FacebookTwitterDiggStumbleuponGoogle BookmarksRedditTechnoratiLinkedin

Danger: High Voltage! [UPDATE #5 - Testing of HV noise suppression]

Written by Robert Greenyer on .

We have great news on the progress made by Stoyan Sarg with his triggering circuit design he is providing us. Before getting onto that, a bit of background.

For a good many years people have been attempting to trigger or enhance LENR events through the use of various high voltage signals, pulses, waveforms, glow discharges, arcs etc, some of these techniques we noted in our triggering mini project open document which can be found here:

New Fire triggering open document

We’ll first mention some of those players who are actively employing these techniques in this field of research below. The most high profile of which is Defkalion Green Technologies (DGT) who use high voltage in order to both stimulate and regulate/control their reactors output.

Defkalion Green Technologies

Whilst the specific structure of the core of the Hyperion reactors is unknown, what can be revealed is the nature of their driving circuit. This has been deduced from a picture that Mats Lewin took at last year's live demonstration, where he was the only external witness. You can go and see the original photo on his blog here:

Defkalion Milan update

From manipulating one of the images in photoshop, we extracted this key product information.

Fig 1: Product label on DGTs HV source

We subsequently discovered that these Tecnolux devices are widely available across the world as they are standard neon signs high voltage transform units. Here is one supplier


The specific transformer they are using is listed in the table on that page as


Which you can find near the bottom and it is rated as 575W!

Judging the input power in DGTs test from this high voltage supply is not trivial, looking at the paper published from ICCF-18 here:


It says:

Triggering the effect is accomplished by hydrogen discharge across the two W/TZM electrodes at V = ~ 24 kV, using the current I = ~ 22 mA DC current with ~ kHz frequency”


24,000V X 0.022A = 528W

It was later reported that the input was switched on and off with a duty cycle of 20 percent, Defkalion must have built a bespoke switching circuit to enable short pulses though we have not seen the structure of it to confirm. They claim that the HV power supplied to Hyperion is insignificant (because of short pulses) but we did not see evidence of its measurement.

The reported parameters are close to the rated 575W and it might be argued that there would be an initial surge on the switching on that could get nearer that, so assuming it does, we could be looking at an average input power from the HV of approximately 115W (20% of 575W), though there would be losses in the HV unit when not supplying the reactor also.

The calculations above are a bit simplistic however because it is difficult to measure the power of HV used like this, we assume they estimate by the pulse duration and the repetition rate but other than the 20% figure, we did not see detailed data from them. Given the unknowns we cannot be certain of the input power from DGTs HV.

If you have any ideas for the kind of switching circuit needed, please post them below.

Other relevant parameters were; pressure in the reactor was between 1-8 bar and their operating temperature range of 180 - 849ºC. They said the reactor does not show excess below 180ºC and above 849ºC importand structures inside the reactor to start to fail.



Mizuno has been one of the leading pioneers in LENR and his seminal work has stimulated a range plethora of beautiful looking (and sounding) glowing beakers over the years, but they need very careful work to establish the COP as we know only too well with the careful work being undertaken in France. Mizuno’s more recent research has taken inspiration from those working in Ni+gas systems, however, he has brought two interesting innovations to the table. The first is to work at very low pressures say 0.003bar, enabling much lower voltages and the second is to create his active reaction matrix in-situ, thus minimising contamination of its critical structures.

He is using a glow discharge 600-800V 20-30 mA for preparation, and up to 1000V in combination with resistive heating to over 200ºC during active runs. More information can be found in these document:



Very importantly, he carefully studies the relative effects Ni and Pd with H2, D2, H2O and D2O in the same experiment sequence and this is exactly the kind of parameter sweeping needed to get to the bottom of LENR.

We are carefully considering the implications of this work in our current research thrust. Alan Goldwater has suggested this little controllable device family as a potential solution.


However, Bob Higgins noted that this may still need some kind of ballast due the wide needs in voltage and current between formation of the plasma and the maintaining of it, he says “Normal operation will be for the output voltage across your device to maybe go from 5000V with no arc/plasma formed to 200-400V when the plasma starts.”

