News:

Masm32 SDK description, downloads and other helpful links
Message to All Guests

Main Menu

Capacitor and it's testing ?

Started by Magnum, September 14, 2013, 01:26:17 PM

Previous topic - Next topic

Paulo

Quote from: dedndave on September 15, 2013, 10:26:23 AM

there is a slight time delay between current traveling and visible light   :P
also - light bulbs need some minimum current to be visible

Interesting point.
Light only travels at the....well...errr speed of light in a vacuum (the often quoted 3X10^8 m/s).
In other mediums, it's considerably lower.

Likewise for electricity, hence the concept of velocity factor in high frequency electronics where the dielectric constant
of the substrate ( the printed circuit board) has a huge influence on the speed and hence wavelength.
At very and ultra high frequency, circuit elements such as capacitors and inductors are made (etched) from pcb traces
and the lengths and/or widths must be compensated for because of the dielectric constant.

This also applies to high speed digital circuits.
Next time you have a PCI card in your hands, look very carefully at some of the traces, they "snake" around instead of just
going directly from point A to point B.
This is so that the correct delay is achieved as required.
In PCI cards, this is normally done to the clock signal where the data lines go straight to a latch and the clock lines snakes around
for a bit to create a delay then feeds the latch enable of the latch.
This is done so that the data is present (and steady) before it is latched.

Sometimes it's also done so as to compensate for the length of travel, i.e. the first connection is physically closer than say the last
thus extra trace length is added so that the arrival of data is at the same time.
It all depends on layout.


dedndave

that's how you make an inductor on a PCB   :P
i doubt the propogation delay alone would be signifigant
at least - at the frequencies that are present on that board

Paulo

Quote from: dedndave on September 15, 2013, 11:46:33 AM
that's how you make an inductor on a PCB   :P
i doubt the propogation delay, alone would have meaning
at least - at the frequencies that are present on that board

Well yes and no.
To create inductance, it must have a given length and width for a certain dielectric constant.
So when one creates a thin trace like that, it will be mostly inductive (except for the small amount of capacitance due to the ground plane)
so creating a trace of the same length but wider will decrease the inductance but increase the capacitance.
Which ever is the dominant reactance, it's the length we are interested in here as both capacitance and inductance create delay but one also has to watch the phase
so simply creating a long line is not going to work, it must have the right combination of XL and XC.

The propagation delay is very much a factor considering that the dielectric constant of FR4 PCB is 0.5 and at PCI 66MHz with bit times in the region of 15nS, it's enough to cause very weird errors.
Remember that PCI is a parallel bus so if one of the bits is delayed by evena few nS relative to the others, it could mean the bit is interpreted as a high instead of a low or vice versa.
Of course the situation gets even more critical with PCI @100MHz and above.
This is also the reason why one cannot use "normal" logic gates with PCI (and other high speed digital circuits) as the propagation delay thru these gates and logic chips is simply too great
not to mention their input capacitance which would cause unacceptable bit error rates.
Have a look at the PCI specs and you will see why many PCI cards make use of ASIC technology and highly optimized pcb layouts.

Paulo

Keep in mind that propagation delay is directly proportional to velocity factor for a given substrate which is unfortunately not constant with frequency.
That is why good PCB manufactures always publish VF versus frequency charts of their products.
BTW, you will also see that these traces always have mitered corners, this is done so that there isn't an abrupt change in the magnetic field which
cause impedance disturbances (or bumps).

The fun really starts when one is designing at 1GHz and above.  :biggrin:

Magnum

Meter with ohms is high on my list.

Considering what to charge. (I have a steady income, so I don't need to get rich or whatever)

$6 for part + 1 hr. of actual labor (other time spent was educational)

Andy

Since I moved, I no longer have access to a computer repair shop waste container.  :icon13:

But I have been accumulating furniture, bookcases, etc.

My 2 daughters and hubbies, 2 of my 4 step kids and 3 grand kids live close by.




Take care,
                   Andy

Ubuntu-mate-18.04-desktop-amd64

http://www.goodnewsnetwork.org

dedndave

i think you can get an analog meter for about $10
it doesn't have to be digital
sometimes, an analog meter is actually preferable
i still have an old Simpson 260 that i like to use for monitoring supply current, etc
i also have several Weston meters, but those are more as collectors items   :P
they do work, though

Magnum

I received the new cap. It has a white and red wire.

I remember the can style caps as having a + and - terminal but no markings indicating polarity.

I don't want to miswire that puppy.

Andy

Take care,
                   Andy

Ubuntu-mate-18.04-desktop-amd64

http://www.goodnewsnetwork.org

dedndave

i am thinking....
it's a non-polarized capacitor
they provide a red and a white wire so the directions will jive - lol
these things are put together by trained monkeys
if you gave them 2 black wires, it would never get assembled   :biggrin:

Magnum

When I hooked up the cap and tried the fan, nothing happened.

I looked at the armature and it's a silver color unlike the copper color that I have seen in others.

I replaced the armature and rotated it, and it sort of wobbled instead of being a fixed distance away from the windings.

The owner of the fan ran it about 12 hrs. a day for 7 days a week, so I think he got his money's worth.  :t

The bearings were frozen before using PB Penetrating Catalyst to loosen it up.

If it had oil lube holes for the bearings, the fan would last for a long time :-)

I'll just keep the starting capacitor for something else.

Adios,
         Andy
Take care,
                   Andy

Ubuntu-mate-18.04-desktop-amd64

http://www.goodnewsnetwork.org

dedndave

you may be able to see that one of the windings is "open" with the continuity tester
give the whole thing back to him and tell him to save it for parts - lol
that way, you don't have to store someone else's junk   :biggrin:

Magnum

I will tell him I could not fix it and ask if he wants it back.

He probably won't want it back and I can salvage the cord to use for a short extension cord.

Andy

I may keep the blade too.

I repaired a fan for my daughter that had a small piece of one blade missing,
so I used Gorilla glue and glued a dime to the blade to balance it.



Take care,
                   Andy

Ubuntu-mate-18.04-desktop-amd64

http://www.goodnewsnetwork.org