Jump to content

CDI 2009 LED M5 Samples


Recommended Posts

Uggg, here we go again reverse engineering and designing circuits that should be done by the "knowledgeable" engineers at the manufacturing (Innovalite) end. Don't "they" know how to do all this or can't they hire smart enough people to figure it out. It amazes me that something so simple can be made to be so complex.

Here's our requirements.

Smooth, wide range fading across all color and counts.

Reliable and robust.

Bright.

No Flicker, No noise.

I think low cost is kind of a moot point since we know these things are going to cost more than those you can buy at your hardware store and we are more demanding than most people who do not animate their lights.

Design them right. Be practicle about it. Build 'em and we'll buy 'em.

Did I miss anything?

Edited by Duke
Link to post
Share on other sites
  • Replies 100
  • Created
  • Last Reply

Top Posters In This Topic

Let me get this straight. Did you actually disassemble both ends of a string and carefully examine it to come up with the schematic, or did you use some other method to come up with the schematic? Is it based purely on observation or on some combination of observation and supposition? Don't take this personally, it's just that I am a firm believer in starting with the basics and taking one thing at a time.

--

Phil

I tore both ends apart and followed the traces to confirm the voltage doubler circuit. There is not that much to the connectors/plugs. It is basically a snap together assembly with a lid sonic welded over the PCB with what looks like bathtub caulk trying to seal it. I am hoping these plugs are for prototype only. The wires are not actually soldered to the PCB, but use a pressure fit to make contact. Moisture WILL get in the plugs and I am afraid eventually get to the circuit board.

Link to post
Share on other sites

I tore both ends apart and followed the traces to confirm the voltage doubler circuit. There is not that much to the connectors/plugs. It is basically a snap together assembly with a lid sonic welded over the PCB with what looks like bathtub caulk trying to seal it. I am hoping these plugs are for prototype only. The wires are not actually soldered to the PCB, but use a pressure fit to make contact. Moisture WILL get in the plugs and I am afraid eventually get to the circuit board.

Thanks. I'll be receiving some samples as well on Monday, and that saves me some time and effort.

--

Phil

Link to post
Share on other sites

I tore both ends apart and followed the traces to confirm the voltage doubler circuit. There is not that much to the connectors/plugs. It is basically a snap together assembly with a lid sonic welded over the PCB with what looks like bathtub caulk trying to seal it. I am hoping these plugs are for prototype only. The wires are not actually soldered to the PCB, but use a pressure fit to make contact. Moisture WILL get in the plugs and I am afraid eventually get to the circuit board.

Bob - keep in mind that these are just SAMPLES, the design of the circuit board/PCB has not been finalized. I sent these out for your (the testers)comments/suggestions, let's keep it civilized and constructive! Final plug molds will be made when we/Innovalite get the design to alls liking. I and Innovalite are both following this thread!!!!

I can assure you that the easy thing would have been to walk away from this. CDI and Innovalite are dedicated to fixing this. Innovalite has one customer that is asking for a design change - ME! This change is for ME/YOU only. There are not many Chinese venders (if any) that would give this kind of commitment to a small customer. They have commited a group of people to work on this design and make it to your specifications. We are making efforts to improve this product and actually I feel the product has come a long way. I've been running them on LOR for 48 hours now and I have to say they are not near as bad as some of the comments. The milliseconds of fade are only detectable because some of the testers pointed that out. The average individual wouldn't even notice it. Lastly, don't take the last two sentences for anything other than what they are. We are still improving the product and the final product will be to everyones liking!

I have to tell you that Innovalite is a little upset with the bashing of them and me with the commitment we are making to getting this product up to the hobbists expectations. We have sold a ton of this product to static users that are thrilled with it. Please keep it constructive and lets all work together!!

Link to post
Share on other sites

I apologize for anything that I wrote that may be taken as bashing. So far I haven't received any samples, let alone applied power to them, and my comments should be viewed in this light.

Additionally, I'm glad that you mentioned that these lights are pre-production prototypes subject to change. I was previously under the impression that these were representative of the exact product that would come off the production line. I should have asked about this, of course.

