New Millett Hybrid Maxed Amp
May 26, 2008 at 2:13 PM Post #4,816 of 6,727
Hey guys, Greg at SKA emailed me to let me know the PCBs for the OptiVols are in. Mine are shipping tomorrow. I've been waiting over a week for a kit from Canada, so who knows how long it will take these to show up for Australia, but I ordered one assembled and tested, and one kit. I'll post a pic of the assembled and tested one, but it's not going to be used in my first build, it's going in the second build, the special MOSFET-MAX.
 
May 26, 2008 at 3:20 PM Post #4,817 of 6,727
Thanks for the LED help! The build should be done by next week, so I'll put pics up asap...
 
May 26, 2008 at 3:36 PM Post #4,818 of 6,727
I have found a White LED at radioshack that has a 20mA rating, which is the same as the LED on the BOM. So I should have no problems with this one correct?

The MCD is rated at 1100, and the "Typical Voltage" is 3.6V

I would like to avoid having to swap out my resistors since I have soldered them in already...

Clearly my understanding of this in terms of electronics is poor...any explanation would be a great help!
 
May 26, 2008 at 3:37 PM Post #4,819 of 6,727
Quote:

Originally Posted by agutt /img/forum/go_quote.gif
I have found a White LED at radioshack that has a 20mA rating, which is the same as the LED on the BOM. So I should have no problems with this one correct?


It will work fine. It will just be very, very bright behind the default 2kohm resistor.

As a comparison, I have a 300mcd diffused blue LED on my Max front panel behind a 10kohm resistor. This is borderline too bright for me......
 
May 26, 2008 at 3:48 PM Post #4,820 of 6,727
I am going to try the dull coat, and if that is to bright, Ill swap the resistor.

Thanks Beefy
 
May 26, 2008 at 4:35 PM Post #4,821 of 6,727
Quote:

Originally Posted by agutt /img/forum/go_quote.gif
I have found a White LED at radioshack that has a 20mA rating, which is the same as the LED on the BOM. So I should have no problems with this one correct?

The MCD is rated at 1100, and the "Typical Voltage" is 3.6V

I would like to avoid having to swap out my resistors since I have soldered them in already...

Clearly my understanding of this in terms of electronics is poor...any explanation would be a great help!



The LED positions on the MAX are fed with V+ voltage, whatever you've set for that. Most use the recommended 27VDC. The LED and the LED resistor are in series, so they will share that full voltage. Meanwhile, whatever current flows through them will be the same for both. Ohm's Law establishes this relationship as V = I*R. So, if V is 27VDC, then we can establish "I" by choosing the correct value for R. If we were sizing it for the full current rating of the LED you mention above (it's typical), then we'd use 0.02A (20ma) for "I". Then R = 27VDC/0.02, or 1350 ohms.

Now, we don't really want to set a resistor for exactly that amount, because we can't be too sure of the rest of the components. If that 20ma is an absolute rating and our resistor is 5% low, then more current will flow and the LED might burn out pretty quickly.

That said, there is another factor to consider and that is the power handling ability of the resistor. Power = I^2 * R. For the values we just arrived at, Power = (0.02^2) * 1350, or 0.54W. Unfortunately, resistor power ratings go in fractional increments like 1/8, 1/4, 1/2, and 1W. So, for 0.54W, we'd need to use a 1W resistor. Even then, if anything was the slightest bit off, the LED would still burn out.

Luckily, the brightness curve for an LED is not really proportional. I don't know what it is exactly, but experience seems to indicate you can probably get at least 3/4 of the maximum brightness rating at 1/2 current. If we use 10ma up there as the current - so as to assume a decent safety factor on burning out the LED - then R = 27VDC/0.01A = 2700 ohms.

That's an odd resistor size, so let's pick a 2K resistor and see how that works. I = V/R, so I = 27VDC/2000 = 13.5ma. Hmm not too bad - still a decent safety factor, but with a standard-issue, common resistor size. If we check the power equation, then P = (0.0135^2) * 2000, or P = 0.365W. That's a bit more than 1/4W, but plenty below 1/2W to allow for a decent safety factor.

Hence, we chose 2K - 1/2W resistors for the LED positions.

