Tech Forum Questions

The linear type wallwarts operated on line frequency of 60 cycles, and even though they only have 3-6 parts, these had to be large to handle the current and squelch ripple.  What you are looking at are the newer units which are switch mode type supplies, operating in the 100-200 kilocycle range.  Even though they may have 40 or so parts, these parts due to surface mount technology, and the high frequency, can be much smaller and lighter.  These supplies also provide a stable voltage of more current and less load droop than their older linear cousins.

I think an couple of quad op amps would enable you to scale the temperature range you want to 0 to 5 volts. That way, the resolution would be around 0.5degrees on each a/d input.

I suggest you consider the Maxim MAX31820 temperature sensor. Its accuracy is +- 3,6 Degrees F over the range of -67 F to 257 F. It has a digital output (rather than analog) with a "1-wire" interface using only 1 controller port pin. A large number of these sensors can be paralleled on the one pin so you could even put several together and average the readings. For a datasheet, see: [url=https://datasheets.maximintegrated.com/en/ds/MAX31820.pdf]https://datasheets.maximintegrated.com/en/ds/MAX31820.pdf[/url]

Have you considered a digital sensor, such as the DS18B20? They are inexpensive (about $4), readily available, is accurate to +/- 0.5 C (so, about 1 degree F), wide temperature range, and then you only need 1 digital pin. I used one in a little oven controller that used a PIC, a while back. Programming is a little more work, and it can be a little tricky to use with longer cables.

Your proposal will not work. It reminds me of students who try to connect two 8-bit DACs to make a 16-bit DAC. It just doesn't work. But I can't offer a solution without more information. What's the resolution of the analog-to-digital converter(s)? Do you mean you can set any of the analog inputs for a 0-to-0.5V range anywhere within the 10-volt span, say, from 1.2 to 1.7 volts? What is the voltage output from the thermistor circuit? More information, please.

You don't say what kind of controller you are using but if it has SPI capability, why not use a DS18B20 serial sensor. It is accurate to .5°C in the temperature range you specified and software routines for most MCUs are readily available.

Your calculations are probably correct, however you may have used the worst case current from the datasheet, if you use the nominal current it should be good. A couple of months is the expected retention at room temperature. Just be careful if you use schottky diodes in your circuit (to power the RTC from normal 5V) the leakage current in a poorly selected diode can exceed the standby current. (Schottky diodes are made with 3 doping levels, producing forward drops of 200, 300, 400mV, the 400mV has lowest leakage current, it will usually have a H in the part number). A supercapacitor is a electrolytic double layer capacitor (EDLC), and each electrode is coated in very fine carbon granules, the total surface area is about 1000 times higher than just aluminium foil, with a much thinner dielectric, hence the increase in capacitance (and drop in maximum operating voltage). There is no actual oxide layer for the dielectric, an EDLC has charged layers of ions a few molecules thick instead, and charge and discharge just move the ions back and forth across the layer.

“Super-caps” are, indeed, electrolytic capacitors. They’re construction typically uses tantalum (not aluminum) plates to obtain a large capacitance value in a relatively small package compared to standard aluminum electrolytics. Plus, their electrolyte formulation differs enough from aluminum capacitors to allow a larger charge vs. size capability.

Because of their large charge capability, they are ideal for use in short-term backup applications where low currents (typically, less than 100 microamperes) are required (i.e., real-time clock chips). However, they are not the same as a battery (i.e., lithium coin cell) as, like regular electrolytics, super-caps do self-discharge over time and they can not deliver a large supply current (i.e. >1 mA) for longer than a couple of seconds.

Personally I always use the "machined" gold plated contacts (Machined contacts are round, and have 3 to 4 gold fingers internally) the cheaper tin sockets require more force, and this force can (and does) bend the IC pins. Once you get above 18pins it is really difficult to insert IC's into tinned leaf sockets.

Gold plated contacts are highly resistant to corrosion. If you are designing equipment that is going to potentially operate in humid environments or where reliability is paramount, then using sockets with gold plated contacts can be a good investment. For most applications though, conventional machined-pin sockets provide excellent performance and the added cost of gold plating will not buy you much.