--

Phil

Link to post
Share on other sites

Bob - keep in mind that these are just SAMPLES, the design of the circuit board/PCB has not been finalized. I sent these out for your (the testers)comments/suggestions, let's keep it civilized and constructive! Final plug molds will be made when we/Innovalite get the design to alls liking. I and Innovalite are both following this thread!!!!

Paul - I think I said in every post that I was under the assumption that the plug ends were prototypes and that I assumed the production version would be different. I apoligize if it came off as uncivilized and unconstructive. That was not my intention at all. I was under the impression that you were seeking input to develop the best product possible. Again, I apologize, I thought I was being constuctive in pointing out some possible improvements in the design.

Link to post
Share on other sites

I have to tell you that Innovalite is a little upset with the bashing of them and me with the commitment we are making to getting this product up to the hobbists expectations. We have sold a ton of this product to static users that are thrilled with it. Please keep it constructive and lets all work together!!

Paul, I sincerely appologize if anything I've said can be interpreted as "bashing". I have been very pleased with your efforts to resolve this issue and have every intention of remaining a loyal customer. However, it's possible that some of my disappointment over these strings may have inadvertantly shown through in my posts. I've been very close to tears because the show I've worked so hard on for nearly a year is now not going to happen because my lights died. I honestly didn't mean for this sadness to come off as "bashing".

I don't want anything to interfere with getting this problem resolved, so I'll just quietly watch this thread from the sidelines. You've got some very bright (no pun intended ;)) testers working on your side. I feel sure they'll be able to get these strings to meet the basic expectations Duke outlined so well in his post.

Link to post
Share on other sites

OK, so I've done some testing with the LOR setup now...

I have to say, I think I like the new strings...

The extra capacitance (in the colors other than red), give them a little bit of a incandescent feel. I can see where animated users (such as me) using both LEDs and incandescents would like that. The effect is a little exaggerated however, as it takes a few seconds for complete darkness when you go from 100% to 0%. The greens seem to take the longest, producing some amount of visible light for almost 4 seconds.

All the fades seem to be very smooth, I'm not noticing any "hiccups" like I've seen in some of the '08 strings I have.

I also don't get any of the audible buzzing I experienced when fading the '08 50-ct strings I have.

My biggest complaint is the instant off of the red vs. the other colors. I'd like to see either the red's fade out a little, or the other colors instantly turn off.

Perhaps a little shorter discharge time would be nice too, something closer to a incandescent mini-light. But if that came at the expense of smooth fades, then I can live with the longer time.

I'll probably try to incorporate these strings in to my display this year and give them a bit more of a workout, but I don't expect any problems.

So far I give them an A-.

Get those power-off fade rates a little closer from color to color and that grade will go up. Again, it's not a "deal-breaker", but would be a "nice to have".

Paul S. - Should we be concerned about the sample nature of these strings when it comes to using them outside? I'm sure they are fused properly so I'm not worried about my equipment, I just don't want to report a failure for a moisture (water/snow/ice) problem if they aren't intended to handle those conditions...

--Paul

Edited by pbryan
Link to post
Share on other sites

Paul S. - Should we be concerned about the sample nature of these strings when it comes to using them outside? I'm sure they are fused properly so I'm not worried about my equipment, I just don't want to report a failure for a moisture (water/snow/ice) problem if they aren't intended to handle those conditions...

--Paul

I'd say go ahead - the plugs may not be sealed tightly but the LED's are.

Link to post
Share on other sites

After running these strings through some test patterns and then through a repeating sequence for about 3 hours, here is what I find.

Hardware - Renard 24 channel board

Software - Vixen

Lights - All 5 strings sent by Paul plus a red incandescent set as a control of sorts.

Testing

  • Ramp up to full and down to to zero.

    • % of intensity at first visible light.

      • Warm Wht - 15%
      • Cool Wht - 15%
      • Blue - 15%
      • Green - 15%
      • Red - 15%
      • Incand - 15%


    • % of intensity at full on.

      • Warm Wht - 65%
      • Cool Wht - 65%
      • Blue - 65%
      • Green - 65%
      • Red - 40%
      • Incand - 80%

Note: Red had a noticeable jump in brightness at 20%. At 20% it also had a small flicker. All other colors had a gradual ramp up in brightness. Red was much faster to reach full perceived brightness at 40%. With a long slow ramp up in a sequence, the difference is visible. In a fast ramp up or down, it is less apparent and you'd have to be picky to point out the difference.

