Jump to content
SaltCritters.com

Euroquatics E5 Teardown


evilc66

Recommended Posts

Ok party people, teardown time! I'll do a picture dump first, then I'll go into my initial thoughts.

 

Here it is unmolested

 

IMG_20160203_085137886.jpg

 

IMG_20160203_085203811.jpg

 

IMG_20160203_085223602.jpg

 

IMG_20160203_085259800.jpg

 

Opening it up, some interesting things become apparent

 

IMG_20160203_085949092.jpg

 

IMG_20160203_090007455.jpg

 

IMG_20160203_090919298.jpg

 

IMG_20160203_090934110.jpg

 

IMG_20160203_090027699.jpg

 

First off, I'm going to reserve final judgement on these after I have done all the tests I want to do, which will include electrical and optics tests.

 

Now, with the tube fully assembled, I will say that it feels quite nice in the hand. That may not mean much to some people, but the fact that it's rigid, smooth, and free of burrs is a good sign of overall quality. The plastic materials felt good and had no flex to them. That's a problem with some Chinese products. They skimp on the plastics and makes it feel cheap (Odyssea T5 end caps, I'm calling you out!)

 

Once inside, I was a little surprised. Let's focus on the pcb first.

 

It doesn't get more simple than this. First thing to note is the use of poke in connectors on the pcb to connect to the end caps. It's a nice feature that some manufacturers would consider unnecessary, as it adds cost. Soldering would be the alternative, but connectors make assembly (and disassembly ;) ) much easier. Power from the ballast is fed into a full bridge rectifier to convert the high frequency AC into DC. There is no capacitor after the rectifier like you would normally see, but seeing as the output frequency of most electronic ballasts is somewhere in the order of 20kHz, there really isn't any need for it. No one is going to see flicker at that frequency. Plus, it's one less point of failure.

 

Each end has a pair of 0.2 ohm current limiting resistors in parallel. I'm assuming they are there to limit the inrush current on startup.

 

On the opposite end to the rectifier, there is a chip that I have not been able to identify. The markings are LX8800A2 1542. I have not been able to find any indication as to what it is. If anybody runs across it, let me know. It's even tough to guess as to what it does, as even though it's an 8 pin chip, only pin 8 (LED-) and pin 5 (rectifier -) are connected to anything, and it's in parallel with the LEDs.

 

As for the LEDs, they are a 5630 format LED. Brand is unknown. The Cool Daylight tube that I opened certainly had a very thin application of phosphor, which would line up with the 20K color temperature listed on the exterior label, and on the pcb itself. The pcb even shows the average forward voltage of each LED at 3.1-3.2v. Convenient.

 

Now, the pcb itself is single sided 1mm thick fiberglass. I was fully expecting an aluminum MCPCB. Granted, the overall power of this tube is only 15W (7.5W/ft), which isn't a lot of heat, but I would have thought for the sake of longevity it would have been prudent to get the heat away from the LEDs as best as possible. With the pcb being single sided fiberglass, there is a pretty high thermal resistance between the LED and the heatsink. And with the pcb not being bonded to the heatsink at all (not even thermal paste, and there is basically no clamping pressure), the heatsink is there more for show than anything else. As of right now, it's main purpose is to make the tube rigid. That's not to say that it won't get heat out of the tube. It just won't be much. I'll stick a thermocouple next to an LED and see what comes of it.

 

Moving on to the housing, we find that the plastic cover is a nice milky white with ridges in the interior to help with diffusing the light. I have no doubt that the output will be pretty smooth and T5-like. It has tracks for the pcb and the heatsink to slip into, and like I said before, no burrs. Everything goes together like butter.

 

The heatsink is surprisingly small, but like I mentioned before, it doesn't seem to be removing as much heat from the tube as you would think. Keep in mind that I haven't fired these up, so I have no real idea just now as to how hot these actually get. Again, no burrs, and everything goes together nicely. A single screw holds each end cap on.

 

 

So there you have it. An initial look at what makes a Euroquatics E5 tube tick. Please hold back your pitchforks and torches until I can run more tests and see what these things can really do. Here's what I'm planning on testing:

 

  • Basic electrical profile (might help identify what that mystery chip is)
  • Thermal measurements of the heatsink and the pcb
  • Thermal image of the heatsink and pcb
  • PAR reading with Apogee meter
  • PAR reading with Ocean Optics USB4000 spectrometer (narrowed to 400-500nm and 600-700nm range)
  • Relative intensity using spectrometer
  • Spectral distribution using spectrometer
  • CCT using spectrometer
  • CRI using spectrometer

Hopefully I can get all this testing in a reasonable amount of time, but with work, school, and two kids, finding spare time can be a little challenging. I'll try my best though.