Most consumer equipment uses inexpensive stamped pin sockets and even those are normally adequate.

Depends on the environment that the project will be operating in. Gold plated sockets are usually specified for high-reliability, that is, in mission-critical or life-critical applications. Tin plated contacts don’t like lots of high frequency vibration, such as near heavy industrial equipment. It also doesn’t tolerate humid environments. Gold plating is great for use in humid environments, and is better than tin in high vibration environments.

There’s not a lot of difference in contact resistance, and you won’t gain anything with gold. If your project will be used in a humid or corrosive environment, then I suggest a conformal coating be applied over the entire circuit board. Gold plating comes with its own unique problems, the main one being separation of the gold plating from the plated surface, causing a failed connection, or worse, an intermittent connection.

I’ve used tin plated sockets for many years, and never a problem that could be traced to the sockets. My money is on the tin plated sockets.

Always go for the gold with chip sockets and “wire wrap” socket pins! The main reasons are:
 

1) Excellent corrosion resistance, which translates to more reliable connections. Tinned contacts will electrically degrade due to oxidation, even after the chips are inserted.
 

2) Better spring (mechanical) contact between the socket and chip pin. Tinned sockets are simple “leaf compression” types that are prone to “chip creep” (i.e., the chips push out of the socket over time) caused by thermal effects (i.e., heating when powered on and cooling when powered off), which means you have to occasionally re-seat the chips (i.e., pushing them back into their sockets with a “crunch-like” sound when they re-seat). Gold sockets are machined and use “grip fingers” that, literally, clamp down on the chip pins, virtually eliminating chip creep. Also, after crunching a loose chip in a tinned socket, it’ll creep out again at an increasingly faster pace as the tension in the leaf springs increasingly degrades with each re-seat operation (also resulting in degraded electrical connections - see 1 above).
 

3) Mechanical security of installed chips. Referring to 2) above, a creeped-out chip will eventually fall out of its’ socket if the board is jarred. Chips installed in gold sockets, again because of the clamping effect on the pins, will never fall out if the board gets jarred. FYI: I type this from 35 years of industry experience having to find and re-seat loose chips in tinned sockets that caused intermittent (and frustrating!) device malfunctions. I rarely (almost never) had creep, etc. problems with gold-plated (machined) chip/wire-wrap sockets.

I would get the less expensive tin sockets, unless you will have the project in a corrosive environment (and you would also have to deal with possible wire corrosion). In using many sockets over the years, I have not had any problems with tin. Get the “springy” side-contacing socket like (www.digikey.com/product-detail/en/A08-LC-TT/AE9986-ND/821740) rather than the machined pin type of socket. I have had some machined pin sockets making bad contact with the IC.

The short answer is no, you can’t use a variac on an induction motor.

The speed of an induction motor is controlled primarily by the line frequency, not the voltage. If you try to limit the voltage with a variac the motor will likely overheat as it draws additional current to try to compensate for the lower voltage.

What might work for you is a universal (brushed) motor. They are called universal because they can run on either DC or AC. This is the type of motor that is used in home treadmills, handheld routers, jobsite circular saws, etc. The speed can be controlled with a SCR (variable phase) speed control. These motors are usually rated for 120V single phase, up to about 15 amps or so. You could conceivably use a variac to control one of these but the SCR or Triac variable phase speed control will provide better low speed performance.

The speed of an induction motor is dependent primarily on the frequency of the alternating current driving it. Speed controls for 3-phase induction motors are often called VFDs which stands for Variable Frequency Drive. They work by converting the incoming AC power to DC and then use a 3 phase inverter to convert it back into AC with the ability to vary the frequency from as low as a few Hz to 120Hz or more for motors that can tolerate the increased speed.

If you attempt to reduce the speed of an induction motor by controlling the voltage with a Variac, you will likely overheat the motor without it slowing down much, because the current will increase as you reduce the voltage. Inexpensive variable speed power tools are powered by universal motors that have brushes and a commutator and these can be controlled with a Variac, though the common triac light dimmer type circuit is far more economical.

