Negative waveform Ainslie circuit -- replication help?

[eldarion]

I've been working on trying to replicate your negative waveform Ainslie circuit, with the primary goal goal of replicating the below ambient temperatures you saw. Unfortunately, despite being able to create nearly identical waveforms to the ones you posted earlier (which was a challenge in and of itself), I'm not seeing any temperature drop in the resistors.

The only part of the system I haven't been able to properly replicate is the load resistor you were using. Do you happen to have a part number for it? It looks like an old Ohmite model but I haven't been able to find an exact match, and most of these modern resistors are using some variant of nichrome (rather obvious due to its magnetic properties) or an unusual material that is most definitely not nickel silver. I assume the resistor you were using had a nickel silver winding? Every otherwise similar resistor I've been able to acquire is in the 10-30uH range, not the 96uH range.

You mentioned before you got this to a 5 degree or more drop in private testing. Any chance you could share any hints as to how that was achieved, and whether you had been able to double check the result with a non-contact thermometer? I watched the Youtube video you posted with the non-contact thermometer but the temperature drop wasn't showing up there (I presume the system wasn't operating in temperature reduction mode at that point).
Thanks!

[eldarion]

I'm also curious if you might remember if the resistor was magnetic or non-magnetic, and if you had got similar results with any other load resistor you might have tried. I've got pieces of a possible theory floating around in my mind but really need to see at least one of these frosting up events when I'm at my test equipment to put those puzzle pieces together properly.

[Aaron]

The battery was a 12v garden starter battery from Walmart - Everstart of whatever their generic batteries are - flooded cell. It was not a new battery and with that circuit, it delivers a power pulse then the mosfet turns off and the spike goes right back to the front battery through the fet's intrinsic diode, which are all crappy diodes.

At some point, I used a higher quality diode that went across the source and drain to bypass the the intrinsic diode - it was much faster and I recall gave up to 50-100% more recovery to the front battery. The intrinsic diode killed most of the recovery.

The only thermometers I used were the platinum based thermo-couplers, which are impervious to electric and magnetic fields - that is why they're used for something like this. They're highly accurate. I may have used an infrared thermometer occasionally just to see, but unless those are on flat black, they readings are almost alway wrong.

The Ohmite resistor is a typical nichrome wire-wound inductive resistor. For that experiment, I don't remember which rating it had. Do I mention it somewhere in a video? If it is a small one, it is a 50 ohm if it is a large one it is 100 or 150 ohm. I still have those and could check but not something I'm spending time digging out, but if I run into it.

All experiments were to produce heat. I messed with the settings on the 555 to get all kinds of frequency ranges and at one time, it had quite a longitudinal transmission - I was able to control a touch sensitive lamp across the room.

Where the cold experiment came from was that the 555 circuit's little 9v radio battery was operating it, that battery was dying and did not operate the 555 circuit properly, it was going into all kinds of erratic oscillations - it was then that I noticed the temperature dropping on the resistor up to 5C below the control probe on the control resistor, which was at ambient.

[eldarion]

Thanks much for the additional info! Especially the confirmation I was on the right track by trying 100 ohm and up resistors is helpful. Your original schematic said 10 ohms but I had figured that either the inductance or the resistance had to be a typo, since the 100uH range inductances were only showing up on the 100 ohm and up resistors.

I haven't had much luck beyond figuring out the dying control battery was important, when I had used a dying battery on my driver circuits a few months ago I did see similar waveforms but no temp drop unfortunately. I am using the same type of temperature probes and have the same battery, all to try to get as close as possible to the original circuit. I also run fast diodes across drain and source on all my designs, for the same reason, the intrinsic diodes are really quite awful.

I'll try to work at this some more over the coming weeks, sadly my normal job isn't giving me much free time for fun stuff, not with everything breaking around the house it seems right now too.