Ldanielrosa

I wouldn't recommend it. Perfect matching can't be guaranteed, and failure to use resistors to distribute the current could result in one or more triacs turning off and the remaining would overload and perish.

Ah, I see. I have no experience with the BS series. I did just look at some of the literature though. I have not determined whether or not you can access TMR1, but even if you can the BS1 uses a ceramic resonator so the frequency will wander. Alas, the external hardware is necessary. Just the same, it'll be an interesting clock- a decimal day.

Well jcbarber, you have an almost limitless variety to choose from. You can expect better accuracy from crystals than ceramic resonators, the flip side being that resonators are usually cheaper- but for one-off projects that won't matter much. You said you wanted to use 100Hz for your master clock signal- I will assume that you will be multiplexing the display at this rate, among other things. So it is of primary importance that your crystal frequency divide by this. Beyond that, binary ripple counters are easy to find and simple to implement- minimal circuit complications. So look for frequencies that have binary values and end with "00", like 1,638,400. A lot of these you will not find, but you may find 1,024,000- the key difference being the number of zeroes. So you may find that you need one stage of decimal division. Another option is choosing another frequency for your master clock, that way you can use ripple counters for everything else and have the wierd number worked out in the last few stages. Good luck.

I won't bother quoting the exact words they used, but the way I interpreted it was: "If you're smart and/or educated, we don't want your business!" Did they mean to insult me/us? Maybe I just want an excuse to get irate.

I'm working on a house that needs nearly everything brought up to code. While the walls were open, I put in enough media drops to run an arcade (30 ethernet points, assuming I don't start using the 'phone lines too). I gave it some thought, and decided not to include any heavy wire- so no speakers without a remote amplifier. Anyway, the distance is moderate so the wire resistance may be as high as 2 ohms from the distribution point to the farthest box. This could be highly undesirable if I wanted to power something through the signal wires, however... The query is this: Has anyone encountered anything whereby a moderate impedance, moderate negative voltage (-12 => Vsource => -48) goes through a switcher to provide a low voltage (=> -1V) for an alternate current path to keep ground stable? Does this make sense to anyone else, or am I babbling out my navel?

I used ot tear just about anything apart. Now I try to limit myself to things that I know have something I want in them- like nokia 5160 (and similar) cellular 'phones, they have a graphic LCD with serial input.

I have one that's close to that- I think about 150*200mm. It is buried away at the moment. The other problem is I'm on another continent, to shipping would be stiff. If you're that desperate for it, I'll dig it out and see if I can find a rating on something similar and look into shipping options.

With nickel chemistries, 1% per day. Maybe more early on.

Apparently articles that far back aren't available for online viewing. I will inquire further at Motion, but I may have to look elsewhere. Thanks.

I've got another wild idea. I"m thinking about putting a little ferrofluid into a stepper motor to see if there will be any improvement. I'm wondering if it will do more harm than good. I know there will be dome inertial damping, but I don't plan on running it anywhere near full speed. Will the improved magnetic coupling do terrible things to the power requirements? Will it permanently lock up? BTW, I don't yet have the ferrofluid but I have supplies on order to make some and I'll be trying a few different carriers. If anyone has information on making magnetic viewing film, please let me know.

Well, with a dummy load you're facing a challenge turning the energy into anything besides heat. I don't know if it would be worth your while, but you could build your own resistors from a heating element. This would allow you to make whatever load you want. It would also reduce the startup surge that bulbs have.

That's a good idea too, Ante. I've already started thinking on the disturbances part. I'll use a triac driver with zero-crossing circuitry (I have about thirty on hand) to limit the interference. I'll also use at least one full cycle at a time to balance the load, but I don't think I'll go so far as counting the half cycles for longer periods. As for saving energy (and safety too) I'll include a time limit, after which the oven will turn off. I've left the oven on over night a few times. I'll also have it shut off if it goes far over temperature. The circuitry inside the little bricks will be simple enough, but finding the right hardware is becoming frustrating. Standard nuts are too small/shallow and have no cavities for theepoxy to fill for a good grip. Brass hardware is even more limited, and expensive. If I could find predrilled brass cubes in the right size, then tapping them would be no problem. I'd given some thought to using a base plate. It might simplify a few things, but then I'll also have to cut aluminum rectangles. Keep the feedback coming- it's making me think.