  • 100% On to 0% Off - no ramping or fading - Perceived time to Off

    • Warm Wht - 1 sec
    • Cool Wht - near instant
    • Blue - 2 sec
    • Green - 3 sec
    • Red - instant
    • Incand - instant

Note: Red and cool white are near perfect, no gradual fade to black like the warm white, blue, and green. Green was the slowest to go dark. It was very noticeable in a sequence where I quickly turned off and then on the lights. Blue and green never make it to full off of during the sequence at times.

Failures:

After 3 hours of usage, the only issue I have found regarding a failure is a single blue LED has a random flicker.

Color:

All colors look good and are vivid. While they don't exactly match the incandescent, that is to be expected and they look much better than any incandescent. Warm white is a near exact match to incandescent white.

Physical:

The rectifiers have moved to the male and female plug ends. No blobs inline. While this looks better, it does cause the loss of the piggy back receptacle in the male end. In my opinion, I'd rather blobs on the string and keep the piggy back receptacle, but that is just my preference. The LED bulbs appear to be sealed and not replaceable. I have not yet tested these in incliment weather as it has been nice here lately. I may test these with the sprinklers later.

Opinions: Comparing to the performance of incandescents

  • Red - work very well with sequencing. The only issue with Red is how quickly it gets to full brightness. At 20% the flicker is only noticable if you keep it steady at 20%. I did not see the flicker during ramping. Nice match to incandescent performace if the jump in brightness is fixed.
  • Cool White - Works well with sequencing. No slow fade to black issues. Better ramp up to brightness than Red. Matches nicely with incandescent performance.
  • Warm White - good color match with white incandescents. Good ramp up in brightness. 1 second fade to black with full power off is slightly noticeable on sequences.
  • Blue - Nice color. Good ramp up in brightness. 2 second fade to black with full power off is very noticeable on sequences.
  • Green - Nice color, somewhat on the light side of the spectrum. Good ramp up in brightness. 3 second fade to black with full power off is not acceptable to me. Does not work well with sequencing because of this issue.

Overall:

These samples did not die or result in 10% brightness after ramping/fading like my 2008 C7s did. That is a very good step in the right direction. I still don't like the use of capacitors due to the slow fade to black issues. Also, while red performs well with no fading to black issues, it needs a smoother curve for ramping up to full brightness. If these issues can be fixed, I'd really have no issues with these sets. I'm not commenting on construction at this time as these are samples. Generally, I could care less about the looks of the string, placement of rectifiers, etc. As long as they look good in the dark and perform well, they will work well for me.

Paul - if there is anything specific you would like me to test, please let me know. If anyone has a question or testing suggestion, please inform me. I want to do my best to test these samples.

Link to post
Share on other sites

I'm not a hardware geek so forgive me if this is a dumb question... but could the bleeder resistor across the cap be changed to a different value, and have the cap drain quicker, eliminating the fade-down issues mentioned above?

-Tim

Link to post
Share on other sites

I'm not a hardware geek so forgive me if this is a dumb question... but could the bleeder resistor across the cap be changed to a different value, and have the cap drain quicker, eliminating the fade-down issues mentioned above?

-Tim

The concern here is power dissipation. Changing the resistor to 100K should decrease the bleed-down time by a factor of 10, but the power dissipation in each plug-end would be over 1/4W. You would need to use a 1/2W resistor to have any margin, which is physically larger part. I don't know if there would be any issues with UL/CSA with that sort of change.

Edit - that change would also increase the ripple current in the caps. I don't know if this would be enough to cause any longevity problem with the caps or not.

Edited by pshort
Link to post
Share on other sites

You'll also be adding more parasitic power usage to the string of efficient LEDs. It's not a ton more, but part of the reason for using LEDs is power savings, and part of the design should always be to minimize power usage. Too much power dissipation here causes heat as well, which isn't usually good for encasement in a waterproof housing.

Reducing the size of the caps could help too, but it will change the dynamic of the circuit in it's normal operation. There is likely a trade-off here between the power off decay rate and characteristics that give us the smooth fade up/down operation. It's possible that this hasn't been optimized. It's also possible that any further reduction in the decay time really hurts the fade performance.

