Show Posts
Pages: [1] 2
1  Topics / Home Automation and Networked Objects / Re: Simple RF controlled 4 input usb sharing switch on: January 23, 2013, 04:06:42 am
What about this-

If i understand that correctly, current limiting of the +/- lines, (if thats even necessary when wiring 3 pc's usb +/- up in parallel to one device constantly) and a circuit that would interpret a momentary ttl signal to hold a on ttl voltage to either 1, both, or none of the two control pins,  would accomplish what i need as far as a usb 3 channel switch, correct?

Or am i misunderstanding the logic as far as how sel1/sel0 switches between the 3 inputs? The way i'm looking at it, if neither sel1/0 is charged, its on the first input, charge sel1 and it drops to input 2, charge both and it drops to 3. Or is it no charge to either sel1/0, input1 is connected. Charge sel1 and it changes to the next input. Switch that charge over to sel0 and your at the 3rd input, rather than a no charge-I1, 1 charge-I2, both charge-I3 switching logic?

To me, my first theory seems logical, as it wouldn't know what input to select if both sel1/0 were charged.

If this is as simple as that, i can settle for a 4 channel rf remote using two channels for the usb switch, and one channel each controlling relays for the smps ac in/dc out. Id rather use a 2 channel remote, and a time delay trigger for the dc out relay/(s), but if its as easy as ttl on to sw1 or sw2 for channel switching, and no extra circuitry is required, then thats acceptable using latched on ttl signals instead of momentary ttl signals. 2 small dc 5v relays for the usb switching between USB1/USB2/USB3, and 2 heavy duty spdt's/2x2 parallel spst's  for the ac/dc switching. Correct?

I'm unsure as far as how the logic works on sel1/0 for the fsusb63, clarification would be much appreciated.

2  Topics / Home Automation and Networked Objects / Simple RF controlled 4 input usb sharing switch on: January 23, 2013, 02:46:18 am
If i wanted to design a simple rf remote controlled usb switch, is there an easier/less components required way to do this, than taking a 4x rf relay, and having each one of those relays cascade trigger 4x small dc relays to switch on/off the 4 usb pins?

IE, using the main rf relay board, to parallel close 4x small relays to open a a circuit to the shared usb device, doing it this way, means 16 relays, (4x rf spst's, then 4x relays per usb channel switched, for this application i need 3 usb inputs).

Basically is there a more simple way to design such a switch, using fewer components, and only allowing one usb input to be attached to the shared device at a time?

I want to run 3 computers usb to a switchbox, to share a usb->optical 96khz 24bit dac (well not really a dac, since its optical digital out, but thats how they are marketed). Then be able to only have to use one toslink input on a digital amplifier, since theres only two optical inputs, and i'd like the other for an adc module for an analog input to the amp.

Something tells me theres an easier way to do this than cascading relays, witch if done that way, would offer no protection in the event two inputs were switched on at the same time. I'd like some way to make sure that only one input can be connected to the "dac" card at a time, granted it would be parallel dc, so theres no overvolt risk, but i don't think its wise to risk it either, factoring in the cost of the usb->toslink adapter.

Do i even need 4 relays, or is there some kind of ic i can use to switch usb channels from a a momentary rf trigger coming from a single button rf remote?

The idea, is to be able to pack all this into an aluminum chassis, and with one rf multi channel remote, be able to switch 3 usb inputs to the usb->toslink adapter using only one  rf channel, and switch on a smps ac -/+ lines, with a delay trigger to connect the dc load (the ddx320 digital amp) after the smps gets a few seconds to power up  so i can get away with using the lowest channel count rf remote as posssible.

Ideally in this case, a 2 channel rf control/reciever module. one channel to switch on a beefy spdt(or parallel spst) to switch the ac mains, and the delay circuit for the load connect, two more beefy spst/one spdt for the 35v 9.6a dc out from the smps to the load (d-amp); the other channel triggering some kind of circuit that uses momentary logic to switch a usb device to 3 different computers, so i only have to buy one usb->optical module since the one i want isn't exactly cheap.

Any circuits that can do this, or ideas as to how to implement such a rf control system for ac power switching with a delay circuit to switch a dc load to the smps that is being turned on, and channel shifting between 3 computers usb inputs, ideally using a 2 channel rf remote device?
3  Using Arduino / General Electronics / Re: Blown Caps in 0-30v 0-10a bench supply repair on: January 12, 2013, 02:23:51 am
Ok, so where the positive output is soldered to the board, there is a diode soldered to the same joint, inbetween the replaced cap, and the two white (resistors?).