 

I'll probably fire them up tonight just to see what they look like.

Link to comment
  • Replies 115
  • Created
  • Last Reply

Awesome. Answered one of my questions about how the tubes handle the T5 startup and operating voltage/current.

 

Definitely interesting they didn't use aluminum or some other mechanism to minimum heat buildup around the diodes. I wonder if the suggested 8 year lifespan is in part due to breakdown over time compared to some of the crazy 50K+ hour lifespans of cree's and other LED's under optimal conditions.

Link to comment

That's certainly possible. Commercial LED bulbs are reducing their lifetime ratings so that they can keep costs to a minimum. It's a shame they didn't add a simple thermal interface adhesive to improve things. But, I haven't fired them up yet so I don't know how hot it all gets.

Link to comment
jedimasterben

I looked up the unidentified ic as "IC 1542" got back linear technology LTC 1542

Amplifier, comparator, reference.

 

http://www.digikey.com/product-detail/en/LTC1542IMS8%23PBF/LTC1542IMS8%23PBF-ND/961662

Going off of that leads to the datasheet: http://cds.linear.com/docs/en/datasheet/15412fd.pdf

 

COMPIN+ (Pin 5): Noninverting Input of Comparator. The input common mode ranges from VSS to (VCC – 1.3V). The input current is typically 10pA at 25°C.

 

VCC (Pin 8): Positive Supply, 2.5V ≤ VCC ≤ 12.6V. The supply bypass capacitors are not required if the supply impedance is low. For single supply applications, it is a good general practice to bypass VCC with a 0.1µF capacitor to ground.

Link to comment

I looked up the unidentified ic as "IC 1542" got back linear technology LTC 1542

Amplifier, comparator, reference.

 

http://www.digikey.com/product-detail/en/LTC1542IMS8%23PBF/LTC1542IMS8%23PBF-ND/961662

 

 

Going off of that leads to the datasheet: http://cds.linear.com/docs/en/datasheet/15412fd.pdf

 

COMPIN+ (Pin 5): Noninverting Input of Comparator. The input common mode ranges from VSS to (VCC – 1.3V). The input current is typically 10pA at 25°C.

 

VCC (Pin 8): Positive Supply, 2.5V ≤ VCC ≤ 12.6V. The supply bypass capacitors are not required if the supply impedance is low. For single supply applications, it is a good general practice to bypass VCC with a 0.1µF capacitor to ground.

I'm pretty sure the 1542 part of the marking is just a lot code. The part you need to be looking for is LX8800A2. I've looked everywhere and can't find any trace of a chip with that marking. I hate gray market chips :angry:

 

Plus, on that LTC1542 part, only Vcc and Comp-In would be connected. The chip wouldn't work. Seriously, there are no other pins connected to anything other than pins 5 and 8.

Link to comment
jedimasterben

 

 

I'm pretty sure the 1542 part of the marking is just a lot code. The part you need to be looking for is LX8800A2. I've looked everywhere and can't find any trace of a chip with that marking. I hate gray market chips :angry:

 

Plus, on that LTC1542 part, only Vcc and Comp-In would be connected. The chip wouldn't work. Seriously, there are no other pins connected to anything other than pins 5 and 8.

Didn't make much sense to me, either.

 

i%252520dunno.gif

Link to comment

From the pictures it looks like SMD diodes?

I had thought that too, but there's continuity between the two pins (standard continuity test, not diode test). It's possible that it may have a low breakdown voltage, like a low voltage zenner, but what purpose would it serve?

 

I may prod and poke it some more tomorrow after I at least get them lit up once.

Link to comment

Didn't make much sense to me, either.

 

i%252520dunno.gif

Me neither. If you look at the datahseet you linked to, the markings for that chip do not go along with what Evil said was marked on the chip on the Euroquatics board.

 

I work as a component engineer and have a lot of resources for finding chips like this and I can't find any reference to LX8800A2 anywhere. It could be something that was made proprietorially for Euroquatics....or grey market...but I'm leaning more towards it being a proprietary chip, in which case you will probably never really know exactly what it does.

BTW Evil, nice write up so far. This product has intrigued me since it was first talked about. Nice to finally see what the internals look like. I look forward to seeing some more of what your testing unveils.

Link to comment

A little background...

