Trying to find a driver that can accept 7 to 15 volts DC that can run the MC-E at full rated output. A variable dimmer would be nice. Any drivers out there that have that? How could I configure it as I understand it has parallel, series etc?? Must run continously for 30 minutes minimum.
Driver for a MC-E
- Thread starter videoman
- Start date
Help Support Candle Power Flashlight Forum
Fichtenelch
Enlightened
Gryloc
Enlightened
I believe that it would be easy for you to drive the MC-E when wired in parallel if your voltage is going to vary that much. However, if you are certain that the input will be constant at a certain voltage, we may be able to wire the MC-E differently and use another variety of drivers. So below are some very nice drivers that will power the MC-E (or a P7) at its rated current if you wire all four dies in parallel:
From TaskLED, the HipCC and HipFlex will work great. The single-mode HipCC operates from 4.5V to 25V, and the HipFlex operates from 4.5V to 24V. The HipFlex has a really cool and smart user interface (UI) that gives you many options, from preset levels, to settable max current outputs. In addition, you can change the low voltage cutoff, auto-off time, and much more.
From the Sandwich Shoppe, you can choose from three different Shark Buck drivers. The Shark Buck operates from 4V to 24V (recommended by the creator in this thread). The Shark Buck contains a trim-pot so you can manually set the max current going to the LED (from nearly 0% to 100% of the max preset current). Three products are offered: each has a preset amount of current delivered to the LED(s): 1000mA, 2000mA, and 3000mA. You can remove this trimpot and add your own potentiometer (a variable resistor) that you can mount on the outside of the flashlight to change the output by the turning of a knob.
So, if you go with either of the three above drivers, you will not be disappointed by the quality and/or features. There are a couple of other drivers around here that may work as well, too. As for the KD or DX drivers, I have no clue on the quality of these drivers, or how consistent they are on output current. I do not know if they are as efficient as these above drivers. I also think they do not have a big enough input voltage range for your application (others please chime in). With the three drivers above, you can get a heck of a lot of help by members of the CPF, and the creators of the circuits (just ask for the member names). You might be on your own with the KD/DX drivers.
Finally, with either of the three drivers I listed above, they are typical buck circuits (they drop the input voltage to power the emitters at the driver's designed current level). This means that as long as the input voltage stays above the forward voltage of the LED emitters (you need 0.5V to 1V overhead), you can drive any amount of LEDs in series at ~2.8A per LED! So with a input voltage of 7V, you may be able to drive two MC-E or P7 emitters in series, and up to four emitters if the input voltage stays around 15V.
Cheers,
-Tony
From TaskLED, the HipCC and HipFlex will work great. The single-mode HipCC operates from 4.5V to 25V, and the HipFlex operates from 4.5V to 24V. The HipFlex has a really cool and smart user interface (UI) that gives you many options, from preset levels, to settable max current outputs. In addition, you can change the low voltage cutoff, auto-off time, and much more.
From the Sandwich Shoppe, you can choose from three different Shark Buck drivers. The Shark Buck operates from 4V to 24V (recommended by the creator in this thread). The Shark Buck contains a trim-pot so you can manually set the max current going to the LED (from nearly 0% to 100% of the max preset current). Three products are offered: each has a preset amount of current delivered to the LED(s): 1000mA, 2000mA, and 3000mA. You can remove this trimpot and add your own potentiometer (a variable resistor) that you can mount on the outside of the flashlight to change the output by the turning of a knob.
So, if you go with either of the three above drivers, you will not be disappointed by the quality and/or features. There are a couple of other drivers around here that may work as well, too. As for the KD or DX drivers, I have no clue on the quality of these drivers, or how consistent they are on output current. I do not know if they are as efficient as these above drivers. I also think they do not have a big enough input voltage range for your application (others please chime in). With the three drivers above, you can get a heck of a lot of help by members of the CPF, and the creators of the circuits (just ask for the member names). You might be on your own with the KD/DX drivers.
Finally, with either of the three drivers I listed above, they are typical buck circuits (they drop the input voltage to power the emitters at the driver's designed current level). This means that as long as the input voltage stays above the forward voltage of the LED emitters (you need 0.5V to 1V overhead), you can drive any amount of LEDs in series at ~2.8A per LED! So with a input voltage of 7V, you may be able to drive two MC-E or P7 emitters in series, and up to four emitters if the input voltage stays around 15V.