It shorts both ways, but do i need to take it out of the board to test it? Or can you test a diode in circuit (used both my diode test, and ohmic resisitance functon on my multi, no resistance either way). Also, a resistor, right next to that diode/cap, 1k ohm, is shorted, looks like its burned up, and shows no resistance when measured with a multi.

And i finally found a bit more info, and the brand name for this supply-

From what that thread is saying, these are pos units. If i have spare resistor/diodes of the same rating laying around, i may attempt to replace them, but i'm not spending any more money trying to fix something thats, well, a piece of crap.

Anyway, i'll dig around and see if i can find the resistor/diode needed, i'll update if i can find what i need.

Heres some pics of the diode/resistor-

4  Using Arduino / General Electronics / Re: Blown Caps in 0-30v 0-10a bench supply repair on: January 11, 2013, 08:23:36 am
Alright sounds like a plan. I have some time to tear it apart again and test any diodes that may have been fried w/ the reverse connection, so i think i'll do that and if all is well, fire it up and hope for the best.
5  Using Arduino / General Electronics / Re: Blown Caps in 0-30v 0-10a bench supply repair on: January 03, 2013, 07:11:40 pm
It was just a simple diode/short test, what would be the proper way to test it?

Check the resistance between the terminals?
6  Using Arduino / General Electronics / Re: Blown Caps in 0-30v 0-10a bench supply repair on: January 03, 2013, 05:13:11 am
Caps were replaced and shorts were soldered and epoxy'd over. Checked pos/neg out for a short, and sure enough it beeped at me. I'll tear it apart and start checking diodes, but at this point, i'd considering buying a new one, or looking into building one, with some bad-ass protection circuitry (diodes, fuzes, etc). Too late now to do any more work on the supply.

BTW, it wasn't plugged in when tested for shorts, and tested positive for shorts between the +/- outputs both with the power switch on/off. Does not powering it up have anything to do with it showing a short?
7  Using Arduino / General Electronics / Re: Using mosfets to switch low current from a high current gate voltate on: December 30, 2012, 11:37:11 pm
Alright, i take back what i said about it not working with a 5v input (the pwm dimmer that is). Using a 800ma 5v ac/dc adapter, i can get a 0-5v output using a 3k trimpot. Without the trimpot the voltage range is 4.6-6-2v. I tuned it down to 5v max when there is no dimming taking place, and can dim down to a few ma using this method. It did not work with a similar dimmer (5v input), but they have very different pcb's, and are different models.

So, knowing the buckpuck can read that pwm signal (from the arduino digital out pin), all in need is a low pass to make that 0-5v i get from the dimmer analog and input that to the arduino, correct?

Or can it read a pwm input signal (2560 mega) and no filter is necessary?

But that leaves the problem of current draw for the dimmer, i'd like to power the dimmer/arduino from one dc source. Since the 5v+ on the arduino can only provide 40ma (came across specs that current consumption of the dimmer is less than 500ma), powering the dimmer from the arduino's 5v+ is not a good idea. Assuming i can use digital read, to take that pwm input, and use it as a linear dimmer for a pwm output, something akin to-

But with digital read instead of analog in, since its pwm, is it possible, using the right resistance, to power the dimmer from the 9v1a the arduino is powered from, use a pot voltage divider (might be more/less than 3k, will have to expirement), and get a 40ma max pwm input from the dimmer?

In any case, after reading about rf filters, i assume since the dimmer can't be powered from the 5v+ on the arduino, i'd have to dig out the old heathkit scope to find the frequency of the pwm, and then devise some way to power the dimmer and limit the current it provides, along with building a rf filter circuit to make it analog.

Or am i overthinking this?

TLDR- Assuming i can use digital read to do something like analoginoutserial, to provide dimming without dac conversion from the pwm dimmer, IE pwm input, is it necesssary to limit the current provided to the input pin of the arduino when the power source for the dimmer is the 9v1a the arduino uses? If so, can this be accomplished by using resistors/trimpots alone?

