Use the LUMEX LED, is my recommendation. Why go to all the trouble to buy cool panels to make it look like MOTM, if you aren't gonna make it look like MOTM<g>? The Lumex LED should be set up to draw 8-10 mA, no more. It may be capable of drawing more, but your power supply doesn't need the extra load, it's plenty bright with 10 mA, and that's how Paul sets it up in the official MOTM modules, so you may as well match it. It has been my intention to put up a web page describing the MOTM panel conversion, and give details about the bypass switch and the LED resistor. I just haven't gotten around to it because I'm on a mission to build two big wooden cases, come hell or high water, before I do anything else. <LONG WINDED TECHNICAL EXPLANATION> An LED's current draw, hence brightness, is controlled by the resistor from the LED's cathode to the voltage source. Looking at my Time Machine schematic, the voltage source is the output of the opamp on the first schemo page labelled U12d, which is configured as an open loop comparator. NOTE: THERE IS A MISTAKE ON THE SCHEMATIC. ALTHOUGH LABELLED U12d, IT'S ACTUALLY U12C. THE PIN NUMBERS ARE CORRECT AS SHOWN. R36 controls the LED current. R37 and R38 form a voltage divider between ground and V+ to provide a fixed 1.7V reference voltage to the inverting input of the opamp. The input signal feeds the noninverting input. Since the opamp is a very high gain (10,000x or so) differential amplifier in the open loop configuration, the output clips to the positive or negative supply rail depending on whether the input signal is above the reference voltage or below it. If the output is at the positive rail, current can flow through R36 and the LED, illuminating it. To calculate the current, we need to know the following: what is the maximum output voltage of the opamp (the positive supply voltage for U12), what voltage is the anode of the LED (terminal to the right on the schemo) connected to, and what voltage is across the LED when it's active. Then you can figure out what the voltage across R36 will be, and hence the current through it. Using good old Ohm's law V/I = R (where V is voltage across the resistor, I is current through the resistor, and R is resistance), we plug in our known voltage and desired current to calculate the resistance. Referring to the schemo, the anode is connected to -15V, and the positive supply rail for U12 is +15V. This is good news, because we will be drawing a little more current than the stock LED. Some ICs onboard the TM are powered from onboard regulators, and we would have to be careful to avoid drawing too much extra current from them if they powered this LED. Since your external supply powers this chip, we are safe pulling a few extra milliamps. The LUMEX LED has about 1.5V across it when active (I've measured it). Analyzing the stock circuit and assuming the same voltage drop across the stock LED, we have a total voltage drop across the resistor of 30V - 1.5V = 28.5V. Divide that by the current R36 resistance of 5600 and you get about 5.1 mA. We want 10 mA, so just about halving the resistance should do the trick. 28.5/.010 = 2850, so a 2.7K to 3.3K resistor will do the trick nicely. </LONG WINDED TECHNICAL EXPLANATION> <SHORT EXPLANATION> Replace R36 with a 2.7K to 3.3K resistor. </SHORT EXPLANATION> Here's a general caveat about current capability - always, always make sure your system power supply is capable of supplying the current you need. It's best to have one loafing along at about 60% - 75% of capacity. If I have a system that requires 1A total, I'd be comfortable with a supply that supplied 1.5A. It will run cooler and more reliably. Very bad things will happen if you try to draw more current than your supply can handle. Your lovely high-fi and stable MOTM modules will perform like sh*t if they don't actually burn components. Larry, we need to make sure that as we come up with popular DIY modules such as the 822 and 831, we provide a published current rating for them as well! Moe > ----- Original Message ----- > From: Cap'n F.M. Bleep <bleep@w...> > hey there... > > the time machine is imminent! there's only one last detail to smooth > out... which LED to use? the lumex leds that paul uses (like on the 320) > draw 10 - 25 mA, but mr. blacet's led only draws 2, so a lumex would be > quite dim. (he suggested i use the supplied led with an led holder) > > how are you other time machine owners dealing with this? are you cleverly > placing a resistor somewhere? i'm not very clever with resistors... > > bleep. > out.
Message
Re: time machine LEDs / current draw (longwinded!)
2001-01-21 by Dave Bradley
Attachments
- No local attachments were found for this message.