The team are discussing ideas right now, however, if you have any solutions that would match Mizuno’s experimental parameters, please, as ever, make them below.

Stoyan Sarg

We have great news, not only has Stoyan finished his prototype Tesla technology HV sparker device, he has already shipped it! Here is the UPS tracking code so you can track its progress

1Z 7R6 55X 68 4560 8308

He says it will operate in two modes with the following parameters:

  1. HV AC mode:  pulse amplitude 40 kV; pulse rate 5 – 30 p/s; power consumption from a 6V battery – 5 W.

  2. HV DC mode: pulse amplitude 20 kV, pulse rate 3 – 8 p/s, pulse energy 0.2 J, power consumption from a 6V battery – 7 W.

The pulse energy in HV DC is easier to estimate because it is based on the discharge of capacitor of a known value. The pulse energy of HV AC pulses is indirectly estimated from the input power consumption and the repetition rate.

The efficiency is between 90 and 100% so it has a low power consumption. It could be supplied from a 6V lead acid battery, these can readily be bought for under $10.

Battery supplier

Whilst a large battery would permit longer active run time, using a small 6V 4.5AH one over a discharge cycle would demonstrate that the power draw is as claimed.

Fig 2: A look at the outputs and controls for the HV device

The grounding will be in one point which will help to reduce EMI noise to the LENR control system. In case of noise problems the device could operate in on/off mode, simply by switching the state of the power supply (for example 2 sec on and 2 sec off).

It gives directly short pulses without any need for a pulse generator (5 to 10 pulses per second).

He has enclosed it all in a neat box measuring 12’’ x 9” x 4” in order to maximise safety. I will adjust for 20 KV DC pulse (spark gap of 5 mm at normal pressure). The HV pulse is a discharge of 1nF HV special pulse capacitor charged to 20 kV. The device will be supplied by 12 V battery and the power consumption is about 1A.

Fig 3: A look at the inside of the device

Stoyan has done such a wonderfully professional and thoughtful job with this prototype and we love the way you can see into it, it is very in the spirit of the power supplies and control systems HUG designed that have allowed our LOS approach.

Supply and control

One pole of HV output must be grounded to the power cell body, the present design of the device must be supplied only from a 12 V battery that has no connection to ground. Therefore the turning ON/OFF must be from a transistors switch supplied by this battery but controlled through a HV optocoupler.

The current Champion spark plug used as an output connector has an internal resistance which is not desirable. Alan Goldwater recommends replacing it with a Champion N7YC or similar spark plug, they have a copper core allowing for high heat dissipation, Nickel electrode, helping to avoid cell contamination and can be found here for just $2

Champion N7YC supplier

Stoyan is supplying a HV cable from car to be put between the HV output and the ground at the rear side of the box. It will be for safety protection when testing before connecting to the system. Stoyan is going to lend this version of the device to the MFMP for 6 months whilst he is in Europe as he would like to use it later in the year for his plasma research.

He says “It must be used only by qualified person for research purpose. I will send you also instruction for connection and safety considerations." We are talking up to 40,000V, that can be deadly, so we need to be really careful basically. He goes on "“In this computerised era, where working with very low voltages is the norm, many are not familiar with this kind of technology which is very dangerous if not handled properly. For this reason, this prototype is for research only, not for wide distribution. For open distribution I will suggest a modified schematic that will have better safety.”

Here is a video of Stoyan giving an overview of the device.

UPDATE #1 - Stoyan's connection diagram

Stoyan has provided a PDF (see below) showing how to connect it to a reactor, safety considerations and how to calculate spark gap.

HV Sparker usage considerations

UPDATE #2 - Sparker arrives in Minnesota

For those that were following the UPS tracking, you will know that the sparker has arrived at its destination without any undue delay.

UPDATE #3 - Alan Goldwater makes an HV passthrough

Even before DGT showed us reactors with conventional spark plugs protruding from them, other researches, such as Brian Ahern were employing or designing into their reactors, High Voltage (HV) pass throughs - though Brian never had a chance to properly use his - a situation we aim to rectify. 