Or do i need to figure out the frequency of the dimmer with my scope, and build a rf filter that will limit said current to 40 ma using the appropriate uf and ohm value cap/resistor to get an analog signal, 0-5v, 40ma max, from the dimmer when the dimmer's input is 9v1a, using a trimpot to voltage divide its output to 0-5v?
8  Using Arduino / General Electronics / Re: Using mosfets to switch low current from a high current gate voltate on: December 30, 2012, 11:11:16 pm
I understand what your saying. The application is a rf remote that can dim the buckpucks, possibly with another rf switch to switch between multiple drivers on different pwm pins.

This is the pwm dimmer i have-

Low pass filters may be necessary, like you are saying, but i think the buckpuck can dim via pwm and not just analog, as at one time i had a voltage trimmer pot hooked up to its output, and it worked to dim the buckpuck without an arduino in between to invert the logic. I'd like to either just use this dimmer, and another rf remote switch to choose what pwm digital output pin, and driver/led to dim, or one remote that has can do both. The latter option, i've looked for but havent found anything that fits said application.

If theres an easier/better way to be able to dim these drivers via a rf remote with/without an arduino then i'm all ears. I'm just working with what i have.

Is there a better rf device compatible with the arduino's analog in to fit this application that doesn't require low pass filters and optioisolators to change the voltage range?

I'd power the dimmer from a 5v source and just use a low pass filter to make it analog 0-5v, but it diddn't work when i tried, and i think it needs a higher input voltage to operate.

Like i said, i'm working with what i have, if theres a better rf remote compatible with the arduino's analog inputs, then i'm all ears, so long as its not too expensive and greatly simplifys the application. I haven't been able to find such a device.
9  Using Arduino / General Electronics / Re: Blown Caps in 0-30v 0-10a bench supply repair on: December 30, 2012, 10:44:48 pm

Well yeah, your probably right. Though i cannot be certain it was connected with reverse polarity. The fact remains the cap quality has been said to be very poor for said brand. In all likely hood it was a reverse polarity stupid mistake, i was just commenting that the caps in this supply have been known to be crap quality. So don't buy the model i did.

I'll  concede it was almost certainly reverse polarity that blew them, but the fact remains you get what you pay for, and when you buy the cheapest possible option, issues like this are common.  Not to say the caps would have blown under normal use, just that they don't have a good rep. Perhaps they would have lasted, assuming i did not make the mistake i did, perhaps they would fail due to poor quality.

The point, is that while it was most likely my mistake that blew them, and any cap would blow in that scenario, they used poor quality components in this supply, and i got what i paid for.

In any case, the replacement caps arrived, and i'll be installing them once i have someone (my father) with a background in electronics take a look at it first. I'll update as to if the repair was sucessful or not when we have some time to take a look at it together. Forgive my rudimentary understanding of electronic circuits, and assumptions it was the low quality components and not user error that caused the failure. The fact remains though, the caps are known to be poor quality.
10  Using Arduino / General Electronics / Re: Using mosfets to switch low current from a high current gate voltate on: December 29, 2012, 10:04:55 pm
the function the arduino would provide is using its pwm digital out to dim a buckpuck via ttl.

The issue is converting a ~6-12v 8a rated rf pwm led dimmer, into something that the arduino can read from an analog input and dim the buckpuck accordingly. Linear translation of that 6-12v into 0-5v is the issue.

SEPIC, a google search could have provided you with an explanation of said acronym.

The reason i worry about the dimmer sinking all the current from the 12v dc source it would be sharing with the arduino is because its designed for led's. I could be wrong and that it follows conventional logic that it only consumes as much current as necessary to provide power for the pwm/rf circuitry, and the optoisolator circuit to get a 0-5v linear signal for the arduino input.

Ill draw up a sketch, and edit this post tonight as far as the circuit.

The basic explanation of what i want to to, is power a mega 2560 and 6-12v pwm dimmer from the same 12v supply, and use the dimmer along with the optioisolator linear voltage converter to take tht up to 12v (really 12v all the time since its pwm), and along with the optioisolator circuit, get a 0-5v analog signal to use as an arduino input to pwm ttl dim the buckpuck, per the diagram on how to do so in its datasheet.