Cheers,
-Tony
Justin Case
Flashlight Enthusiast
- Joined
- Mar 19, 2008
- Messages
- 3,797
Just looking at the workmanship of that KD driver would steer me away. And examining the series of product photos shows the in-line component changes that seem typical for these inexpensive Chinese drivers. In one photo, you see two SS34 Schottky diodes. In another photo, some other single diode is used instead (I can't read the writing on the component). And the inductor has no markings on it to indicate the inductance value. I've run into these Chinese drivers that probably use undersized inductors, with the result that the driver doesn't run reliably throughout its rated voltage range. They go unstable at upper voltages (and before hitting the max rated voltage).
I prefer to get a known, supported, quality product from The Shoppe or TaskLED. But I'm also willing to pay the money for that level of product.
That said, you have other options besides the Shark Buck, hipFlex, and hipCC, if dimming isn't a hard requirement.
You could wire the MC-E in 2S2P and use an SOB1400 to run the LED at full power. The main downsides are that the small SOB driver will need good heat sinking, running in full regulation at 7V is probably not going to happen (my guess is that you'll need at least 8V in), and you probably won't find a 2S2P MC-E on a star (4S and 4P MC-Es on a star seem common).
For an inexpensive Chinese driver, the KD SKU S002982 can be easily modified for 1.4A output (the stock driver is 1A out to drive a 2S2P MC-E at full power (I'm actually using one of these boards for this purpose). However, my bench power supply measurements suggest that the board becomes flaky above 12V input. The nice thing is that the board goes into full regulation at 7V and runs at about 90% efficiency between 7V-12V.
To do the mod, remove resistor R1 that is next to the AX2002 buck IC -- it is marked R220 in the images on the KD web site -- and replace it with an R560. For the empty resistor location R2, solder an R250 resistor, which will be in parallel to resistor R1. The resultant sense resistor value is 0.173 ohms, giving nominally 1445mA out.
I think if you replace the stock 22 uH inductor with a 33uH (or higher) version, the driver will run stably to its rated Vin max.
I prefer to get a known, supported, quality product from The Shoppe or TaskLED. But I'm also willing to pay the money for that level of product.
That said, you have other options besides the Shark Buck, hipFlex, and hipCC, if dimming isn't a hard requirement.
You could wire the MC-E in 2S2P and use an SOB1400 to run the LED at full power. The main downsides are that the small SOB driver will need good heat sinking, running in full regulation at 7V is probably not going to happen (my guess is that you'll need at least 8V in), and you probably won't find a 2S2P MC-E on a star (4S and 4P MC-Es on a star seem common).
For an inexpensive Chinese driver, the KD SKU S002982 can be easily modified for 1.4A output (the stock driver is 1A out to drive a 2S2P MC-E at full power (I'm actually using one of these boards for this purpose). However, my bench power supply measurements suggest that the board becomes flaky above 12V input. The nice thing is that the board goes into full regulation at 7V and runs at about 90% efficiency between 7V-12V.
To do the mod, remove resistor R1 that is next to the AX2002 buck IC -- it is marked R220 in the images on the KD web site -- and replace it with an R560. For the empty resistor location R2, solder an R250 resistor, which will be in parallel to resistor R1. The resultant sense resistor value is 0.173 ohms, giving nominally 1445mA out.
I think if you replace the stock 22 uH inductor with a 33uH (or higher) version, the driver will run stably to its rated Vin max.
Wow, thanks for all that info, My project seems to be getting along nicely.I have a couple of questions but I would like you to know first that I have located a source that makes black aluninium enclosures that are well suited for heatsinking and heat dispersion. I will be using a case that measures 3 inches X 3 inches X 4 inches deep and also has fins on it for heat dissipating.. This case must contain 2 led's with each led having it's own driver and that each led have(if possible) it's own dimmer control mounted externally behind the case.This unit would be powered from various power sources (DC) and types of batteries ranging from 6 volts up to 15 volts DC. The light will be used as a on-camera high output light for video purposes. Price to make is no object, I want the most reliable drivers, best led's for the job, best potentiometers out there and anything else that might be necessary to accomplish this.The main criteria is that it has an equivalent light output of a 100 watt or more halogen bulb, meaning that it has to illuminate a subject at 30 feet away with a horizontal evenly lit angle of no less than,60 degrees.If the P7's are better let me know, reliability is paramount. Later if this project is successful, I would like a smaller unit using 1 led (MC-E or P7) with a built in 7.2VDC li-ion. Many thanks for your great support and suggestions.
