How to make a 100X oscilloscope probe

This shows howto to make you own homemade 100X oscilloscope probe.

''Warning: if you plan on using this probe for high voltages, you will need to use more insulation than in shown in this howto. Standard heat shrink tube is rated for 600V continuous, and spikes of 2500V.''

Specifications

 * Tested at voltage up to 1200V RMS
 * Compatible with standard probes
 * Tip same shape as standard probes
 * Head same shape as standard probes
 * Ground clip same as standard probes
 * Compatible with spring hook attachments
 * Does NOT have ground ring, near the tip, like most probes
 * Has a decent bandwidth (I don't have a value)

Parts

 * Components
 * R1 - 10.0MΩ
 * R2 - 113kΩ
 * Oscilloscope input impedance assumed to be 1Meg ohm. (Should confirm on your scope)
 * C1 - adjustable 2-5pF, high voltage
 * Can use fixed value, see below
 * Capacitor Voltage > 0.9*Vin_max
 * BNC cable to splice
 * BNC connector to use center pin
 * Flexible Coaxial wire
 * Shielded box (optional)


 * Other
 * Note that 113k in parallel with 1Mohm scope input resistance = 101527 ohms, gives approximate 100:1 division
 * Heat shrink tube, various sizes
 * plastic pen
 * plastic block to make probe's tip
 * rubber stopper



Instructions
The steps described below are separated in to a few arbitrary categories. Some of them must be done in parallel so read all the steps before starting.

BNC connector to R2 to coax wire
I used a shielded box because I thought I might need to add more components. Its probable not necessary, The cable could probable be spliced and R2 could be soldered inside the cable.
 * Get precision resistors R2 = 100kΩ. Or use two or three resistors to get as precise as you can.Use 1/8W resistors if you can, they are much smaller.
 * Cut the BNC cable as close as possible while still leaving room to work.
 * Strip cable
 * Pull back shielding braid, to attach R2
 * Put R2 in a shielded box, connecting signal and ground
 * If you don't use a shielded box, surround R2 in shielded braid
 * Attach coax wire to the BNC cable inside the shielded box or if you don't use a box directly to the BNC cable.
 * Cut the coax with to about 4 ft, or the length you desire. Length will effect the value of C1.

Creating Input impedance unit (R1//C1)
I am calling the components R1 and C1 soldered in parallel as the Input impedance unit. You want to make this as thin as possible so it can fit in the pen with braided shield surrounding it. These steps must be done in parallel with setting/selecting C1, as described below.
 * Get precision resistors R1 = 9.9MΩ. Or use two or three resistors to get as precise as you can.Use 1/8W resistors if you can, they are much smaller.
 * solder them close together in a straight line.
 * Solder C1 in parallel with R1, making sure there are no shorts, use heat shrink tube in needed, but make it all as thin as possible.
 * See Setting C1 Compensation for selecting or setting C1
 * Cover the unit with at least two layers of heat shrink tubing. This with reduce the capacitive effects when you surround the unit with the braided shield
 * Cut leads to be as short as possible

Setting C1 Compensation
This step must be done in parallel with Creating Input impedance unit (R1//C1) and Attaching Coax to Input impedance unit, because the value of C1 is very sensitive to the shielding that surrounds it, and other factors. An uncompensated probe has an over-damped response to a square wave due to capacitance in the coax wire. See image below.
 * Compensation can be added by adding a capacitor, C1, in parallel with R1
 * This will reduce or remove the over-damping.
 * If too large a capacitor is used signals will become under-damped, causing spikes. See image below.
 * Ideal compensation would make the input signal identical to the output signal. With a square wave input, ideally, the corners are as sharp as possible.
 * The level of damping is highly effected by the shielding surrounding the input impedance, so when selecting the value of C1, many tests may be required to get it right.
 * Using a variable capacitor will make this easier, but they are much larger, so it may require some work to get it to fit in the probes casing.
 * The input impedance, R1//C1, needs to be covered with at least two layers of heat shrink tubing. This with reduce the capacitive effects when you surround the unit with the braided shield, See Creating Input impedance unit (R1//C1) for details