Since the logic of said buckpuck is 5v 0% power, and 0v 100% power the arduino could also switch that so the up/down buttons on the rf remote aren't reversed.  Alternatively just don't use the arduino at all. Just the dimmer/optioisolator and the buckpuck to dim the led, albeit with reversed logic on the remote (down is up up is down) due to the way the buckbuck reads the 5v ref it supplys, and its control pin (short them, and no power flows to the led, put a potiometer in between and no resistance is 0% dimming, 5v flowing, full resistance, or not shorting the buckpucks 5v ref/control and you get 100% power to the led's). Basically just backwards from how most other ttl dimming works.

Edit- Circuit Diagram attached. The only thing i might need to add is a low pass filter to change the digital pwm coming from the dimmer out, to an analog signal. No idea if the optioisolator will work with a pwm digital input.
11  Using Arduino / General Electronics / Re: Blown Caps in 0-30v 0-10a bench supply repair on: December 29, 2012, 09:43:59 pm
true, true.

The transistors tested ok, so i'm just waiting for the new caps to see if thats going to fix it. The cap's in question, i'e heard nothing good about, so quality issues definately could have been the cause of failure just going off that.

Or the caps could have been from that bad batch made with the stolen seret suace electrolyte formula from a few years back. Nothing is certain, other than the quality of this supply fits in line for what i paid for it. Lesson learned.
12  Using Arduino / General Electronics / Re: Blown Caps in 0-30v 0-10a bench supply repair on: December 28, 2012, 06:12:50 am
With all due respect, since you seem to be somewhat lacking in knowledge on what a power output transistor looks like, then your chances of identifying the problem are pretty low.  Without a circuit dagram, someone who really knows what they are looking for will have to dig deep into their knowledge of circuitry to identify where the faults lie.    As previously stated, your best bet is to simply replace the damaged capacitors, switch ON and if it works, all well and good.  If it doesn't work, and you really want to keep it, then pass it onto a colleague who has the skills to trouble shoot.  Otherwise you could well be throwing good money after bad, with little chance of success.

This sounds like good advice. I don't have much skills in circitry, but can test componenets and understand the basic principles behind the supply. I taught myself alot about dc electronics, and am just wading into learning how circuits work, so this may be beyond my skill level. I ordered some quality caps, ~5 or so of each (35v, 50v 470uf), i don't much care for radioshack, and would rather give an ebay seller my business, but regardless. They should be here in a few days.

If the caps/shorted lead patch/solder joint fix the problem, then huzzah. If the transistors are bad, i can replace those if necessary as well, so long as the repair costs are under $25 total. I'll have someone with a proper electronics education take a look at it (my father) after i replace the caps, fix the trace, and replace any transistors/diodes that may have been damaged, before powering it up again for saftey reasons. If he says scrap it for parts, thats going to be the gameplan unfortunately, he's fairly well educated in the matter, though has not worked in the field for almost a decade now. Still, he might be able to show me how/what failed, and if its repairable. Busy week with lots of snow in norcal, so it might be a few days before he has some time to take a look at it though.

Might make a nice vlra battery case you could stuff a charger in hooked up to a solar panel if its unrepairable. And i'm sure theres plenty of undamaged components i can scrap for future projects so its not all bad.

Its also relevant to know, it did not pop, and die immediately. The voltage/current display remained lit, but the loud cap pops and smoke made me disconnect everything as quickly as possible. I have not powered it up since then. i'll wait for the caps and trace/fix before plugging it in again.

Live and learn i guess. Should have just spend a  bit more on a mastech. Don't buy these knockoff pos's. Its like 50-75$ more for something quality and protected against stupid mistakes like reverse polarity connection of a fla.
13  Using Arduino / General Electronics / Re: Using mosfets to switch low current from a high current gate voltate on: December 27, 2012, 01:59:21 am
After looking at my mega 2560's input voltage specs, it looks like i can just use 12v to power the board, and eliminate the sepic for the dimmer. So all i need to figure out, is how to limit the current that flows through the 12v pwm dimmer, and its ~5-12v output that gets converted to linear 0-5v using the optoisolator circuit. Thats simpler and more effecient than adding a dc/dc converter to bump 9v up to 12v for the dimmer, when drawing the power input for the dimmer from the arduino input supply.
14  Using Arduino / General Electronics / Re: Using mosfets to switch low current from a high current gate voltate on: December 27, 2012, 01:37:30 am
What i'm thinking, after giving this some thought, is as follows-

Use the 9v input supply for the arduino to power a sepic converter to give me 12v dc cv to power the pwm dimmer. Take the up to 12v pwm out from the dimmer, and use this circuit to allow it to take the pwm 12v, and give a linear 0-5v out for the analog in-

The second optoisolator one.