I just read a thread on Custom and modified lights by Mick and he has a cave light that uses 3-P7"s in it with a hipflex driver.Wow, That's wat I'm talking about. He made this with a lathe and looks awesome.Now if that light can be dimmed with a rear mounted pot and those lenses can be floods. How reliable is the driver, considering that I will probably only need 2 MC-E's or P7's ?
Checked out the hipflex driver and looks good. however how can I put an external dimmer (pot) outside the case to it, and what value ohms. Also I saw the luxdrive driver that is also nice and can be ordered with the dimmer. Question is, which one of these is more reliable to drive either the MC-e or ssc-P7 ? With the dimmer, is the color temperature going to change as it is dimmed ? Thanks
Gryloc
Enlightened
Videoman,
The Hipflex has a great built in UI with several modes, but from what I read on the TaskLED's website, I do not see that it can be dimmable using an external potentiometer. The sense resistance is so low (less than a quarter of an ohm), that you will be unable to find a pot that even has a resistance that low (the ones I played with usually had a min of 1 ohm, and a max of 10 ohm, 10K ohm, 100K ohm, etc). Usually drivers that use very low resistance sense resistors are extremely sensitive to tiny changes in resistance, so any variable resistor that could work would be tricky to use. The same with the HipCC. These can be PWM dimmed, which means you may be able to make a very basic external circuit that will do this, or use a separate master driver. You can email the owner of TaskLED to see if there is certainly a way. He is very friendly and helpful.
The only driver that would work with a parallel wired MC-E that is fully dimmable is the Shark Buck. With this one, you can attach a standard logarithmic potentiometer (50K ohm?), and dim away. These are one of the drivers that have a separate part of the circuit that varies the current (dims from 0-100%) using much higher values of resistors (and therefore potentiometers). There is still a part of the circuit that uses a low value sense resistor (it sets max current), hence the 1A, 2A, and 3A versions.
As for the Luxdrive drivers (which can be dimmed on some models), I do not see any drivers capable of delivering over 1A to the LED. If you wire the emitter in parallel, you need 2.8A delivered to the MC-E to drive each die at 700mA, and 1.4A if wired in 2S2P. This would only work if you wire the MC-E in series (in which then a 700mA or 1000mA driver could work). However if you do that, your battery voltage would have to be greater than ~13.6V (assuming around 3.4V per die at 700mA).
As for reliability, any of these drivers are golden (shark buck, hip- series, and luxdrives). These should outlast the LED (maybe not completely, but you may want to swap emitters several times to stay up with technology many times before the driver may have issues). Even if the driver starts to mess up or fail, either dat2LED (for the Shark) or Georges80 (for the TaskLED) will take care of you!
EDIT: Oh, I forgot to answer the post before. All three drivers I mentioned before will reliably drive 2 or 3 MC-E or P7 emitters in series. Just make sure the input voltage is high enough so the driver has the headroom it needs to dish out the designed current. TaskLED's drivers do operate more efficiently with more series connected emitters (look at my previous links). The Shark Buck still operates at more than 77% efficiency, which is great. These high quality drivers will last longer if they are happy and efficient. Since these are designed to deliver high currents to LEDs, then they will do so reliably. It is when you force other drivers to deliver higher currents, by just changing a few components on them, is when you get into reliability or stability issues...
The Hipflex has a great built in UI with several modes, but from what I read on the TaskLED's website, I do not see that it can be dimmable using an external potentiometer. The sense resistance is so low (less than a quarter of an ohm), that you will be unable to find a pot that even has a resistance that low (the ones I played with usually had a min of 1 ohm, and a max of 10 ohm, 10K ohm, 100K ohm, etc). Usually drivers that use very low resistance sense resistors are extremely sensitive to tiny changes in resistance, so any variable resistor that could work would be tricky to use. The same with the HipCC. These can be PWM dimmed, which means you may be able to make a very basic external circuit that will do this, or use a separate master driver. You can email the owner of TaskLED to see if there is certainly a way. He is very friendly and helpful.
