Helpful Helping Hands

Tired of those  not-so-helpful helping hands letting you down?  Well not any more!

They look mean, and they wrestle those “old hands” to the mat!    The old hands are shown here crying for mercy (in mid-air) just before being thrown down:

and hammered mercilessly with 100,000 volts (can yours do that?):

But seriously, these hands are the best I’ve ever had, and they’re easy to make.   The joints are flexible, yet stay firmly in-place; perfect for soldering and holding probes.   The end-effectors (shown here with clip-leads) can be easily swapped out to support different tooling requirements.   Here’s a shot of the hands doing real work:

The jointed segments are “Loc-Line” modular hose.   Connected to the base (Panavise PV-300) is 1/2″ diameter Loc-Line (from ModularHose.com) rising to a bifurcated fitting splitting off into two 1/4″ segments.   Here’s how they come (the 1/2″ kit has two segments, only one is shown):

To make the helping hands, you’ll need a 1/4″ hose kit, a 1/2″ sample kit, a pack of 1/2″ to 1/4″ reducers, 1/4″ and 1/2″ assembly pliers (supposedly they aren’t necessary, but I wasn’t able to make all the connections without them), and additional 1/4″ round nozzles (a parts list is at the end of the blog entry).

To make the tool-holders on the ends of the “arms”, I ran over to Ace Hardware, and picked up some 10-32 bolts, nylon spacers, and 8/32 set-screws:

I threaded the nylon spacers, and fed the screw up through the bottom of the 1/4″ nozzle, and tightened it into the nylon.   The other end of the nylon spacer was drilled out to accept a clip, then a set-screw hole was drilled & tapped.  After assembly, the nozzle was snapped into place on the end of each 1/4″ arm.

A 1/2″ NPT connector (provided in the 1/2″ kit) was snapped onto the bottom of a 1/2″ segment,  then sanded down to fit into the Panavise base.

Parts list (other colours are availble):

1 40413-Blk   Loc-line 1/4″ hose kit – $7.85
1 50813-Blk   Loc-Line 1/2″ sample kit – $8.85
1 51822-Blk   Loc-Line 1/2″ to 1/4″ reducer (pack of 2 “splitters”) – $6.65
1 41404-Blk   Loc-Line 1/4″ round nozzle (pack of 4) – $3.90
1 78003          Loc-Line 1/4″ and 1/2″ plier set – $17.78
Shipping:        Around $10.

Panavise Base:  $$priceless$$

Total: $37 in parts not including the plier set & base.  You’ll have enough “spare parts” to make a second set of helping hands just by ordering another 1/4″ hose kit (40413-Blk) — I did.

UPDATE – If two hands are good, four are better!!

“May the hands be with you!”

Posted in Hardware | Leave a comment

ThunderPower RC TP820CD 800 Watt Dual-Port Multi-Chemistry Battery Charger/Discharger/Cycler/Balancer

Wow, that title is a mouth-full, and this monster delivers!   With the new generations of Lithium-Polymer (LiPo) batteries capable of being charged at up to 12C (12X battery pack capacity – YMMV), the “last generation” chargers were holding my batteries hostage when I  could have been flying — not any more:


(Shown charging a 3-cell Turnigy and a 6-cell ThunderPower pack at the same time.  The “spider-monkey” charge adapters are from DraganFly).

The ThunderPower RC TP820CD comes with two balance paddles that accept ThunderPower and JST-XH balancer connections.  It seems a bit strange that the TP820CD doesn’t appear on the ThunderPowerRC web site, but they can be found all over the web.  The users manual can be downloaded from here.

After upgrading the firmware via the mini-usb connection (running Windows XP, in VirtualBox, under Linux),  I returned the unit to ThunderPowerRC when I noticed my battery packs weren’t being charged to full capacity (only about 80% or so).   Mark at ThunderPowerRC called me back and explained that if the charge balance feature was disabled (I didn’t have balancer adapters for my Hyperion packs), as a safety feature, the batteries won’t be fully charged (doh).   He was very nice, and let me know that he had carefully verfied the operation of the charger, and then promptly returned it.   Apparently this is SOP for ThunderPower, but isn’t for Hyperion Chargers (at least for the 606i).  Mark also referred me to Dan at RC Lipos where I was able to get JST-XH to Hyperion balancer adapters.

To get the full 800+ watts out of the TP820CD, you’ll need a power supply that can provide 24 to 28 Volts at 40 Amps (maximum input current).   13.8V at 40 Amps will allow the charger to deliver 550 watts.   If your input source is less than “full power”, you can set how much of the input power is directed to which port (1 or 2).

The main feature I needed in a charger (besides more power) was one with a “storage mode” for LiPo packs.   The TP820CD will either charge or discharge LiPo packs to roughly 50% capacity (depending on its charge-state), to maximize cell life when the batteries are not going to be used right away.

There are several safety features on the TP820CD to ensure batteries are not over-charged (a very dangerous condition for LiPos), including charge capacity, maximum charge-time, internal cell-sense, and maximum cell-voltage.   It is, however, possible to force the charger to do something bad for your batteries (and perhaps your domicile as well).

Of course, the TP820CD will also charge Lithium-Ion, Lithium-Iron, Lead-Acid, NiCd, and NiMH cells.   The cycling (charge/discharge) mode on the charger can  help rejuvenate some of these chemistries.

I only wish the charger could dissipate more that 50-watts (per channel) on discharge;  several times while discharging cells, the TP820CD would overheat and have to temporarily halt the discharge cycle until it cooled off (a minute-or-so).

Overall, I’m very happy with the unit, and it’s a welcome addition to my “family” of battery chargers/conditioners.

Posted in Electronics | Leave a comment

Quantum Random Number Generator

Tired of unpredictable pseudo-random number generator (PRNG) seed artifacts?  Need a high-speed/high-volume source of truly random data for Monte-Carlo, gaming, or security?   The Quantis Quantum Random Number Generator from IDQ could be what you need.   Certified by the Swiss Federal Office of Metrology,  the Quantis devices can crank out up to 16 Mbps (in-machine version), or 4Mbps for the USB version:

This white-paper discusses how they work, and why they should be used.