The question i have, is regarding resistor values, or potiohmeter values for a (lets say 9v 1a ac/dc wallwart, or a current limited dc/dc sepic), to power the pwm dimmer and arduino with minimal power loss to heat in the resistors to limit the current draw from the ardiunos dc supply (9v 1a) that powers the 12v dimmer, and where to put them (IE before the sepic, inbetween the sepic/pwm dimmer, etc).

Since it needs very little current at the input of the pwm dimmer, but 12v regulated (where the sepic comes in handy, but i'd rather use resistors than a more expensive current limited sepic), whats the most effecient resistor values to use, and where in the circuit to put them to cause the sepic drawing from the arduino 9v input, to only pull 100ma max, but allow 12v cv to the dimmer?  100ma, derived from some effeciency loss at the converters, power for the rf pwm dimmer and 40ma as the usual current when using the 5v+ arduino pin for analog input applications.

IE, i want the dimmer circut to be current limited to ~100ma (give or take, i'd like to use a pot if thats too high/low after testing) when drawing from the 9v 1a dc source that powers the arduino. My thinking is without current limiting, i would be sinking all the supplys amps into the dimmer/optoisolator 5-12v -> 0-5v circuit, when only 100ma max is necessary, to power the rf pwm dimmer, effeciency loss at the sepic, and the 40ma for the 0-5v to the arduino. Since this is something that may be drawing from backup lead acids at some time, i'd rather it use as little wattage as possible for the pwm/optoisolator 12v to 5v conversion, and theres no point wasting power anyway, regardless of a ac/dc source, or dc battery source. I just want it to be as effecient as possible.

Basically, what resistor values/pot values and where to put them in the described circuit is what i need help with. I can draw up a complete circuit diagram including the sepic and pwm dimmer if that makes my question easier to understand, just let me know.

15  Using Arduino / General Electronics / Re: Blown Caps in 0-30v 0-10a bench supply repair on: December 27, 2012, 12:24:52 am
Alright. So when you guy are talking about the transistors that may have been shorted/fried, are they the two square white ones on board near the blown cap? Can someone point it out to me in one of the pics? There just a good # of transistors (mostly the small black d shaped ones) on that board, and i don't know what ones are the output ones.

And i definately agree its a pos supply. The only reason i'm attempting to fix it, is because in all likely hood its going to cost a few dollars for the caps, mabey a few more for the transistors. Even if its not capable of high amp output, if its repairable, the parts cost justify doing so relative to how much i paid for the device. At 3 amps it has plenty of uses, and eq charging fla's doesn't need high current (nor is it advisable) anyway. Just to have a cc/cv supply for charging batteries would be nice, even if it cant pump out 10a. I can't really imagine a scenario i would need the amperage its "rated" for anyway.

But a quality switching supply with overload/overvoltage and reverse polarity correction, like one of those EX mastechs, is defnately what i want, eventually. Just don't want to spend ~$175 on one atm. I'd rather buy a few caps, and possible a few transistors/diodes and make do. And if this is a linear supply, then i could have both, considering they both have their advantages/disadvantages when i do get a quality switching supply.

Also, if i got some heavy duty diodes, is there any way i could wire them up to the output terminals to prevent reverse polarity damage in the future? Perhaps some heavy duty schotty diodes? If so, would it just be connnecting them from + to -, one only allowing flow in one direction, and vice versa for the other? Mabey add some fuses there as well, since the fuses on the board and ac input plug diddn't prevent this failure? I'd imagine there some fairly simple way to use a few diodes and a fuse to prevent this from happening again (reverse polarity to a fla starter battery), or any load that can push back into the device the way the battery did.

I ordered some quality caps to replace the blown ones, and will test the transistors with my multi once i know which ones you guys are talking about. As far as the trace damage, i'm thinking a solder joint, and an epoxy coating (After replacing the cap of course since its fairly close to the solder joints) is the best way to fix that prob and prevent future arcing there. Is that correct? Or should i just epoxy the trace gap, find the nearest solder joint on each end of the trace, and solder a 16gauge copper wire to bypass the broken trace entirely?

Pages: [1] 2