The only driver that would work with a parallel wired MC-E that is fully dimmable is the Shark Buck. With this one, you can attach a standard logarithmic potentiometer (50K ohm?), and dim away. These are one of the drivers that have a separate part of the circuit that varies the current (dims from 0-100%) using much higher values of resistors (and therefore potentiometers). There is still a part of the circuit that uses a low value sense resistor (it sets max current), hence the 1A, 2A, and 3A versions.
As for the Luxdrive drivers (which can be dimmed on some models), I do not see any drivers capable of delivering over 1A to the LED. If you wire the emitter in parallel, you need 2.8A delivered to the MC-E to drive each die at 700mA, and 1.4A if wired in 2S2P. This would only work if you wire the MC-E in series (in which then a 700mA or 1000mA driver could work). However if you do that, your battery voltage would have to be greater than ~13.6V (assuming around 3.4V per die at 700mA).
As for reliability, any of these drivers are golden (shark buck, hip- series, and luxdrives). These should outlast the LED (maybe not completely, but you may want to swap emitters several times to stay up with technology many times before the driver may have issues). Even if the driver starts to mess up or fail, either dat2LED (for the Shark) or Georges80 (for the TaskLED) will take care of you!
EDIT: Oh, I forgot to answer the post before. All three drivers I mentioned before will reliably drive 2 or 3 MC-E or P7 emitters in series. Just make sure the input voltage is high enough so the driver has the headroom it needs to dish out the designed current. TaskLED's drivers do operate more efficiently with more series connected emitters (look at my previous links). The Shark Buck still operates at more than 77% efficiency, which is great. These high quality drivers will last longer if they are happy and efficient. Since these are designed to deliver high currents to LEDs, then they will do so reliably. It is when you force other drivers to deliver higher currents, by just changing a few components on them, is when you get into reliability or stability issues...
Last edited:
Hi and thanks for the detailed info. The configuration I have in mind is an aluminium enclosure with ridged heat sinking all around that measures approximately 3 inches diameter by 4.5 inches long.Using 2 MC-e's and each one with it's own dedicated buck driver with dimming control.Later I would have same enclosure with 3 MC-E's and 3 bucks. The seating of the drivers can be on a 3 inch diameter X 1/2 inch thick aluminum stock.The buck's can also be mounted likewise near the back. Using the 2 driver configuration for now, can I wire the 2 bucks off the same battery? I have 2 battery options, one is a 7.2V DC li-ion, the other is a 12 volt li-ion.Having 2 dimmers mounted in the rear, the only buck I see is the Lux drive ones. Any thoughts on improving the set-up?
Gryloc
Enlightened
videoman,
Actually, if the battery voltage is high enough, you can wire two, maybe three, MC-E emitters (each wired in parallel) from a single driver. Those three drivers can deliver up to 2.8A, at any voltage necessary in order to keep delivering 2.8A, as long as the input voltage is higher than the combined Vf plus a bit more than 0.5V for the driver electronics overhead. This is how buck drivers work. They do not care what LED you put on its output, as long as they can handle the current, or if the Vf of the LEDs (at a certain current level) is not higher than the max voltage that the driver can output.
How many li-ion cells are in that "12V" battery? I wonder so we can determine the voltage range during the life cycle of the cells. So, if you use the 12V li-ion battery, then you can easily drive both MC-E emitters in series (where all four dies in each emitter are in parallel) from a single Shark Buck using a single potentiometer. With the 12V battery, it would be a waste to power each MC-E separately with their own driver. It would also seem to be easier to quickly dim the light output up or down when needed using a single potentiometer.
I guess I did not understand what you are asking about the two potentiometers and the Luxdrive drivers. Just to clarify, the Luxdrive drivers may not be suitable for your application if you are only using 12V (V in is not high enough to wire the four dies of the MC-E in series to drive them at 700mA).
If you use the 7.4V li-ion cell, you may be able to drive the two MC-E emitters in series like above, but only if the forward voltage is low enough (the Vf does vary per LED as well as how other characteristics vary due to the manufacturing processes). It would be the safest to drive each MC-E with their own drivers if you use the 7.4V li-ion battery, just in case your MC-E emitters have a high Vf.