 

I'm currently running a Fish Need It 4x24W fixture with the stock reflectors and ballasts. It's not great, but it's WAY better than anything from Odyssea. Bulbs were all UVL T5HO-R (T5 with the built in reflector), consisting of 2x454, 1xActinic White, and 1x Super Actinic. None of them are particularly new.

 

 

 

So, I swapped out a 454 and the Actinic White with the Blue Pop and the Cool Daylight respectively. The first thing you notice is that they are bright. Very bright. I know my bulbs are pretty old (going on a year), but the increase in brightness is quite substantial. I'll get some comparison pictures soon once I have a little more time.

 

The other thing I noticed is that they get hot. They are hot everywhere, but particularly the heatsink which I was kinda surprised by. I can only imagine the temperature of the LEDs considering the poor thermal path.

 

Color wise, they are about what you would expect. The Blue Pop is a straight royal blue setup. The Cool Daylight is a very crisp high kelvin white. I can't really comment on the color rendering of the Cool Daylight, but time will tell.

 

I think tomorrow I'm going to get a fresh set of T5 bulbs from my LFS. They only deal with Giessmann for quality bulbs (the rest are Coralife and Aquatic Life), so I'll probably get something similar to what I have now.

 

More to follow

Link to comment

I wonder what the efficacy of the chips is.

 

If the RB chips are the same package then radiometric data from the blue pop lamp would be very useful.

Link to comment

So how can we DIY one of these? If the biggest benefit is from the plastic diffusing the light to make it T5ish wouldn't getting something similar create the same result? Maybe a white ridged plastic cover over a strip of leds.

Link to comment

So how can we DIY one of these? If the biggest benefit is from the plastic diffusing the light to make it T5ish wouldn't getting something similar create the same result? Maybe a white ridged plastic cover over a strip of leds.

You may be able to do something with these.

http://ckleddistributions.mybigcommerce.com/aluminium-profiles/

I have a friend that is using one of these to add some lighting under her kitchen cabinets. You would have to make a board similar to the one Euroquatics made though, although it is not that complex.....Hey Evil maybe this could be The Evil Par38 Round 2...."The Evil tube"

Link to comment

You may be able to do something with these.

http://ckleddistributions.mybigcommerce.com/aluminium-profiles/

I have a friend that is using one of these to add some lighting under her kitchen cabinets. You would have to make a board similar to the one Euroquatics made though, although it is not that complex.....Hey Evil maybe this could be The Evil Par38 Round 2...."The Evil tube"

 

 

EVIL TUBE! EVIL TUBE! EVIL TUBE!!!!!!! :D:D:D:D:D

Link to comment

I'm a component engineer, and looked up the LX8800A2. The LX88 appears to be the base number for a Microsemi LDO Regulator. I wouldn't think they would go with a custom IC, but who knows. ICs aren't my commodity, so I don't want to email Microsemi for a datasheet. Granted, if this is a custom IC, Microsemi might not release the datasheet.

Link to comment

I had run across Microsemi in my searches too. They were the closest to the part number, but obviously no matches. I don't mind pinging them to see if it's one of their parts.

 

A linear regulator doesn't make a ton of sense though based on how it's connected to the LEDs. It's run in parallel to them, not in series. Only thing I could think it would be is a clamping diode to help snub any voltage spikes at startup.

Link to comment

I'm a component engineer, and looked up the LX8800A2. The LX88 appears to be the base number for a Microsemi LDO Regulator. I wouldn't think they would go with a custom IC, but who knows. ICs aren't my commodity, so I don't want to email Microsemi for a datasheet. Granted, if this is a custom IC, Microsemi might not release the datasheet.

I wouldn't think they would either, but nothing else seems to make sense. I just looked into the LX88 based on your observations. I can't find any datasheets for the LX8800, but I did find similar ones for others in the LX88XX series. None of them are 8 pin chips and the markings that would be on them do not correspond with what Evil is seeing on the mystery chip, so I'm leaning away form this being what it is as well.

I had run across Microsemi in my searches too. They were the closest to the part number, but obviously no matches. I don't mind pinging them to see if it's one of their parts.

 

A linear regulator doesn't make a ton of sense though based on how it's connected to the LEDs. It's run in parallel to them, not in series. Only thing I could think it would be is a clamping diode to help snub any voltage spikes at startup.

Was questioning this a bit too.

Link to comment

Pretty happy with the new Giessmann bulbs. Ended up with an Aquablue+, Pure Actinic, and two Actinic+. Came out to just about the same color as the UVL bulbs. So now I have a good fresh set of bulbs to compare against :)

Link to comment

Archived

This topic is now archived and is closed to further replies.

  • Recommended Discussions


×
×
  • Create New...