--Paul

The concern here is power dissipation. Changing the resistor to 100K should decrease the bleed-down time by a factor of 10, but the power dissipation in each plug-end would be over 1/4W. You would need to use a 1/2W resistor to have any margin, which is physically larger part. I don't know if there would be any issues with UL/CSA with that sort of change.

Edit - that change would also increase the ripple current in the caps. I don't know if this would be enough to cause any longevity problem with the caps or not.

Link to post
Share on other sites

There is likely a trade-off here between the power off decay rate and characteristics that give us the smooth fade up/down operation. It's possible that this hasn't been optimized. It's also possible that any further reduction in the decay time really hurts the fade performance.

--Paul

Well spoken - this is a follow up from Innovalite:

We only can do a little trimming, like change size of resistor. We can not modify too much, like circuit diagram or basic construction. It may hurt the integral performance. Now we know what you require, and we like to do our best to satisfy you, but please understand, patent, UL approval, circuit diagram had been completed, and raw material, electronic components had been prepared. We will discuss with JC for his help and will let you know the result ASAP

Link to post
Share on other sites

Well spoken - this is a follow up from Innovalite:

We only can do a little trimming, like change size of resistor. We can not modify too much, like circuit diagram or basic construction. It may hurt the integral performance. Now we know what you require, and we like to do our best to satisfy you, but please understand, patent, UL approval, circuit diagram had been completed, and raw material, electronic components had been prepared. We will discuss with JC for his help and will let you know the result ASAP

Perhaps another stupid question, but can they go back to a design similar to your 2007 stock? The LED's I bought from you then look and work like a champ...

-Tim

Link to post
Share on other sites

Perhaps another stupid question, but can they go back to a design similar to your 2007 stock? The LED's I bought from you then look and work like a champ...

-Tim

The big problem is those were made in another factory with their patents. 2008 and further were switched over to our factories with our/Innovalite patents. The earlier versions are wired parallel in series of 25 - the latter is in series of 10 - 140 and with the high counts (over 50 from what I am understanding) need a capacitor which causes a slight fade rather than a complete shut off (we are talking milliseconds). JC is the factory owner and he is seeing what he can do with that as we speak (E-mail). More to come.

Link to post
Share on other sites

need a capacitor which causes a slight fade rather than a complete shut off (we are talking milliseconds).

If it's really milliseconds, it wouldn't be a problem for me. Someone above in this thread was claiming 4 seconds, and that wouldn't be acceptable for my own uses...

For what it's worth, I much prefer the smaller series (25-LED), as if it's all one big series, if one LED fails, the whole string will go out. We had one fail last year, and only 25 LED's went out so it wasn't nearly so drastic.

Good luck on getting this resolved!

-Tim

Link to post
Share on other sites

Well spoken - this is a follow up from Innovalite:

We only can do a little trimming, like change size of resistor. We can not modify too much, like circuit diagram or basic construction. It may hurt the integral performance. Now we know what you require, and we like to do our best to satisfy you, but please understand, patent, UL approval, circuit diagram had been completed, and raw material, electronic components had been prepared. We will discuss with JC for his help and will let you know the result ASAP

CORRECTION FROM INNOVALITE:

l The earlier versions are wired parallel in series of 25 & 35, so need extra rectifier housing after the end of series.

l New version is in a series of 10 to 150.

l The purpose of capacitor is involving electric charges of equal magnitude, but opposite polarity, building up on each plate. No heat builds up on rectifier housings, thus no short circuit situation.

l Using capacitor is much safer than the other competitors.

Link to post
Share on other sites

Looks like a translation from (I assume) Chinese.

I think what he's saying here is that a standard bridge rectifier requires a current limiting resistor in series with the LED's inside of each rectifier block. Since we're working with 120V AC, that resistor, in many cases, would have to dissipate a lot of power. In the case of the 50 count red's that some were converting to bridge rectifiers a 3W resistor was needed. That resistor will also get hot (or at least warm). Heat is bad for preventing moisture build up, and likely bad for passing UL (or other fire prevention) certifications.