If you can find a grinder powered by one of these then your idea will work, but most bench mounted tools use induction motors.

Also note that the nominally fixed speed motors of any type usually  contain an integral cooling fan under a cover, mounted on the shaft. These are good enough, but if you run the motor slower for a significant time, you could fatally overheat the motor.

If it gets hot, add a separate cooling fan so the motor can take It!

Fuses are specified according to the circuits they are designed to protect. Unless you are certain that the original design specified a fuse of inadequate capacity or excessively rapid response, you should never attempt to substitute a higher current or slower acting fuse. If your power supply is blowing fuses, you need to determine why this is happening. Are you overloading the power supply? Does it have an internal fault? Do the fuses you are using match the original type specified by the power supply?

Properly designed and operated equipment that is in good working order should not ever blow fuses.

It’s not a good idea to over-fuse your power supply (PSU). The fuse was designed to protect the supply from damage (and fire) if it becomes defective internally or is operated beyond its design limits. You are probably exceeding its capabilities in some way, such as overcurrent due to too heavy load. If you’re not exceeding its specifications, there might be something wrong inside the PSU, such as bad filter capacitor(s) or a defective power transformer.

Have you checked the output under load with a scope? That will tell you if the PSU has high ripple under load, an indication of poor filtering inside the PSU. I suggest that you do a little investigation to determine whether the blown fuses are due to trying to operate it beyond its capabilities or bad component(s) in the PSU.

Your multimeter is a good tool to help do this. Watch the needle or display and see what the output voltage does right before the fuse blows. Analog meters are better in this situation. The first thing that comes to mind; does your load have a large capacitor that needs to be charged by the PSU? Large capacitors need high values of surge current from the supply until they acquire a full charge. If that’s the case, you might lower the value of the capacitance at the load. Use your multimeter as an ammeter to watch the current to the load. Is it at or beyond the specificied rating of the PSU? Again, an analog meter is best.

If you’re operating the PSU at its limits, the internal circuitry could be overheating. Mount a fan or blower so that it directs air over the heat-producing components (heat sink, power transistors, power transformer). If you absolutely need to run the PSU at its limits, you might consider getting a more robust PSU. It will be more likely to survive.

Cheap PSUs are sometimes over-specced, meaning that they meet specs only under very controlled conditions. Also, is the PSU rated for full output continuously? It might be overheating if it’s not rated for continuous operation.

The last thing I can suggest is to check your mains voltage to the PSU. Is it at or near the high limits of the PSU? If so, you might use a Variac or bucking transformet to lower the mains voltage to the supply. Hope that gives you some ideas that will help determine why the fuses are blowing so frequently.

The first thing you need to discover is why you blow fuses!

1) Are you trying to directly measure “current” by putting your meter leads directly across the supply’s output terminals? (a mistake always made by first week electronics students)

2) Does the device you want to power demand more current than the supply can deliver (i.e., 1A supply trying to power a 10A device)?

3) Is there a short circuit in the device you’re trying to power? Use your meter to measure the resistance between the device’s power terminals: a reading less than 10 ohms indicates an unwanted short circuit.

Find out what’s causing the supply fuse(s) to blow before you try substituting higher-value/slo-blo types. The power supply will be much happier overall and you’ll minimize the chance of having the supply blow up altogether!

Do not use a fuse with a higher amp rating. Try the same value fuse in a slow-blow fuse. A fast-blow fuse with a higher rating could lead to blown components, melted wires, and possibly turn your power supply into a smoke generator.

Search online for “firefly circuit 7555” and you will find what you want. (Or “unijunction transistor flasher” for the old fashioned method). The LED should be a high brightness type with diffused lens, at least 300mcd. The LED’s above 1CD typically use clear lens, so can’t be seen off axis. As an alternate, if you only need it to run a couple of days, then just solder a L-36BSRD flashing LED inside a 9V battery snap, I make these up for cavers, they leave them behind like breadcrumbs to find their way back.