These units are supported under Windows,  Linux, FreeBSD, and Solaris.   The USB version needs libusb1 under Linux, so a recent distribution is required.   The installation instructions are well-written,  and good sample programs are included in the C, C#, C++, Java, and VB.NET languages (in addition, a run-time application “EasyQuantis” to test the installation is included).   The maximum number of bytes that can be read at-a-time is 16MBytes;  this took about 33 seconds via the USB device (4Mbits/S).    As might be expected, trying to compress the (binary) random data yields either no, or negative compression (output file is larger).

Taking 16MBytes of data and putting the values into 1 of 256 “bins” (byte values), the mean value was 65536, with a median of 65513.5,  a standard deviation of 292, a minimum of 64798, and a maximum of 66278 (as computed by SciPy).

There are several on-line sites for random numbers;  Random.org (uses atmospheric noise), Hotbits (radioactive decay), or the Quantum Random Bit Generator Service (photonic emission from semiconductors).  I’ve noticed that one or more of these sites may be unusable at any time (device/network issues).  Security/privacy of the data may be an issue, and the quantity of available random data may be limited from these sites (Hotbits  generates about 100 Bytes/Second).    One of my favorite (fringe) techniques is LavaRnd, conceived and first implemented at Silicon Graphics, just before the turn-of-the-century.

The delivery time for the Quantis devices is very fast (2 days), but in the event of device failure (they are self-monitoring), having a “reservoir” of random data that can be tapped which is large enough to cover any down-time is a good idea.   I built an RNG daemon that switches between the Quantis and a reservoir transparently to applications (and, of course, logs & emails notifications).

At a price of under $2000 USD, these units solve a lot of problems.

Posted in Electronics | Leave a comment

Fedora 14 Quad Monitor Configuration (and issues)

Need (or want) a single desktop using dual graphic cards and 4 displays?

Here’s what I found upgrading to Fedora 14  from Fedora 11:

First of all,  I had to perform a fresh install since there is no direct upgrade path from Fedora 11 to Fedora14.   WATCH OUT!   Half my disk was partitioned (by default) for root (/), and the other half for /home.   Since my disk is relatively small, I ran out of space setting up users in /home (boo, hiss).  After a few hours trying to resize the file systems and partitions, I gave up, and did a second full re-install, but not before deleting the /home partition.

Next, the system wouldn’t come all the way up (even to run-level 3); it would just hang.   Using the rescue DVD to change the initial run-level in /etc/inittab to 1, I finally saw that the kernel was krashing when enabling IPV6.    So I added NETWORKING_IPV6=no to the /etc/sysconfig/network file, and created a file named blacklist-ipv6.conf to the directory /etc/modprobe.d, containing the lines:

install ipv6 /bin/true
blacklist ipv6

I was finally able to boot completely.   The first install involved several hang-ups as well, but it started working (for reasons unknown).

Next, the swap partition wouldn’t mount at boot, or even later with swapon -av due to memory allocation problems.   After a couple of hours of mucking around, I added vmalloc=256MB to the end of the kernel line in /etc/grub.conf, and suddenly swap was mountable.    See the “Kernel virtual address space exhaustion on the X86 platform” section of Chapter 9 “Known Issues” in the Nvidia driver README.

As a side-note, using the Fedora Nouveau drivers are not an option, since they don’t support hardware acceleration and don’t support 4 displays as a single, unified desktop (I read rumours of folks getting Nouveau to support two graphics cards, but only as 4 separate desktops).   So,  Obi-Wan Kenobi, the Nvidia drivers are our only hope.

After grabbing the latest Nvidia driver (x86-270.29), installing it, configuring  /etc/xorg.conf for quad displays (configuration below), things seemed to be running well.   But then during the “shakedown”, flash seg-faulted.   Un-installing nspluginwrapper made no difference.   Even trying the open-source flash player “Gnash” didn’t work (“Too many attempts to read from the player!“).    After a few hours of trying things and many reboots, it came down to the use of the Xlib Xinerama extension in the configuration (more issues later).

The only way to make quad monitor support with two graphics card work (nicely) is with a combination of TwinView (left/right) and Xinerama (top/bottom).  However, when using Xinerama and the current (Nividia) drivers, flash support is broken in all the browsers I tried (Firefox, Seamonkey, Opera).

At this point, trying to get new drivers from Nivida while Xinerama is enabled is problematic since flash doesn’t work, and the Nvidia site is all flash-based (sigh). So, plug: ftp://download.nvidia.com/XFree86/Linux-x86/ (or x86_64)
into your browser to get driver access.

Trying the drivers for FC13, the last release, an undefined Symbol miEmptyData in nvidia_drv.so (found in Xorg.*.log) keeps the drivers from loading, so they are “right-out”

Trying the latest legacy drivers NVIDIA-Linux-x86-96.43.19-pkg1.run, and NVIDIA-Linux-x86-173.14.28-pkg1.run, both allow flash to work, with 173.14.28 performing the best.   The downside being, there’s no CUDA support for the GTX 260 in order to get flash to run (sigh).

Currently, using Xinerama has a couple of other downsides; there is no Xlib RANDR (display manipulation) extension support, so some programs like the scanner driver “xsane” crash, since it (apparently) blindly assumes the extension is available.   I have found  the program gscan2pd works very well for me (much better for monochrome PDF generation).  In addition, the Xlib Composite extension is not available, so the Compiz desktop effects can’t be used 🙁  In addition, I have to log-in twice to get the second graphics card activated.  Dropping back to dual-monitor (single graphics card) support (no Xinerama, just TwinView), everything works fine on the newest drivers.

With all that said, you’ll have to decide if the trade-offs are worth it for you.     I’ve heard some scuttlebutt (since 2008) that the Xinerama extension may be deprecated,  so the future of quad-displays may be in-doubt with Nvidia.    But, in the mean time, they are very cool!