Their design using the capacitors creates a current limited power supply, with enough voltage to drive 100 LED's in one string. And since it doesn't need a big limiting resistor, it doesn't get hot.

It sounds like from the other e-mail, they are pretty set in their component selection. It would be nice if they could tweak it a little to speed up that decay rate (especially for the greens). But it may not be as easy as we make it sound...

--Paul

l The purpose of capacitor is involving electric charges of equal magnitude, but opposite polarity, building up on each plate. No heat builds up on rectifier housings, thus no short circuit situation.

l Using capacitor is much safer than the other competitors.

Link to post
Share on other sites

My samples arrived today. So far all I've done is plug them into the AC outlet to check for DOA, but (as expected) all were alive. I did notice, however, that it took a noticeable amount of time for all of the LEDs except the red ones to fade to off when unplugging them from the AC outlet.

Now I'll have to dig my spare Renard setup out of the garage to do more testing.

Link to post
Share on other sites

it took a noticeable amount of time for all of the LEDs except the red ones to fade to off when unplugging them from the AC outlet.

Could you define "noticeable"? So far we have people saying milliseconds, and others 4 seconds... ;)

I would say something around 1/2 second would be the minimum acceptable for me. It would be sad if they couldn't get them at least this low, since other brands out there work great with controllers, and I think a lot of customers might flock away from these... (I don't mean that in any way as "bashing" -- just as feedback on my own needs if I were to buy these...)

-Tim

Link to post
Share on other sites

Could you define "noticeable"? So far we have people saying milliseconds, and others 4 seconds... ;)

I would say something around 1/2 second would be the minimum acceptable for me. It would be sad if they couldn't get them at least this low, since other brands out there work great with controllers, and I think a lot of customers might flock away from these... (I don't mean that in any way as "bashing" -- just as feedback on my own needs if I were to buy these...)

-Tim

Tim,

I'm still testing, but it seems to be a factor of how charged the capacitor is that determines the "fade rate" when power is removed.

Using the LOR HW utility, an ON immeditely followed by an OFF, as fast as I can click the mouse, seems to be (eye ball test) similar to a .7 sec fade down.

Leaving the string ON for 10 secs or so, then an OFF, seems similar to a 1.8 sec fade down rate.

This was with the 100 Ct Green.

Link to post
Share on other sites

I made a 10 channel test sequence, all channels identical, with the 5 test strings and 5 strings of incadescents. The M5 reds seem to mirror the regular minis, with the M5 Greens being the worst to mimic an incadescent, with its longer afterglow, depending on perhaps how long it was ON before receiving the OFF.

Tonight I'm adding 5 more channels and will test adding the 2007 LEDs.

This will allow comparing 2 different mfgs LEDs side by side with regular minis.

Others have done a great job with component and circuit design evaluation, but to me now, seeing how they would work with a display, and how much consideration for the different OFF and FADE rates would have to be compensated for, is prime consideration. I will use all 5 of my test strings in the display this year.

I will be preordering the 2009 versions from Paul regardless, at least these can be dimmed without failing.

Link to post
Share on other sites

Could you define "noticeable"? So far we have people saying milliseconds, and others 4 seconds... ;)

I would say something around 1/2 second would be the minimum acceptable for me. It would be sad if they couldn't get them at least this low, since other brands out there work great with controllers, and I think a lot of customers might flock away from these... (I don't mean that in any way as "bashing" -- just as feedback on my own needs if I were to buy these...)

-Tim

OK, I'm set up now (Renard + DIY SSR) for use with Vixen.

I'm using a sequence with a 1-second on-time followed by 1-second off time. The red turns off instantaneously, the whites turn off quite quick (my estimate is around 1/10 sec, with the yellowish-white slightly faster than the bluish-white). The blue turns off in just under 1 second (i.e. it's pretty much totally off before it turns on again), while the green is still visibly (but faintly) on when it turns on again. Of course you must bear in mind that I'm over 50 years old, so my eyes don't respond as fast as they used to.

Link to post
Share on other sites

. ........Of course you must bear in mind that I'm over 50 years old, so my eyes don't respond as fast as they used to.

lol, that's my situation also, but I'm almost 61 :D Thats why I added the regular minis to my test sequences, so I can see if there is a difference

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...