Gray Tube Replication

[Electrotek]

Hi Mike,

You've reversed the two caps. 16 is in series with the diode, and it's around 20nF. Also, 36 is the lazy D, at the top, with a circle in it. This inductor is bifilar, so it's equivalent to the two motor coils in opposition. What you labeled 36 is the power supply, a NST connected to mains, putting out 7.5 kV at 30 mA.

That's about it, except for the diode T-tap. This won't work with a FWB, but Gray and MC used half wave rectification on the early motors. When the switch to full wave was made, the motor didn't work. But that may not be the reason. In the CSET patent, the T-tap runs through the resistance of battery 18 and the motor, to the grid and the Overshoot Switch.

Jerry's Site - Diode T-Tap Circuit

[mlurye]

Jerry,
Now I can see what you are doing. Thank you for explanations. Do you have door knob caps? Try to replace 20n cap with door knob cap should be something in pF range. It will accelerate charging process.

[Electrotek]

Mike,

You're right, this is a frequency dependent effect. A smaller cap produces a higher frequency. With the very small capacitance of a 1B3 HV TV diode, I can get well over a GHz.

I can also get a very interesting effect when I use both caps at 16 uF. Multiple discharges in the spark, with very hard sparks. Almost like a lot more power than 30 mA, with more voltage, too.

When testing with my CSET, I found it was a lot better to use a larger cap 16. Hence, the value I gave.

[mlurye]

Mike,

You're right, this is a frequency dependent effect. A smaller cap produces a higher frequency. With the very small capacitance of a 1B3 HV TV diode, I can get well over a GHz.

I can also get a very interesting effect when I use both caps at 16 uF. Multiple discharges in the spark, with very hard sparks. Almost like a lot more power than 30 mA, with more voltage, too.

When testing with my CSET, I found it was a lot better to use a larger cap 16. Hence, the value I gave.

Yes it is possible to get frequency to any range, but in Gray's system it was only 6 kHz. And based on my understanding that what cap 16 did.

Why do you need hard sparks? How hard sparks will help in driving electric motor?

If you will decrease space between electrodes and grid you might find that it's easy to drive CSET with smaller cap.

And what I don't understand is, why everybody looking for something in Gray's patents, but not the main purpose of it - driving electric motor with HV impulses.

One more thing that bothers me. According to Mark, stats of Gray’s motor were taken once. Were these stats accurate? Because cap 38 discharges through 2 batteries backward and it will void any accurate readings.

[Electrotek]

Since cap 16 is placed in parallel with the constant arc in the Overshoot Switch, it would have a resonant frequency which matches the front end power supply. According to the stat sheet, this was 200,000 pulses per minute, or around 3.3 kps. However, the diode T-tap is capable of producing a much higher frequency, acting as a sub system. I've found that if I charge a cap from both directions, using this circuit, the cap charges much faster and to a higher voltage. In fact, this diode circuit can blow a cap which will normally stand up to the rated voltage. The hardness of the sparks it produces is, I think, related to the longitudinal nature of the sparks. This may or may not relate to the motor operation and may only be a charging benefit.

I've seen a patent for a thyratron which included a 20" ring around the arc region, to recover energy. This grid provided enough energy to trigger each subsequent firing. When dealing with Tesla's Radiant Energy, the grids can be as large as you want. The important thing is the efficiency of the EMP producing spark. (ie, my Puff Spark or the Water Sparkplug arc.) I've found that a longer arc works much better for this purpose, but this doesn't relate to grid spacing, unless the inner grid is part of a different circuit that the others.

The reason people are looking for clues in Gray's patents is because he used an unusual type of energy produced when a current pulse from a battery displaces electrostatic energy in a capacitor. And the motors may have used some form of pulse compression. If so, would this also affect the battery's current? For some reason the motors were returning more current to the batteries than it took to initialize the pulse.

If cap 38 discharges through the motor with a pulse which is pulled by a magnetically quenched Overshoot Switch, it will only affect one battery. I don't see how it can discharge through the CSET unless the diode polarity is shown accurately in the patent, which is unlikely. If it does affect both batteries, their resistance can be measured or calculated. At the least, one battery would be brought up while the other runs down. Discounting any OU.

Here's the stat sheet:

[mlurye]

Electrotek,
Thanks a lot for good explanations. Now I do understand schema a little bit better.
Please could you repost your attachment 1 more time in bigger size? For some reason I cannot open it

[Electrotek]

Correction

Sorry about the attachment size; I'm still learning how to post these things.

Also, while looking at the stat sheet, I see that the frequency I mentioned applies to the Recovery System. I remember that the front end supply had a vibrator which ran at 100/sec. This strengthens my thought that cap 38 provides the pulse for the motor's operation, rather than cap 16 which is shorted by the arc.