Here’s the /etc/X11/xorg.conf for 4-head display:

#
# xorg.conf-3840x2400_quadhead-2x2-xinerama-twinview+twinview
# (C) Bill Bishop, all rights reserved, but free-use granted.
#
# Update the Monitor and Device sections to reflect your configuration.
# You probably want to delete the GTX 260 device section too (unless you
# have one)
#

Section "ServerLayout"
        Identifier   "Layout0"
        InputDevice  "Keyboard0" "CoreKeyboard"
        InputDevice  "Mouse0"    "CorePointer"
        Option       "Clone"     "off"
        Option       "Xinerama"  "on"
        Screen       0           "Screen0"
        Screen       1           "Screen1" Below "Screen0"
EndSection

Section "Files"
#	RgbPath      "/usr/X11R6/lib/X11/rgb"
	FontPath     "unix/:7100"
EndSection

Section "Module"
	Load         "dbe"
	Load         "extmod"
	Load         "glx"
	Load         "freetype"
	Load         "type1"
EndSection

Section "InputDevice"
	Identifier   "Keyboard0"
	Driver       "kbd"
	Option	     "XkbModel" "pc105"
	Option	     "XkbLayout" "us"
#	Option	     "XkbVariant" "nodeadkeys"
EndSection

Section "InputDevice"
	Identifier   "Mouse0"
	Driver       "mouse"
	Option	     "Protocol" "IMPS/2"
	Option	     "Device" "/dev/input/mice"
	Option	     "ZAxisMapping" "4 5"
	Option	     "Emulate3Buttons" "no"
EndSection

Section "Monitor"
	Identifier   "Monitor0"
	VendorName   "Dell"
	ModelName    "2407WFP"
	HorizSync    30.0 - 81.0
	VertRefresh  56.0 - 76.0
	Option       "IgnoreEDID" "true"
EndSection

Section "Monitor"
	Identifier   "Monitor2"
	VendorName   "Dell"
	ModelName    "2407WFP"
	HorizSync    30.0 - 81.0
	VertRefresh  56.0 - 76.0
	Option       "IgnoreEDID" "true"
EndSection

Section "Device"
	Identifier   "Videocard0-Output0+1"
	Driver       "nvidia"
	VendorName   "Nvidia"
	BoardName    "NVIDIA GeForce 7600GT"
	BusID        "1:0:0"
        Option       "CursorShadow" "1"
	Option       "NoLogo" "1"
	Option       "Coolbits" "1"
	#Option       "IgnoreDisplayDevices" "crt,tv"
	Option       "ConnectedMonitor" "DFP,DFP"
	Option       "TwinView" "true"
	Option	     "TwinViewOrientation" "RightOf"
	Option       "MetaModes" "1920x1200,1920x1200"
	Option       "SecondMonitorHorizSync" "30-81"
	Option       "SecondMonitorVertRefresh" "56-76"
EndSection

Section "Device"
	Identifier   "Videocard1-Output0+1"
	Driver       "nvidia"
	VendorName   "Nvidia"
	BoardName    "NVIDIA GeForce 7600 GT"
	BusID        "10:00:0"
        Option       "CursorShadow" "1"
	Option       "NoLogo" "1"
	Option       "Coolbits" "1"
        #Option       "IgnoreDisplayDevices" "crt,tv"
	Option       "ConnectedMonitor" "DFP,DFP"
	Option       "TwinView" "true"
	Option	     "TwinViewOrientation" "RightOf"
	Option       "MetaModes" "1920x1200,1920x1200"
	Option       "SecondMonitorHorizSync" "30-81"
	Option       "SecondMonitorVertRefresh" "56-76"
EndSection

Section "Device"
	Identifier   "Videocard2-CUDA"
	Driver       "nvidia"
	VendorName   "Nvidia XFX"
	BoardName    "NVIDIA GTX 260"
	BusID        "5:00:0"
EndSection

Section "Screen"
	Identifier   "Screen0"
	Device       "Videocard0-Output0+1"
	Monitor      "Monitor0"
	DefaultDepth 24
	SubSection "Display"
		Viewport  0 0
		Depth     24
		Modes     "1920x1200"
	EndSubSection
EndSection

Section "Screen"
	Identifier   "Screen1"
	Device       "Videocard1-Output0+1"
	Monitor      "Monitor2"
	DefaultDepth 24
	SubSection "Display"
		Viewport  0 0
		Depth     24
		Modes     "1920x1200"
	EndSubSection
EndSection
Posted in Computers | 3 Comments

Building the EVR miniBrute Tesla Coil – Part 1, The Advanced Modulator

On this blog I’ve shown how to build a classic Spark-Gap Tesla Coil (SGTC) and small Solid State Tesla Coils (SSTC’s).  It’s now time to step-up into a Dual-Resonant Solid State Tesla Coil (DRSSTC).   Advantages of the DRSSTC include reduced line-power requirements (more efficient), low-voltage operation (hundreds of volts instead of tens of thousands),  and digital control due to solid state electronic replacement of the spark-gap.    This results in quieter operation and much more precise power control over the SGTC.

Eastern Voltage Research (EVR) supplies a partial kit for their miniBrute DRSSTC.  In addition to the miniBrute kit, a modulator (controller/interrupter) must be constructed, the construction/reference manual “DRSSTC: Building the Modern Day Tesla Coil – miniBrute Reference Manual” must be purchased, and additional material/components acquired to complete the coil.    The additional parts & pieces came to about $500 (toroid, plastic, screws/knobs/inserts, shipping & handling), not including a Variac (variable autotransformer) and spare IGBT‘s.   More detail will be provided in later posts.

This post covers construction of the EVR Advanced Modulator (complete kit with knobs).   Pages 75-81 of the construction manual contains pictures, waveforms, schematics, and parts list for the modulator.   A preview of the manual can be found here; specifications for the advanced modulator can be found here.   Be aware that the construction difficulty of the miniBrute is level 2, advanced.

When ordering the Advanced Modulator kit, custom text for the front panel can be requested.

At this point, it’s assumed that you have the book and modulator kit in-hand, so lets dive-in.   Here’s what came in the complete kit (with knobs) – excluding the clamp & breadboard:

There are no instructions on how-to assemble the modulator, so proceed by installing fixed resistors, diodes, and small capacitors as noted on the schematic into their locations marked on the circuit-board by the white screen printing.  IC sockets are optional, but since the modulator will be used near high-voltage, I recommend them.    My kit came with single-pole, double throw switches, but double-pole, double-throw are required (for at least two of them).   Dan quickly sent me the proper ones.   After assembling the board, your kit should look something like this:

After aligning the faceplate to the top of the plastic box and using a silver “sharpie” to mark the outlines of the switches & potentiometer holes, a Dremmel tool and 1/4-inch drill bit made quick work of the openings (LED holes not shown):

After attaching the front panel, the components were mounted thusly:

The circuit board was secured using four 1/4-inch nylon spacers and #8 screws & nuts.  Shoe Goo was used to secure the 9V battery tray (it’s not just for shoes), and four little rubber feet (LRF’s  not included in the kit) where attached to the bottom of the box.  The Dremmel was used again to make the indents on the inside of the box to accommodate the potentiometer tabs. After installing the proper switches and wiring it up, here’s how it looks (YMMV):

I started by wiring up the power switch, then the Aux (low frequency) potentiometers, and output to J2.   Dan sent me a revised schematic that mentions that the outputs (J1 & J2) should have 750 ohm resistors in series with them when being used with the miniBrute since the logic levels from the modulator are 9-volt, and the miniBrute control board expects 5-volt logic (it has a 1k-ohm pull-down resistor).