It appears the output of the Sirius unit is supplying “phantom power” to the TransPod. IOW, there is a small DC voltage (1-2 VDC) present on the output signal that the TransPod is using as its’ power source. In a way, it is “magic” as phantom power schemes are regularly used in audio PA systems to supply power to electret microphones without using external batteries.

A quick Google search for “speaker reconing” will result in 100’s if not 1000’s of hits for this service or parts. Many offer a DIY kit while others are businesses offering the service or an exchange program for speakers.

We know of one business in our area that does the reconing/refoaming service:
TV Service in Crete, Nebraska. His webpage is www.reconespecialists.com
Phone: 402-826-3540
My personal opinion, I would enlist the services of someone experienced unless you really feel confident in doing the work yourself.

If the cone on your Cerwin- Vega speaker is still good, why not consider purchasing a surround kit. They are available online and maybe at your local electronics parts store. Go online and look up your particular model of speaker’s surround kit. They are usually priced around $20-$30 depending on the make and model. It’s an evenings work, but if you have some manual dexterity, with some care, you can have your speakers back up and running like new.

I have been successful doing this on several JBL, Altec, and Bose speakers. I suggest that you stick with U.S. made surrounds. Follow the instructions, and make sure to use a low frequency signal from either a signal/function generator/oscillator or the CD that comes with some of the kits to make sure that everything is centered, and that the voice coil doesn’t rub on the pole piece, etc. Good Luck!

I was recently given an old pair of Cerwin Vega speakers from a co-worker in much the same condition, it sounds, as yours are. The woofers in both cabinets were electrically good as was the physical condition of the voice coils and cones. Both woofers though, exhibited the common problem of foam rot. Not only does the foam separate but seems to dissolve leaving the cone and voice coil to rattle around without direction. 

I performed a web search on the subject and found a solution that seemed very reasonable. Simply Speakers (simplyspeakers.com) has re-foam kits for numerous speaker manufacturers including Cerwin Vega.  I’m not touting Simply Speakers as the only provider, but they were excellent to work with, sent me everything I needed to do the job and provided a youtube video on how to do it. Search youtube for “Simply Speakers” to see the video. I just finished mine last week with excellent results.

If the cone and the driver are in good condition, I would do it. It isn’t very expensive or difficult. I replaced the flexible foam on a set of old Speakerlab speakers about 5 years ago. There are a number of places on the internet and on ebay that sell the parts and they have instructions. I’d check out some of the sites and look at the instructions to see if you feel comfortable with the procedure.

For the subwoofer, I had to repair the foam on both the 10” speaker and a passive radiator. I did decide to make a ring out of 1/4” plywood to make the assembly of one of the larger rings easier but you probably wouldn’t need to do that.

That’s a very old circuit and it has at least one glaring flaw: The value of R3 is way too high. I suspect it should be 3.3K instead. Also, D3 and D4 seem unnecessary because there can never be more than about 2mA of meter current. R1 protects D1 and D2, but microamp circuit levels wouldn’t threaten them. The input bias current of a 741 op-amp through a 10K resistor produces up to 5mV offset that’s not temperature-compensated. It’s nullable, of course, but the null can drift.

Figure 1

While it’s pretty easy to make a DC voltage inverter with circuits like the one in figure 1, I believe there’s a better solution: Single-supply op-amps are available with offset null pins, completely eliminating the negative power supply! See figure 2 for my (untested) circuit suggestion. TLE2021 chips are available in plastic DIP packaging from the usual suspects, such as Digi-Key and Mouser.
 

Figure 2

The current sampling resistor should be selected for full-scale reading with 10mV drop, which is 10X lower measurement burden relative to the original circuit.

If you add the input diode protection resistor back in, then you incur only 0.7mV max uncompensated offset because of the TLE2021’s enhanced performance. And, if you do this, then there’s really no reason to keep those diodes -- you can rely on the op-amp’s input pin protection circuit.