If money is no object, might as well use three MC-E emitters with the 12V battery. You would only need one driver, and one potentiometer. You do not have to drive the emitters at their max current. LEDs become more efficient as the current decreases, so three emitters operating at, say, 20W would put out more light than two emitters operating at the same power. If you demand a certain max output, then three emitters will allow longer battery life than if you used two. Finally, if you really need the performance, you can crank it to the max (as long as the case can handle the heat).
I hope that I do not sound pushy, but I would like you to be the happiest with your project. Professionals need the best tools. Anyway, the Sandwich Shoppe is perfectly legit. I believe that it is nearly as old as the CPF, and practically created for us as far I know. I have ordered from them many times, and were happy with the service. They responded to my emails helping me with any issues or questions I might have had. I always thought it was weird that they do not have a number listed, too. Well, if you have any concerns, just ask around.
I hope what I said made sense. If not, just ask. Next time I need to start tossing in some numbers and math to clarify. These drivers can be set up so easily once you determine the LEDs used and how much current you want to send to the LED, how they are wired, and the type of power source.
-Tony
Actually, if the battery voltage is high enough, you can wire two, maybe three, MC-E emitters (each wired in parallel) from a single driver. Those three drivers can deliver up to 2.8A, at any voltage necessary in order to keep delivering 2.8A, as long as the input voltage is higher than the combined Vf plus a bit more than 0.5V for the driver electronics overhead. This is how buck drivers work. They do not care what LED you put on its output, as long as they can handle the current, or if the Vf of the LEDs (at a certain current level) is not higher than the max voltage that the driver can output.
How many li-ion cells are in that "12V" battery? I wonder so we can determine the voltage range during the life cycle of the cells. So, if you use the 12V li-ion battery, then you can easily drive both MC-E emitters in series (where all four dies in each emitter are in parallel) from a single Shark Buck using a single potentiometer. With the 12V battery, it would be a waste to power each MC-E separately with their own driver. It would also seem to be easier to quickly dim the light output up or down when needed using a single potentiometer.
I guess I did not understand what you are asking about the two potentiometers and the Luxdrive drivers. Just to clarify, the Luxdrive drivers may not be suitable for your application if you are only using 12V (V in is not high enough to wire the four dies of the MC-E in series to drive them at 700mA).
If you use the 7.4V li-ion cell, you may be able to drive the two MC-E emitters in series like above, but only if the forward voltage is low enough (the Vf does vary per LED as well as how other characteristics vary due to the manufacturing processes). It would be the safest to drive each MC-E with their own drivers if you use the 7.4V li-ion battery, just in case your MC-E emitters have a high Vf.
If money is no object, might as well use three MC-E emitters with the 12V battery. You would only need one driver, and one potentiometer. You do not have to drive the emitters at their max current. LEDs become more efficient as the current decreases, so three emitters operating at, say, 20W would put out more light than two emitters operating at the same power. If you demand a certain max output, then three emitters will allow longer battery life than if you used two. Finally, if you really need the performance, you can crank it to the max (as long as the case can handle the heat).
I hope that I do not sound pushy, but I would like you to be the happiest with your project. Professionals need the best tools. Anyway, the Sandwich Shoppe is perfectly legit. I believe that it is nearly as old as the CPF, and practically created for us as far I know. I have ordered from them many times, and were happy with the service. They responded to my emails helping me with any issues or questions I might have had. I always thought it was weird that they do not have a number listed, too. Well, if you have any concerns, just ask around.
I hope what I said made sense. If not, just ask. Next time I need to start tossing in some numbers and math to clarify. These drivers can be set up so easily once you determine the LEDs used and how much current you want to send to the LED, how they are wired, and the type of power source.
-Tony
roguesw
Enlightened
Yes, they are very reliable, they have been part of CPF since the beginning and many modders utilise their convertors , some people like McGizmo.
A lot of modified lights use their convertors for many years before KD came online.
I have a light that is built with a shoppe convertor and its pushing 6 years old now and still going strong.
Order from them with confidence.