Most of the potentiometers will need to be wired so that their resistance increases with clockwise rotation.   The exception to this is R7 (PRF ADJ).   After wiring up each potentiometer, I powered-up the circuit, and checked the output on pin three of the associated 555 timer for proper output.   This way I debugged problems as the circuit was being completed.    This works until you get to U4 and U5; they have a codependent relationship, so R15 and R18  both need to be wired up before that portion of the circuit can be tested.  After initial check-out, the box can be buttoned-up, and final testing performed:

Part 2 will cover the assembly of the miniBrute electronics.   Get your kit ordered, and stay tuned!

4hv.org has a forum covering the miniBrute here.   If you don’t already subscribe to the Tesla Coil mailing list, you might consider subscribing.

Posted in Tesla Coils | Leave a comment

Probing Your Megavolts

So, you’ve invented this amazing process for converting water into wine, but investors want to know exactly how much power it takes before they cut checks.   After all, vineyards do it for free, with just a bit of sunlight.

Your process involves the use of extremely high voltage with direct electron disassociation (big sparks) in special electro-chemical reactors.  Any electronic equipment within a 6-foot radius of the reactors is fried like a mosquito in a bug-zapper.

Being a start-up, there’s no cash to set-up at Powertech Labs, or buy a custom divider. But to keep the lights on and suppress a mutiny, you desperately need a check this week.   What do you do?  What do you do?

Fortunately, a quick trip to local shops will supply the parts needed to make a 30,000 to 50,000:1 extremely high voltage (megavolts) resistive divider that draws less than a milliamp at 1 ONE MILLION VOLTS.   These parts and a current monitor will allow you to discover your true power (bwaaa haa haa)!   ahem.., sorry; move along, nothing to see here….

Discussion: As you might have figured, that’s still a thousand watts of power (at a megavolt), iff the voltage is continuous.   Measuring hundreds of 50 micro-second pulses/second yields an average power dissipation of a dozen watts, and much lower if the voltage is less and/or decays exponentially;  YMMV.   Since we’re talking about a pulsed-power application, a digital oscilloscope is required to capture and record the measurements.

Enter stage rightThe water resistor.  The big idea is to use small vinyl tubing with very pure (deionized/distilled) water to create a high-value resistor as part of a voltage divider.   The voltage divider is used to scale the million-or-so input volts down to something off-the-shelf test equipment can measure (dozens of volts).  As with most things, the rub is in the details.

Initial experiments with 1/4″ ID (inside diameter) vinyl tubing with 1/4″ bolts as end-caps & electrodes yielded low resistance and a tendency to leak:

Plain tap (drinking) water is easily 20 times more conductive than distilled water due to impurities.   The 20 foot resistor above was measured at 23.1 megohms; or about 1 megohm/foot (boo, hiss).    10 feet of the same tubing (1/4-inch ID) filled with distilled water measured ~215 megohms (20 megohms/foot).    At this point, you may be wondering, “How the heck do you measure something with that much resistance?”   Good question; using a home-brew 1kV MOT power supply (limited to 1kV due to the diodes laying around):

High voltage is applied to the water resistor, and current flow is measured; calculating the resistance using ohms-law (R = E/I):

Switching to smaller tubing, the 3-feet of 3/8-inch ID vinyl tubing above, filled with distilled water, with 1/4-inch brass plugs (which worked much better than bolts),  has a computed resistance of 494V/.00000315A = 157 megohms; the HP-34401A DMM reads down to 10 nano-amps.   In case you don’t have a nano-ammeter available, the divider calibration only requires a voltmeter (and a high-voltage source).

Based on the above experiment, a gigohm water resistor should be about 20 feet of 3/8-inch ID vinyl tubing filled with, you guessed it:

Before filling the tube, it was flushed several times with distilled water.   Using a large syringe & needle makes filling the tube much easier, but it still takes time and patience to ease (knock) the bubbles out.   Also, tapering the ends of the 1/4-inch brass plug/electrodes (cut to 3-inch lengths from a longer rod) makes them much easier to insert, and there has been no problem with leakage.

Once the tubing is filled and sealed, the voltage divider is completed using a standard resistor (27K-ohm) as the low-voltage leg of the divider between the water resistor and ground.   The output voltage is measured across the small resistor.   Changing the value of this resistor is the easiest way to change the divider value.    It was found that making this resistance too large caused apparent non-linear behaviour (divisor changed with voltage):

Since standard fuses (in the US) are 1/4-inch in diameter, fuse holders work great for the electrical & mechanical connections with the brass electrodes.    The water resistor tubing is run far away from the high-voltage source, then coiled around an insulating column (plastic sewer pipe) with 8-inch or so spacing between wraps:

More rub:  Water resistors are not very stable; they tend to change value over time.  Causes of this include impurities leaching from the tubing and/or electrodes, temperature, humidity (condensation on the tubing), bubbles on the electrodes, etc.    What this means is that the divider ratio (value) needs to be measured before and (it’s a good idea) after each use.

Measuring the divisor is straight-forward, but requires a high-voltage DC source and a voltmeter.   It doesn’t have to be fancy (NST, Variac, current limiting resistor, diode & filter cap).

For these tests, 5kV and 10kV was applied to the divider input, and the output voltage (millivolts) was measured.   The divider value (or measurement multiplier, depending on how you look at it) is simply the input voltage / output voltage.    Applying 5000 volts, there was a 139.81mv drop across the small resistor.   For 9,990 volts, there was 279.81mv measured.  So, 5000/.13981 = 35,763X, and 9990/.27981 = 35,702X; taking the average = 35733.   So, the voltage readings on the scope should be multiplied by 35,733 to read actual voltage.  One million input volts should read 1,000,000/35,733 = 27.99 volts on the oscilloscope.

