Testing the TK Tar Baby

[MileHigh]

Hey TK!

Very nice workmanship.   That could be a useful tool in general, just try not to over-current it.  It should be able to resolve ambiguities in the net current flow direction for some circuits.  You know how those wonderful DSOs are only as good as your skill set with respect to making proper measurements.

I suggest you do a preliminary current reading with a multimeter first.  Always do a visual check and feel for excessive heat between thumb-forefinger.  You probably have a good 1 1/2 seconds before a catastrophic failure.

High heat mode would spell DOOM for the LEDs of DOOM.

The logic is that if you have say a 48-volt power supply, the LEDs of DOOM are only "stealing" about one volt (?) from the supply source.  (I forget the voltage drop across a vanilla LED.)  So in theory you are not going to disturb the device under test too much.

So the big question is we saw the GREEN LEDs light up when connected to the Tar Baby.  Were they LEDs of Salvation or were they LEDs of DOOM?

MileHigh

[TinselKoala]

Well... it is the Cathode side of the Green LEDs that is connected to the negative battery pole.

[MileHigh]

Eeeeeeeeeeeeeek!

[TinselKoala]

Meanwhile..... back at the DeepBunker..... I finally got a round two it, so I hooked up the Tar Baby to the Clarke-Hess 2330 Power Analyzer aka Sampling V-A-W meter.

It is true that this instrument says that it is supposed to be accurate below 600 kHz, so it might be missing some of the action in the high frequency oscillations. But there it is, for what it's worth.

I used 4 x 12 V batteries, and set the Tar Baby's 555 driver to make the highest amplitude stable oscillations, which usually causes my inline cheapo DMM to indicate 200-210 mA.

Figure 1: Clarke-Hess 2330 readings: Input power to circuit.
Battery >>> CH2330 >>>> Tar Baby >>> water heater element load

Figure 2: CH2330 readings: Output power to load
Battery >>> Tar Baby >>>> CH2330 >>> water heater element load

Figure 3: Waveform during CH2330 testing. Note the 60Hz ripple envelope on the overall waveform. The CH needs shielded input leads, I guess. Top trace is pin 3 output from 555 timer @ 10 v/div, bottom is common drains @ 20 V/div. The channel zeros are on the nearest graticle line to the grey dots on the bezel.

[TinselKoala]

Eeeeeeeeeeeeeek!

For most LEDs the cathode is the little cup-like structure inside, and the anode comes over the top and contacts whatever is in the cup with a little wire. I always remember it as "cup==cathode" and usually you can see this inside the LED.  And I just confirmed it with my bench power supply: when the negative lead from the supply is hooked to the cup-cathode side of the green LEDs, and the positive lead to the anode side,  they light up and the red ones don't.

So what are the implications of this, especially considering two things:
First, there is the AC oscillation and it is possible to extract power from it...
and
Second.... Rosemary now says that she has never claimed that the batteries are _recharging_, just that they aren't _discharging_. And you know what happens next: the current comes out of one end of the battery, and a current goes into the other end of the battery, right? So they are equal and opposite and the battery remains charged. The LEDs prove nothing because they do just what's expected: they indicate the flow out of one end of the battery (or in the other end.)

I think the oscillations are affecting my brain or something.