They even have a subforum here on CPF in the modded section
Justin Case
Flashlight Enthusiast
- Joined
- Mar 19, 2008
- Messages
- 3,797
hipCC can be controlled via the d2Flex PWM controller.
Besides the pot, you can add the Remora UI daughterboard to the Shark Buck for multimode control.
Justin Case
Flashlight Enthusiast
- Joined
- Mar 19, 2008
- Messages
- 3,797
If you go to the Shoppe's web site, you may run into problems with Google trying to block access because they report the Shoppe as an "attack site". The Shoppe's web site is legit and not an attack site. Their server got infected. See this thread for explanation.
Tony, thanks a million for all the info. From what I understand is that I can drive the 3 emitters from the same buck with only 1 dimmer. The voltage supply is 14.5VDC tapped from a professional Anton Bauer Pro Video camera. These batteries are used for powering video cameras as well as having a TAP or Power Brick connector on them. For one type of video light that I am building, will take its power off of it. They are conservatively
rated at 8000ma if that means anything. I imagine they are li-ion or better.These batteries are not cheap and typically go for 500 dollars each.Videographers usually have 3 or 4 of these fully charged for the event. They normally run a 50 to 100w halogen sucking somewhere near 10 amps. Keep in mind that the cameras are not cheap either, averaging 10 to 15 grand a piece. That is why cost no problem to build one. Tony, I don"t mind buying a driver for each led with a pot for each led for dimming. You say that underdriving them would be more efficient. How about 3 of them in the same enclosure, so if I underdrive them, they should have the same output or more as 2 at full capacity. Would I see any problems using 3 drivers as I am afraid that 1 driver can overheat, stressed ? Also I think that having them as 3 separate circuits can be more reliable as if 2 would shut down for any reason, the third can be used to at least illuminate the subject. Please correct me if I am wrong as I will be building approximately 100 of these type of units for our videographers. The other unit I am thinking about is a 2 led unit with a built-in battery and dimmer. Thanks
rated at 8000ma if that means anything. I imagine they are li-ion or better.These batteries are not cheap and typically go for 500 dollars each.Videographers usually have 3 or 4 of these fully charged for the event. They normally run a 50 to 100w halogen sucking somewhere near 10 amps. Keep in mind that the cameras are not cheap either, averaging 10 to 15 grand a piece. That is why cost no problem to build one. Tony, I don"t mind buying a driver for each led with a pot for each led for dimming. You say that underdriving them would be more efficient. How about 3 of them in the same enclosure, so if I underdrive them, they should have the same output or more as 2 at full capacity. Would I see any problems using 3 drivers as I am afraid that 1 driver can overheat, stressed ? Also I think that having them as 3 separate circuits can be more reliable as if 2 would shut down for any reason, the third can be used to at least illuminate the subject. Please correct me if I am wrong as I will be building approximately 100 of these type of units for our videographers. The other unit I am thinking about is a 2 led unit with a built-in battery and dimmer. Thanks
Illum
Flashaholic
full rated output would be 350ma per die, maximum rated at Iled = 700ma
Assuming your power supply is DC...all you would need is to wire your MCE emitter in series by either using individual hookup wire or a series connected LED star.
Since this is posted in "beyond flashlights" I'm assuming this is for fixed lighting or any other purposes that the recommendation will not be constrained by the ability of it to fit in a flashlight...
MCE wired in series would yield 3.2x4 = 12.8V, your DC power supply may not be adequate as some buck [step-down] converters require minimum of 16V input for 4 LEDs [such is the case IIRC with the LuxDrive power puck].
I'd recommend the LuxDrive 4015 BoostPuck [Datasheet here] available from www.Ledsupply.com, there may be other alterenatives to choose from. Note that the boostbuck stated above is limited to Iled = 350ma
Hi and thanks for a quick reply. Perhaps I am confusing the issue with 3 led's in a fixture. It is not a flashlight. It is an on-board video light with 3 led's with 3 drivers ( one driver dedicated to a led) with 3 dimmers ( one dimmer dedicated to a driver) in other words it's like having 3 redundant units in case one or 2 fails. The only common thing is that they will be powered from the same battery 14.5 VDC rated at 8000ma li-ion. Can this be built ? Cost or money is absolutely not a problem.Reliability is the key factor. Thanks