NOTE:  Over the course of a week, the divisor dropped from 43,120 to 35,733, but the readings changed only 0.2% between runs on a given day.

Zooming in on captured data (using gnuplot) shows voltage & current oscillations in the neighbourhood of 15MHz and peak voltage of  over 800kV  (23V x 35,733):

Since the current probe used (Pearson 411) is good to 20MHz, the bandwidth of the water resistor divider appears adequate:

The Pearson is shown on the left around the secondary coil RF ground cable.    Real (active) power is then computed by multiplying the current and voltage samples together times the sample interval of the oscilloscope (one-at-a-time) to get Joules; watts (power) is simply Joules per second.   Excel can be used to compute the power (from .csv files) if the number of samples is less than 65,000.   Here’s a Python program to calculate energy if you have a large sample memory scope (greater than 65,000 samples), or run Linux:

#!/usr/bin/env python

"""
Filename:    stats.py
Date:        2010-12-29
Author:      Bill Bishop <wrb at wrbishop.com>
Description:
    Display energy & statistics for the specified .csv file.
    The first column is the time index of when the sample was taken.
    Channel 1 (col2) is assumed to represent voltage, and channel 2 (col 3)
    current.  The specified scale factors are applied before computing stats.
    This program works with Python 2.6 through 3.1

    Be sure the .csv file has no extraneous header information, and
    that the first column is monotonic.
"""

import csv, sys, os

#
# Grab the program (script) name
#
prog = os.path.basename(sys.argv[0])

usage= prog + ": csv_file chan1_vscale chan2_iscale"

if __name__ == '__main__':

    numargs = len(sys.argv);

    if numargs < 4 :             # Sanity check
        print(usage)
        sys.exit(0)

    csvFile = sys.argv[1]        # 3-column .CSV file
    vScale = float(sys.argv[2])  # Voltage scale factor
    iScale = float(sys.argv[3])  # Current scale factor

    reader = csv.reader(open(csvFile, "r"))

    rc = 0           # Initialize row (sample) counter,
    joules = 0.      # initialize total energy value,
    minV   = 10000.  # and min/max values.
    maxV   = -10000.
    minI   = 10000.
    maxI   = -10000.
    for row in reader:
        #
        # Convert text input values to floating point
        #
        t = float(row[0])            # This sample time
        v = float(row[1]) * vScale   # Scale voltage
        i = float(row[2]) * iScale   # Scale current to amps

        #
        #  On the first row, we have to set our first and last time values
        #
        if rc == 0:
            firstT = t  # Grab first time value
            lastT = t

        rc = rc + 1

        #
        # Power (Joules) is computed by current * voltage * sampleInterval
        #
        joules = joules + (i * v * (t-lastT))   # bump energy by this slice
        lastT = t                               # Step to next time value

        if i < minI:                            # Update current stats
            minI = i
        if i > maxI:
            maxI = i

        if v < minV:                            # Update voltage stats
            minV = v
        if v > maxV:
            maxV = v

    #
    # Display results
    #
    seconds = lastT - firstT    # Compute total time-span of this run
    print("%s: Power: %.2f Joules, Period: %.6f seconds" %
        (csvFile, joules, seconds))
    print("\tMin/max V(%.2f,%.2f), Min/max I(%.2f,%.2f)" %
            (minV, maxV, minI, maxI))

As Always, be careful when working with high voltage, and use the buddy system.   Here’s a link to the CSU high voltage safety manual.  It has good information, and few accident examples (nothing serious).   We all want to be able to enjoy the fruits of our labor!

Posted in High Voltage | 2 Comments

World Robotic Boxing

Cool: The domain “WRB.COM” I sold to Disney several months ago (through a 3rd-party negotiator) has now surfaced as the web-site “World Robotic Boxing” for the movie “Real Steel” starring Hugh Jackman.  (Real Steel Trailer)  Excellent!!

Posted in Internet | Leave a comment

The 12 Days of Tesla

While successfully transmuting lead into gold with a small Solid State Tesla Coil (SSTC), it became apparent that a much larger Tesla Coil (TC) would be necessary to make a dent in the national debt.  The new coil would need to have more output (500-700kV), the ability to run for long periods of time, and be relatively immune to mis-tuning (since the transmutation process causes considerable field distortions).  After some consideration, a classic Spark-Gap Tesla Coil (SGTC) would fill the bill.

After absorbing the 20 years of Tesla Coil Builders Association Newsletters, which are now being hosted at TeslaUniverse, along with The Ultimate Design Guide to Tesla Coils, it became apparent that an eclectic mix of electro-mechanical components (and some fair amount of engineering) were going to be needed.

Desiring to get a larger TC up and running ASAP (since time IS money), I scoured the Internet (and Ebay, in particular) for parts.  After firing off a list of questions to Alan Majernick at TeslaStuff about his “Fast Start” packages (see his site for what’s included), he responded with a set of plans (gratis), and thoughtful answers.   Alan disclosed that he had sold a large number of these packages with NO failures, and at just under $1000, the price sure seemed right.   Alan had all the parts in-stock, and I could take delivery within a week.    A week after the order was placed, the parts arrived.

Now, just to set the proper expectation, Alan’s fast-start packages are not kits.  They contain the hard-to-find, long lead-time, and more esoteric parts.    General construction tools and materials are necessary to complete the coil.  At the end of this post is a list of materials and costs for additional required bits & pieces.   What follows is the 12-day “build-log” of the coil.  Acquiring a copy of Alan’s plans from his store, and following along will help you get the most out of this build-log experience.      If you want a spoiler for the (almost) completed coil,  look here.

Day 1 – Lay of the land:  After unboxing and checking the parts for damage, and buying wood & PVC parts for the base:

NOTE:  The two 30ma 12kV transformers in the background are ones I already had.  This blog entry covers their demise.