Attaching Coax to Input impedance unit
These steps describe attaching the Input impedance unit and covering it with the braided shield from the Coax.
 * Strip about 3.5 inches of insulation off the end of the coax, leaving the braided shield in place.
 * Pull back the braided shield as far as you can, exposing the insulated center wire. Try not to mess up the braiding when pulling it back
 * Cut insulated center wire to be 3/4 inches long. This make the braided shield 2.75 inches longer than the center wire.
 * Keeping the braided shield pulled back, strip the center wire about an 1/8 of an inch
 * Solder the Input impedance unit to the center wire and insulate the solder joint with heat shrink tubing.
 * Pull the braided shield over the Input impedance unit, so it covers it the complete length of the unit.
 * Unbraid the portion of braided shield that stretches beyond the Input impedance unit.
 * Twist the unbraided portion into a wire. This will be used as the probes gnd.

Creating the probes head
Using these steps the probe should be compatible with other oscilloscope probes. The only feature this doesn't use is a ground ring close to the tip.
 * Take the plastic block and cut it into a rectangle, about 1.6in x 0.35in x 0.35in (a little bigger than the width of a normal probe)
 * Use a knife to round off the sides to make it into a rod
 * Use a drill as a mini-lathe
 * Put the rod into a drill or drill press
 * turn it on
 * Use a chisel to shave the piece of plastic into the shape described below.
 * Make sure it slides into the pen's tube
 * Using a 1/8 in drill bit, drill down the center of the rod. Like shown in the diagram.
 * Using a 5/64 in drill bit, drill down the center of the rod, in the front. Only drill 0.3 inches. enough to insert the pin. Like shown in the diagram. Use tap on the drill bit to mark 0.3 in, so you know when to stop drilling.
 * Making the pin to the tip of the probe
 * Break open a BNC plug to get the center pin, keep a little bit of the wire attached so its easier to solder to.
 * Use the drill as a lathe on the pin. put it in the drill and with a file, shave it thinner and pointier. Shave it so that it will fit in a bread board easily, but not so easy that it bends.

Final construction
See images for visual representations of instructions below Follow the images bellow to finish the construction
 * Cut a thin wire the a little longer than the length of the plastic piece.
 * Solder wire to metal tip
 * Slip wire in the front of the plastic head and push the tip in.
 * Drill hole in pen 3/4 inches from the front of the pen
 * Slide the input impedance(connected to the coax) into the back of the pen
 * Pull ground wire through pens hole.
 * Slide the input impedance(connected to the coax) 3/4 inches further forward, and solder it to the wire in the plastic piece.
 * Push the plastic head in pen tube, while sliding back the input impedance(connected to the coax)
 * If the plastic piece is not tight in the pen tube, add some heat shrink tube.
 * Make sure the ground wire stays in the hole
 * Put heat shrink tube over the back of pen and an inch of the coax (like in image)
 * Pull rubber stopper over pen tube, past the ground wire
 * Take alligator clip attached to a wire.
 * Slide some heat shrink tube on wire.
 * Solder alligator clip wire to ground wire
 * Pull heat shrink tube over solder joint and ground wire, and shrink it
 * Use heat shrink tube where ever else needed
 * And you are all done


 * Notes
 * Make sure all solder joints are well done, because they will not be easy to access.
 * Make components leads as short as possible, and connections as tight as possible. This will reduce noise, and make it easier to encapsulate.
 * Make sure everything is insulated so there are no shorts, now or ever.
 * Make sure nothing can move around, over time small movements can break connections.
 * its probable a good idea to not seal up the probe permanently in case a modification needs to be done. Heat shrink tube can easily be cut, epoxy glue cannot.

Ideas for a (possible) better design
This isn't tested but should be easier than trying many different various different fixed capacitors, hoping to get one right.
 * Use a fixed value C1, that causes slight overshoot.
 * Add an extra adjustable capacitor 0-2pF in parallel with R2
 * Adjust the extra capacitor to ideal