You can follow the same logic by setting a higher resistor for a super-bright LED in order to knock down the brightness to something reasonable for a panel LED. Let's say we used the 10K resistor. Then we have I = 27/10000, or I = 2.7ma. Checking for power, we'd have P = (0.0027^2) * 10,000, or P = 0.073W. Power should have a much higher safety factor (extreme failure could result in flames), so even though a 1/10W resistor might work, a 1/4W would probably be better.

Hope that clears up some of your question.
 
May 26, 2008 at 7:00 PM Post #4,822 of 6,727
tomb...you are the man. Thankyou!
biggrin.gif
 
May 26, 2008 at 8:03 PM Post #4,823 of 6,727
OK so I did some math...

Since my LED runs at 1100MCD I want to run this at about 10%.

which would mean 2mA. Plug that into the math...

P=IR 27=.002*R
R=13500 ohms... again an odd resistor size so I will jump to either 15K or 10K

For 15K I get...

I=V/R
I=27/15000
I=1.8 mA (close to my desired 2mA)

Therefore:

P=(I^2)R
P=(.0018^2)15000
P=.048W

Now this value is well under 1/8 W, BUt at radio shack I cannot find a 1/8W resistor at 15K. Will this work if I use a 1/4W or 1/2W at 15K???

(If I go for the 10K which I might, It will be very easy to find a resistor, but will be brighter than what I desire)
 
May 26, 2008 at 8:05 PM Post #4,824 of 6,727
Also this resistor be getting very hot because of such a huge difference correct?
 
May 26, 2008 at 8:44 PM Post #4,825 of 6,727
agutt:

No worries with using a higher power rating. Power ratings ~ Max power. Its 'kind' of like saying "this resistor can be heated up to 100 degrees" so heating it up to 70 is fine, and getting one that can go up to 200 is redundant but in this case, harmless and even desirable.

Regarding the heat, the lower the current the less heat it will actually produce and the higher the resistor, the lower the current.

As you've seen, P=(I^2)R and and V=IR (I=V/R) you could substitute the second into the first and get P=(V^2)/R and since V is pretty much constant, P is directly proportionate to 1/R
Also, the actual power calculations here give you more margin than you think, since the actual voltage is actually [(V+) - Vled ]~(27-3.6).
The current can be more accurately calculated with that voltage (Vled can be more accurately obtained from the LED specs per a given current).. but since you dont need such exact values, and its easy to stay within the margins, if you have a ratshack around, it may not hurt to buy one of those cheap resistor assortment packs and play around with it..

-R
 
May 26, 2008 at 8:46 PM Post #4,826 of 6,727
Quote:

Originally Posted by agutt /img/forum/go_quote.gif
Also this resistor be getting very hot because of such a huge difference correct?


No. Power is what's dissipated from the resistor. Looking at the equation, Power is proportional to the square of the current. Therefore, very tiny current - not much power. Even if the rating of the resistor gets very big, it's not enough to overcome the effect of squaring the current.

Your calculations show P = 0.048W. That's not quite 50 millwatts. A 1/2W resistor dissipates the same as 500 millwatts.

BTW, the board will easily accept up to 1/2W resistors for the LED positions. There's no difference in these ranges if the resistor has a higher power rating - as long as it fits in the board.


EDIT: Uh-oh: dual post again.
smily_headphones1.gif


EDIT2: Umm ... as an ME (meaning this is one of the few things I understand around here), I'm not sure I agree with that analogy, ruZZ.il. The resistor will burn at the same temperature in either case. The difference is that one will take much more power before it reaches that same temperature.
wink.gif
Still, thanks for the post!
 
May 26, 2008 at 10:20 PM Post #4,829 of 6,727
Thanks guys! it helped alot!
 
May 26, 2008 at 11:44 PM Post #4,830 of 6,727
just an FYI, I went to pick up another power jack from RS today and after going to two different stores and only finding the new version, which won't work nearly as well decided to get one to see if the mounting hole was at least the same size, which it isn't. Even if it was it would have to be isolated as it's metal and the ground would ground to the case.

so the BOM should probably be updated and people encouraged to buy the ones from Mouser.

here's the new RS version:
RadioShack.com - Cables, Parts & Connectors: Connectors & connectivity: DC power connectors: Size M Panel-mount Coaxial DC Power Jack
 

Users who are viewing this thread

Back
Top