After testing the fan and Variac, I bought PVC tubing and end-caps for the base (stand) from a local hardware store, and ordered a sheet of 3/8″ XX phenolic from Amazon to mount the fan and spark-gap on (it should have been grade XXX, but more on that, later).   I also ordered more ceramic stand-offs from Ebay to mount the second tier of (MMC) capacitors with.   Finally, I purchased a polypropylene cutting board from a local grocery store that I later butchered to fabricate the primary coil supports.

Day 2 – Cutting up:  Since I started with a plank that was cut into 18″ squares, and the plans call for circular discs, I cut, sanded, routed (and sanded again) the squares to shape.   If I had “do-overs”, I would have left them square (more mounting area for assemblies):

I then purchased stain, polyurethane, brushes, sandpaper, etc.  to finish the platform discs.  Three of these discs are joined at roughly 1-foot intervals to create the base (with casters on the bottom disc).  The discs were then fine sanded in prep for finishing.  I bought more stand-offs from a local shop and electrical connectors for the MMC bank from a hardware store.

Day 3 – Whew, busy, busy:   I applied the first coat of stain to the discs, and sanded the PVC endcaps flat on top so they could be bolted to the discs.   The rest of the endcaps were sanded to remove gloss in preparation for painting.   Screw holes were drilled in the end-caps (using the nubs on the inside for centering).    The holes allowed me to lift & move the end-caps during painting, using a piece of 12-gage solid wire.  The end-caps were cleaned with lacquer-thinner to remove markings, fingerprints, etc,  wiped down with isopropyl, then painted (two coats):

The second coat of stain, and two coats of polyurethane were applied to the discs:

Lengths of PVC pipe were cut to 1-foot lengths, sanded, cleaned, and painted.  They are the supports between the discs.  After painting, the tubing wouldn’t fit nicely into the end-caps since they had an increased diameter (doh).   The last 1.25″ of each end of each tube needed to be sanded down to fit easily into the PVC caps.

Day 4 –  Base station complete:  The painted plastic parts were coated with Rust-oleum crystal clear enamel, and the base construction completed with casters:

A trip to the hardware store yielded the Lexan and ten feet of copper tubing to fabricate the strike bar.

Day 5 – Zappy, Zappy:  The 32-capacitor MMC array was fabricated:

The bleeder resistors are mounted below each capacitor, underneath the plastic.   After this, the primary coil construction was started by drilling the large Lexan circle and mounting the large stand-offs.

Day 6 – Primary drive:  The polypropylene cutting board purchased earlier was sliced-up and prepared for mounting the primary coil (copper tubing).   The secondary end-caps and mounting hardware were then finished up:

The cut-outs for the primary coil are on 5/8″ centers.    The plans detail how to create the mounts.   The holes on the ends of the polypropylene are for mounting the strike-bar.  At this point, I started gathering parts for the coil control console (box, switches, AC connectors, cable, etc) on a trip to the local surplus store (OEM parts).

Day 7 – Great progress:   I wound the primary coil with some help from a friend (Jeff). 4-hands are VERY useful for this part.   A hot-glue gun was used to secure a few “problem” mount points:

The Lexan strike-guard mounts were then cut, drilled, sanded, and installed for the strike-guard.   This component is critical, as it keeps stray arcs from hitting and damaging the secondary (and other parts).  Notice that the ends of the strike-guard tubing are joined with a small piece of flexible plastic tubing.   This is important to keep the guard from altering the resonant frequency of the primary coil:

The secondary was then mounted to the base, connectors soldered to the primary coil & strike guard, and the toroid mounted:

Lastly, I started laying out the components for the Terry Filter.  This filter is named after Terry Fritz, the designer, and well regarded Tesla Coil researcher.    Some of his papers can be found here.

Day 8 – Pushing hard:  I cut the plexiglass for the Terry Filter, finished the electronics, and mounted the components:

The safety gaps on the Terry filter were set just far enough apart so that the high-voltage from the neon sign transformers (NST) didn’t arc at the highest AC setting of the Variac.  The transformers were then painted  and sealed:

Brian then helped me with mounting the fan (arc quencher) and spark-gap on the phenolic board (as-well-as making another trip to the hardware store for nylon stand-offs):

Before firing up the coil, the primary needed to be approximately tuned by connecting one side of the MMC bank to a position on the primary coil.   Alan’s plans specify somewhere around the 9th or 10th turn.    A more accurate placement was determined by first measuring the resonant frequency of the secondary, then setting the tap on the primary for the same frequency.

Day 9 – First fire:  We hurriedly finished up the spark-gap and fan board.  In the rush, I neglected to cover the spark-gap bolts (bottom-side of the phenolic) with corona-dope as Alan outlines.    The high-voltage wire was cut, tinned, and connected to all the components.   A plexi-glass base was added to the MMC array for stability.    An (RF) ground rod was driven into the earth, the TC carted outside, and initial (test) wiring of the coil completed.    The ubiquitous aluminium ladder was then erected as a strike target:

The first-fire was performed at dusk, and the coil ran for about a second before the spark-gap shorted out.  There was flash-over between the bolts on the bottom of the phenolic.   So the carbonized trace was “dremmeled” out, and slathered with corona-dope.

Day 10 – Flash-fry and victory:  The coil was fired-up, only to have arcing on the TOP of the phenolic board from under the insulators & stand-offs.   My thoughts are that grade XXX phenolic might have prevented this:

Once again, the carbon path was cut-out with a dremmel tool, and covered with corona-dope.  When the coil was fired-up again, the arc just ate through the dope.   I replaced one of the metal bolts with a nylon bolt, and was able to get a successful run for a couple of minutes before arcing was noticed between the fan and the other spark-gap mounting bolt (still metal) at 6+ inches.   A quick trip was made to the hardware store for another nylon bolt, and there has been no problem since.     We shot a few pictures, and shut it down for the night (10PM, and it’s very noisy).

Sparking Tesla Coil

Day 11 – Safety first:  Now that the coil was working, the final wiring needed to be completed.   The control box parts:

In process:

There has been some controversy about whether to install AC line filters “backwards” to keep interference from getting “back” into the power mains.   Alan has had no problems with any of his systems when installing them in the proper “forward” direction.   After a bit of research, I am in agreement.  The finished control box has Variac connections on the left, Coil connection on the right, and AC input/fusing on the back.

Once this was wrapped-up, I wrote a safety/operations manual for the coil.