I don't know what more there is to tell about the oven. The thermostat is off by enough that a cooking thermometer has to hang from one of the racks to give a better indication than the setting on the knob. By cabinet mounted, I mean that it is not in the same enclosure as the range top- that is mounted on the counter a few meters away. I figured I'd replace the thermostat with a microcontroller, since I'm looking for a cause. I'll have to use four SSRs because there are two heating elements, and I don't want to leave one half-energized when it's off. An LCD and six button keypad for human interface, and a MAX6675 to convert the thermocouple to a digital value for the uC.

I might. I have a cabinet mounted oven with a failing thermostat. The appliance shop tells me that the oven is so far outdated that they can't locate a replacement part for it. Since it is in otherwise good condition I was thinking about using it as an excuse to build a PID controller. After seeing the leakage specifications on SSRs, I may use mechanical relays. Thanks for the lead, MP. Their specifications are fairly clear. I haven't found their prices to be too different from other manufacturers', I may need to find a different distributor though. My searches on JB Weld have turned up some interesting discussions. There seems to be some disagreement on how electrically insulating it is. The fact that it _is_ a filled epoxy (steel) came up early. For an SSR it shouldn't matter that the potting compound is magnetically responsive, but that may complicate things for a power supply. Please keep the input coming.

Thanks for the fast reply, Ante. I haven't looked further than a couple online catalogs for price, but they seem to start at $30 each for the molded block type. I'll definitely have to look on ebay, but the selection there is not necessarily stable. Anyway, the application that prompted me to consider this will require four of them that can handle 30A/240V. I've no doubt that the circuit is easy, even if I'm using two SCRs back to back instead of a triac. The hard part is making it pretty. You're probably right in the frustration/cost analysis that improvising a mold and working the component layout and optimizing the mechanical hardware may very well take twelve hours plus (at $10 per hour) giving me at best a break even on buying them.

There are some projects that may be easier if I had a few SSRs on hand, but between the price of them and the parts I have laying about I'm considering making my own. I have SCRs good to 50A and 600V with isolated tabs (in rediculous quantity), a few triac optoisolators, and probably all of the other components to make them. Here are some of the issues I'm curious about. Will JB-weld be a good potting compound? I believe it is electrically non-conductive, but it may contain metal dust to improve thermal properties. I'm pretty sure I'll want to use screw terminals for durability and repetitive connecting/removal. Will brass (likely more expensive) be desirable to steel after considering strength, solderability, conductivity? Is there a specific piece of hardware that lends itself well to this? Since a lot of optoisolators have compatible pinouts with wildly different internals, I was considering leaving a 6 pin DIP socket exposed. This would allow me to make the relay in advance of deciding upon the driver (zero-crossing detect or not), but would also have a few drawbacks (more exposed mains level metal, a potentially loose part). edited for spelling

A low dropout voltage regulator should do the trick. They often have a very low quiescent curent. Depending on some details pertaining to your email address, you may be able to get free samples from Microchip. http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=64 Look under analog & interface products. If you are not disqualified due to a bulk email address, you should be able to get some goodies within a week.

What is the difference between the left hind wing of a duck? The higher it flies the much!

My dad has one that looks very similar to that one. He was very proud of having a torch that needs no batteries. I looked inside and guess what- I found two lithium coin cells (non-rechargeable type) hidden inside.

At that size, you're probably better off making your own 7-seg from arrays of individual LEDs. That will give you ultimate control over color, size, and intensity.

Thanks Mixos. Berty's page helped a lot. I had a glitch for a while (probably messing up the !reset pulse) but now its running. Right now it's displaying the output from a PRBS routine. I saw right away that wasn't going quite right, now it's fixed too. It ended up being a great test platform for that. So my LCD is showing systhesized static as fast as the microcontroller can send it. Now I need to make something.

By this do you mean that the relay you are controlling is running on DC? If so then that's likely your problem. Triacs (and SCRs too) are regenerative switches- when you turn them on they latch. To turn them off you need to interrupt the current. They are very useful for AC becauce for every voltage zero-crossing the current crosses zero as well and the thyristor turns off.

Your scope leads may be causing the problem. If the either output form the bridge rectifier is electrically grounded then the scope ground lead may be providing a current path that you hadn't planned on.

Um, mind if I ask what this is for that a transformer won't fit? And assuming it's not a terribly hazardous application, can you be more specific about the power requirements of the fan?

It might have something to do with your profile. TI/BurrBrowm has always sent me everything I've asked for (and some of it was pretty expensive), but then I've never requested that many part numbers in one fell swoop.