing, and thought it seemed ideal, but I'm not sure how to select one. Thanks for the links though.
That's the DC/DC converter I'd need?

In case it wasn't apparent, the DC/DC converter is exactly the kind of spice I wanted to add to this recipe. Taking the rectifier + capacitor formula and add an inexpensive tweak to improve the result.

The ~10-12W Incandescent auto bulb was my plan to do this weekend with the multimeter. Seemed like the suggestions were getting to a point where no further detail could be added without knowing some actual measurements from my circuit. But it seems if I go with the DC/DC converter, I don't have to worry much about pre-testing since I'll dial in the converter to output what I need?

Amazon is pretty much already charging shipping in their regular prices (rectifier is $6 on Amazon even before shipping, and capacitor is $5) so the total pretty much comes out the same as mouser ($2 + $3 + $8 shipping).

On the rectifier. Is there any benefit of massively exceeding the Amperage of the circuit? I see SS linked a 6A rectifier compared to the 35A one I was eyeing. Is a "smaller" rectifier better or more accurate?
Main: i5-3570K, ASRock Z77 Pro4-M, MSI RX480 8G, 500GB Crucial BX100, 2 TB Samsung EcoGreen F4, 16GB 1600MHz G.Skill @1.25V, EVGA 550-G2, Silverstone PS07B
HTPC: A8-5600K, MSI FM2-A75IA-E53, 4TB Seagate SSHD, 8GB 1866MHz G.Skill, Crosley D-25 Case Mod
 
just brew it!
Administrator
Posts: 54410
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: LED Light for Snowblower

Fri Nov 15, 2019 1:51 pm

For basic silicon rectifiers/bridges lower current rating is typically just going to mean physically smaller and cheaper. As long as you have a safety margin of at least a factor of 2 you should be good to go. 35A is massive overkill in your case, since you're probably going to be dealing with current draw of around 1A.
Nostalgia isn't what it used to be.
 
SecretSquirrel
Minister of Gerbil Affairs
Posts: 2722
Joined: Tue Jan 01, 2002 7:00 pm
Location: North DFW suburb...
Contact:

Re: LED Light for Snowblower

Fri Nov 15, 2019 11:15 pm

DPete27 wrote:
Sorry SS. Didn't see your earlier post about the DC/DC converter. I'd seen that component on browsing, and thought it seemed ideal, but I'm not sure how to select one. Thanks for the links though.
That's the DC/DC converter I'd need?

In case it wasn't apparent, the DC/DC converter is exactly the kind of spice I wanted to add to this recipe. Taking the rectifier + capacitor formula and add an inexpensive tweak to improve the result.


I make no guarantees the whole thing won't go up in a flash and cloud of magic smoke. :o Never took the PE exam, so you can't blame me! :lol: Computer engineers weren't required to.

As far as picking the DC/DC converter, when I was looking, here are the parameters I was thinking about:

Load current of 1A (12W) based on the light. This means you want a converted rated at least 1A, I went for 2A as it will keep the temperature down and things will last longer. You might give up a little bit in efficiency, depending on the converter design, but it's going to be like 5 points of efficiency or something like that. Since you're dealing with cheap electronics, the extra de-rating is more important.

Input voltage somewhere between 12VAC and 24VAC, rectified and smoothed. Since the DC voltage peaks are going to be close to the peak voltage of the AC input, 24VAC RMS means a peak DC in the 33V range. 12VAC RMS would be around 16-17VDC. So a regulator that could accept at least 35V at the input.

Beyond that, I was looking at Amazon an there are a number a regulators that are cheaper per unit, but I figured you didn't need six, or ten. Having two gives you one to blow up in experimenting. :D Also, if you look at various types, some have little single turn trimmer pots on them. The wouldn't do so well in a high vibration environment and you would definitely need to fix them in position with glue or nail polish. The one I linked has a multi turn trimmer which will make it easier to dial in 12V and be a little less susceptible to vibration related movement, though you might still want to fix the trim screw once you have it set.

For what you are doing, the parameters aren't too tight and there are lots of options. The one I linked is probably what I would order unless I had use for several more, in which case I might look at some of the higher count multi-packs.

--SS

Who is online

Users browsing this forum: No registered users and 2 guests
GZIP: On