Day 12 – In-to production:   The final wiring was neatly run on the coil, and cables from the coil to the control box completed.  After this, the coil was transported to the (secret) lab, where it has been creating copious amounts of shiny precious metals!

Ok, that last part about the precious metals,  I just made up.

A costing of materials above and beyond the “Fast Start” package ran something like:

Remote Control power/control panel parts: $50

The Base:
Wood: $25.00 (5’x18″ board)
Wheels: 2 locking, 2 not:  $25.00
Paint, stain, brushes, sandpaper, polyurethane: $45.00
Pipe caps & PVC pipe: $20.00
Rust-Oleum crystal clear enamel: $4.00
Screws: $5.00

Primary Coil mounting hardware:
Polypropylene cutting board (primary coil mounts): $10.00
Strike guard plastic and nylon (nuts & bolts) hardware: $7.00
Strike guard 1/4″ copper tubing (10′): $9.00
Strike guard copper lugs: $3.00
Plastic tubing to slip over strike guard ends: $3.00 (lots left over)

Phenolic Board for spark gap(s) & fan mounting: $35.00

Terminal strip for fan AC wiring: $2.00

(Threaded) ceramic stand-off’s for MMC (4-base, 3-top tier): $24.00

Misc Nylon nuts, bolts, screws, spacers: $30.00

Hot glue gun & glue for mounting primary coil
Polypropylene and stand-offs for mounting Terry Filter (& resistors)
Misc hookup wire

Grounding strap: $3
Grounding rod:   $12

Special thanks to Jeff & Brian for their help on this project.

I would like to mention that Alan at TeslaStuff has been very supportive and responded quickly to my emails during the purchase/build process.   Thanks Alan, for helping make this coil a reality!

Above all, be safe.   This project generates potentially lethal voltages.  If you have never worked with high voltage before, find someone who has, to help you.     Never operate a Tesla Coil around someone who has  a pacemaker or other biological implant.   Never operate a Tesla Coil when drowsy or (self) medicated.  All warranties void where prohibited, YMMV.  Keep all pets at a safe distance.    Never put food in the toroid to attract squirrels.   Do not put a Tesla Coil near a gate or fence to keep the bear out.

This guide has lots of other useful information.

Posted in Tesla Coils | 2 Comments

A Mini-Solid State Tesla Coil (SSTC) Kit

Here’s a compact little Solid-State Tesla Coil (SSTC) that you can run INside your home (with the appropriate RF ground).   It’s quiet, doesn’t produce a lot of Ozone, and is fun to play with (as-well-as being a good introduction to SSTC’s; it’s my first).   Many people have built coils around this design, so there’s a good body of knowledge for it;  that’s very important if you find yourself needing some help (search for “Steve Ward mini-SSTC”).     Steve spent several years iterating on this design, and the final schematic is here:

If you go-to his website, there are several other (earlier) variations along with pictures and additional information (spark effects, etc).    Dan McCauley at Eastern Voltage Research, has a nice “development kit”, the SSTC 2.0, incorporating this design for only $99.00 (resonator not included), which is what I’m covering today.

First, a little background:    SSTC’s are differentiated from “classic” TC’s by way of not having spark-gaps (noisy & inefficient, but capable of handling lots of power), or high-voltage capacitor banks.   The voltage is low (comparably – hundreds vs thousands of volts, still hazardous), and the power is switched via solid-state electronic components; in this case a pair of IRFP260 MOSFET semiconductors (replacements are a buck or two on ebay) in a half-bridge configuration (don’t worry if you don’t know what that means).   A brief historical background on SSTC’s can be found here.

Another neat thing about this design is that it incorporates a self-tuning interface (Jan Wagner/Justin Hayes).   The “classic” SSTC has to be manually tuned by adjusting a high-voltage tap (wire placement) on the primary coil.

With that said, let’s get started.    The kit as it comes from Eastern Voltage Research contains only parts and a schematic.   There are no build instructions and no technical support, so as such, it’s considered an intermediate kit.   Access to a voltmeter and oscilloscope is recommended for debugging.   Here’s what I received (not shown is the transformer and aluminium frame that the circuit board is mounted in):

Notes on parts:   The bridge rectifier labelled  KBL01 is for low-voltage (from the transformer) in place of the specified KBL005.  The two diodes across the IRFP260’s (labelled FFPF08S60) are not needed, and are not included in the kit.   A 2200uf 35V capacitor was substituted in place of the specified 4700uf  filter capacitor.

I started by populating the resistors, small capacitors, sockets for integrated circuits, diodes, and DIP switch (selects 555 timing capacitors).   Then I populated the components for the low-voltage power supply (5 and 12 volts).   There is an important jumper outlined in the errata that needs to be installed.  I placed the circuit board into the aluminium frame to set the height for the two voltage regulators; their screw holes need to align with the holes in the aluminium chassis before soldering them in place (same for the MOSFETs, later).

The black and red wires on the transformer (T1) are connected together (red-to-red, black-to-black).   The blue output wires are used to power the circuit (connect to CTRL+ and CTRL-); the yellow center-tap wire is not used.

Once the low-voltage supply operation and power paths were verified (checking the power & ground pins on the IC sockets),  I installed the diodes, integrated circuits, and remaining capacitors.   On my board, pin 4 of U6 was not grounded as the board appeared to be “over-etched”, causing the ground plane to be disconnected.   Dan said this wasn’t a problem on his other boards (YMMV).   A small jumper fixed the problem.

I wound the gate transformer according to the instructions on page 18 of the Flyback Driver Version 2.0 manual.   It’s important to get the wires wrapped tightly and neatly around the core; this effects coupling efficiency and driver capacity.    Page 11 of the SSTC 1.0 manual describes the circuit operation, including the antenna circuit.  Most of this is the same for the SSTC 2.0 kit.    After installing the rest of the components, here’s what it looks like (the missing screw-post arrived later).

Dan suggested using a 100-watt light-bulb as a “dummy load” for testing (across OUT+ and OUT-).  Too bad they’re going to be outlawed next year, so be sure to stock up (I’m still not clear on why using less energy is “better” than being poisoned by mercury).  I used a Variac and isolation transformer connected to AC+ and AC- for testing:

I had to play with the position of the antenna a bit to get it working.   I also found that if my arm was too close to the antenna, the circuit would stop working.   I did my testing with position 1 of SW1 closed and the other two open (1000pf capacitor enabled), and the two potentiometers R3 & R4 centered.