[mlurye]

Electrotek,
What do you think is the purpose of diode 28?

[Beshires1]

WOW! that large enough Mike?

[mlurye]

WOW! that large enough Mike?

Beshires1,
If you can see it, then I can see it

[Electrotek]

Gray's patent says diode 28 prevents arcing at the commutator. Other than this, it may prevent HV current through the CSET. The patent says the HV charge is stored on the HV anode. (Not in the capacitor.) Since this anode is connected to the positive side of the cap, with the capacitive grids connected to the negative side, there will always be a charge on the anode, even with the commutator open. When the commutator closes momentarilly, the LV current interacts with the static charge on the HV anode, producing some kind of "effect" in the grids and the connected circuit. But I don't think the capacitor's voltage can jump the diode, even if a higher voltage pulse from an ignition coil can. (PIV) So the only HV is the static charge on the anode, without HV current. That's why it's static. Produce this "effect" 200 or 300 times between each pulse of the motor, and quite a bit of power can be built up on cap 38, between motor pulses.

When the commutator closes, negative electron current will flow from the battery 40 into the negative side of capacitor 16 - if the diode has the right polarity. This will increase the potential on the HV anode by 12 volts. This increased potential will pull electrons from the motor side of cap 38, giving it a positive charge, and will also pull electrons from the positive pole of battery 18, charging it. Do this a few hundred times in a fraction of a second and a lot of power will build up on 38, ready to pulse the motor. But this would also require your extra diode, shown in the Investor photo, since the motor has its own low speed commutator.

[Aaron]

cap 38 charging

When the commutator closes momentarilly, the LV current interacts with the static charge on the HV anode, producing some kind of "effect" in the grids and the connected circuit. But I don't think the capacitor's voltage can jump the diode, even if a higher voltage pulse from an ignition coil can. (PIV) So the only HV is the static charge on the anode, without HV current. That's why it's static. Produce this "effect" 200 or 300 times between each pulse of the motor, and quite a bit of power can be built up on cap 38, between motor pulses.

Electrotek,

With the power supply on the front end, with HV output - we don't know what the output is I don't think but safe to say at least high enough to charge the cap but probably higher since it appears to be some modified mallory racing coil, wouldn't that easily spark over to the lv rod when connected giving the cap a discharge path over the gap when lv rod is closed?

Also, when the lv rod is disconnected, with the power supply running at 6khz, the hv from that can directly jump to the grid and charge the cap 38 in between commutator switching so the cap 38 can charge just fine without the lv rod even being a part of the equation...charged up with just the front end power supply.

The power supply jumping directly to grids will charge cap 38 directly through the coil without giving the coil any kind of magnetic charge...it just passes the potential. But of course the coil will charge when that cap discharges to rod.

[Electrotek]

Electrotek,

With the power supply on the front end, with HV output - we don't know what the output is I don't think but safe to say at least high enough to charge the cap but probably higher since it appears to be some modified mallory racing coil, wouldn't that easily spark over to the lv rod when connected giving the cap a discharge path over the gap when lv rod is closed?

Also, when the lv rod is disconnected, with the power supply running at 6khz, the hv from that can directly jump to the grid and charge the cap 38 in between commutator switching so the cap 38 can charge just fine without the lv rod even being a part of the equation...charged up with just the front end power supply.

The power supply jumping directly to grids will charge cap 38 directly through the coil without giving the coil any kind of magnetic charge...it just passes the potential. But of course the coil will charge when that cap discharges to rod.

The only information the patent gives about the charging coil is that it takes several pulses to charge the cap. This is consistent with your measurement of 400 volts increase on the cap with each click of your relay. The vibrator driving the coil clicks 100 times per second and the motor pulses 40 times per second, at 2400 RPM. If capacitor 18 were discharging through the CSET, it would never have time to build up a High Voltage charge. This is the front end supply. It averages 1.5 clicks between motor pulses. The few thousand pulses per second is the switching rate of the LV rod.

[mlurye]

Electrotek,
Even with my flimsy setup, my charging cap was keeping up with driving one. To begin with I was using 0.8 uF microwave cap (it has discharging resistor built in).
And now I’m completely confused, how cap 38 will drive the motor. I don’t see discharge path
And you had good idea about collecting radiant energy around the gap, may be that is why early CSETs had so many layers.

[boguslaw]

The problem is the it's quite probable that capacitor 38 was added later and was not there in original Cole circuit. My theory is that ( maybe contrary to what Spokane1 said) , Hackenburger added this capacitor because of exploded batteries and Gray opposed that , because it didn't existed in orginal circuit. That would be resonable to assume.