Once things seemed to be running, I bolted the MOSFETs and voltage regulators to the aluminium chassis, using the pink insulating pads between the devices and the chassis.

Now for the moment of truth: hooking it up to a Tesla Resonator (coil)!  I kept the Variac in-line for power control.   Using full power, the MOSFET’s got hot very quickly (30 seconds or so, depending on the frequency settings).    Running at 1/2 power, they stayed cool to warm.    A three-amp in-line fuse and AC filter were added for line conditioning (some of the “touch lamps” upstairs were observed to be self-actuating 😉

Steve (and others) recommend a 4.5″ by 10″ secondary and 4 turns of 16-gage wire around the bottom as a primary.     I had a 4.5″ by 20″ secondary “standing around”, and ran 5 turns of 16-gage around a 6″ form for the primary.   I got a bit of corona when I tried wrapping directly onto the secondary.   I’ll try again with higher-voltage insulated wire (instead of “speaker” wire):

The sparks & spray shoot off the top of the sphere break-out.   A range of effects are achieved by selecting the different capacitors in SW1 and by adjusting the two potentiometers (pots). High-frequency “hissing” with position 1 of SW1 selected.   Selecting position 2 yields a range of tones by adjusting the pots, and with position 3, the frequency is so low, that the effect is of intermittent operation.  Adjusting Variac voltage yields longer or shorter sparks.   Be sure to monitor the temperature of the MOSFETs and don’t let them overheat!

I bent the end of the antenna away from the resonator, as corona was visibly emanating from the tip (it sizzled my finger too!).

Even though this design is relatively low power, it can still pack a wallop.  Be sure to follow proper safety guidelines when using your coil, and NEVER use it around anyone that has a pacemaker or other biological implant.

In discussions with Dan, he says that when his current stock of boards runs out, he’s going to discontinue this kit.    For $99, I’ve found it to be a great value; I’m ordering my second one tonight!

Posted in Tesla Coils | 11 Comments

Making Printed Circuit Boards (PCBs) With Your Laser Printer

So you have a picture of a circuit board, or maybe a Gerber file, but you need that hard PCB STAT (right now)!  Pop-Quiz: what-do-you-do?, what-do-you-do? There are quite a few sites out there on how-to make PCBs with your laser printer, a clothing iron, and an old magazine; I found this one quite informative.    Beware, some of the new laser-printers don’t use toner that transfers (and ink-jet printers are right-out).

Not to be outdone by the simple (or being subjected to untold rounds of trial and error), I found a couple of companies that offer enhanced transfer paper and systems.   One is Techniks Press-N-Peel PCB film, and the other, PCB Fab-in-a-box.   The Techniks site doesn’t have a lot of information about their products, but Frank Miller’s Fab-in-a-box site is chock-full of great information and tips.   After making a call to Frank, and getting quite the “download” of information from him, I decided to try his products (also, Frank boasts 5-mil accuracy with his techniques).

You can use a clothing iron with the Fab-in-the-box process, and Frank has articles on how-to calibrate the iron, and how-to get the best transfer.   However, Frank has done a lot of research on alternatives to the clothing iron, and has settled on the GBC H-220 as a good low-cost “Toner Image Applicator” (TIA).    I found one at Office Depot for $95, and had a $20 off coupon, so here it is:

Frank emailed me with a reminder to perform a modification to the H-220 to “slow it down” for better transfer:

This probably voids your warranty,  so be sure to try out your unit before digging in.   It took about 4 and 1/2 minutes for the unit to warm up, after which, I ran the “roller cleaner” sheet through it.    Be sure to use the 5-mil (hot) setting, and let it warm up for a good 1/2 hour before running a board through;  this gives the rollers a chance to absorb heat and create a good “thermal mass” for the transfer.   If you buy the PCB kit from Frank (instead of Digi-key, Mouser, etc) select the “Free PCB Samples” for an extra bonus.   Here’s what came:

and inside:

The kit comes with images that you can compare the output of your laser-printer with.  They should be dark and contrasty.    I had to scale-up the antenna image so, in Linux, I used the following commands:
mogrify -quality 100 -resize 3033×1656! Image136.jpg  (the size determined experimentally)
gimp Image136.jpg
Bump contrast by 50%:
Tools->Color Tools->Brightness-Contrast

Then, from within gimp, using the “advanced” printer settings, I set the “contrast” and “toner darkness” to maximum (after a few test pages were run on regular paper to verify this worked well), then printed the antenna image onto the toner transfer paper.  Lining up the paper onto the cleaned, blank circuit board (I used the 1/2 oz “scrap” that Frank had sent with the kit), I ran the paper and board through the TIA (after letting it warm-up for 1/2 hour).   The first pass through, the paper didn’t stick to the board, so I ran it through a couple of more times (see instructions).   This worked very well:

The next step was a bit trickier for me, and I didn’t quite get it right.    To make sure that the toner has no “small holes” that the etchant will get into, the green TRF foil is then applied via the TIA (laminator) to make a good seal:

I didn’t do it carefully enough, so there were lots of wrinkles, and the TRF didn’t stick very well to the toner.     Since you can’t just do that step again, Frank suggested I scrub off the toner and start over, but I had a resist-pen, and went over the whole pattern by hand (not that critical):

I used the contact etch technique (sponge) with an oz or so of Ferric Chloride (from Radio Shack), and a few minutes later:

Using the 1/2 oz .032 copper boards made a huge difference; the Ferric Chloride etched quickly, and the board can be cut with scissors (instead of a shear).     Frank’s “Ancillary Items” page has some neat stuff;  I bought some of the silver plating powder from cool-amp, and after scrubbing the resist off the circuit board with Acetone (Naphtha didn’t work very well), I rubbed on the silver plating powder:

Notice the raw copper on the left side of the board compared to the (now silver coated) patterns.   Pretty neat stuff!   Now, on to assembly!

Printing and transferring using the white TRF foil, component placement outlines can be created on the circuit board.     Now that I have the TIA (laminator), I’m looking forward to trying the Decal-Pro system!    Frank has been very helpful, and extremely responsive to email queries — overall I’ve had a great user experience.

Posted in Electronics | 5 Comments