Now, off the shelf HV passthroughs can cost $300 to well over $1000, look at the MDC site for examples. Now we are all about bringing this technology to the masses, so we needed to find inventive ways to bring the costs of research down from these lofty heights.

With Stoyan Sarg's sparker in hand and positive reports from Mizuno's HV driven work in Japan, MFMP team member Alan Goldwater was spurred on to make a prototype HV passthrough he had envisioned, by machining one of the many motorbike spark plugs he had in his workshop.

He started with a carefully selected spark plug that can be purchased for as little as $2.

Alan takes over the making of story for his first prototype flexible passthrough in his own words below.

The HV feed-through design is based on a Champion automotive spark plug, chosen for its Nickel electrode with copper core for heat dissipation, and no internal resistor or air gap. 

Figure 4: Alan Goldwater's design for HV passthrough using affordable Champion N7YC spark plug

After looking at some dimensions, a Champion N9YC was used. The higher heat range gives a longer and slightly thinner insulator nose, allowing more working length.

First, the 14 mm x 1.25 mounting thread was cut to 10 mm long, using a .050 thick steel cutoff tool:

Figure 5: Using a lathe to shorten the thread section of the spark plug

Next, the insulator was cut using a Dremel 7/8" diamond wheel in a tool post grinder at 18,000 RPM. Hand feed at about .002"/sec seemed to work well

Figure 6: Modifying the shape of the insulator using a Dremel grinding wheel

Finally, the electrode was hand-threaded 4-40 UNF. This size is a good fit to the .095" electrode diameter:

Figure 7: The spark plug electrode has been hand threaded to allow flexible shape electrodes and insulators to be mounted

This 'first article' was made from a used plug so it shows some stains on the insulator and rust on the hex base.

UPDATE #4 - Stoyan Sarg proposed HV experiment design

Stoyan Sarg has proposed an experiment design for use with his sparker. Given that Tadahako Mizuno's has shown excess heat with Nickel nano structures and electrical discharge in addition to the claims of Defkalion it warrants further discussion. Here is the overview of the proposed reactor.

Figure 8: Nickel Hydrogen Rydberg cell

This assembly is part of a complete apparatus as detailed below.

Figure 9: Complete research system

What is nice about this system is the use of thermal transfer fluid that, through proper flow control, could maintain the reactor temperature above a critical level - In the case of Mizuno's teams work, that was 200ºC - and extract excess heat, perhaps sinking it into a large insulated water tank, giving opportunities for accurate cumulative energy assessment.

A suitable fluid might be Phenyl Methyl Silicone Oil, some of which can operate at over 320ºC, we would appreciate your suggestions.

Also, you can see in the apparatus design, a magnetic powder trap, expanded in Fig 10.

Fig 10: Powder trap

In Dr. Brian Ahern's work with QSI 10nm Nickel powder, it was found that the particles were highly mobile, getting into and damaging valves and vacuum pumps. The valve sealing issue was sort of solved by using nylon based ones. The magnetic trap is designed to take advantage of the magnetic properties of Nickel in order to prevent damage to other equipment.

Please make any suggestions you can to the proposed design which is discussed in the pdf below.

UPDATE #5 - Testing of HV noise suppression

Below you can see some images and videos of testing of the HUGNet boards in conjunction with Stoyan's sparker.

The result was that nothing blew up. The Thermocouples and pressure sensor and even RTDs were all thrown off while the sparks were happening. That may be tolerable if we only spark for a little while and then wait and see what happens.

Angie is going to characterize the current profile on the DC power supply while the sparker is running to help us understand if the HUGnetLab power metering function is adequately fast to get an accurate power measurement, or if the power is drawn in sharp spikes of current that aren't getting sampled well enough.

Add comment

Security code


0 #6 Nell 2024-06-22 11:46
Hi there, just anted to tell you, I enjoyyed this blog post.
It wwas helpful. Keep on posting!

my page: Broken key extraction: mobilelocksmithwallsend.co.uk/
0 #5 Tatiana 2023-12-04 05:15
Over the previous 10 years, internet site
generation and management has fully altered. From building static,
HTML based web-sites we have moved to constructing gigantic databases ԁriven internet webѕites.

My web site Agen Togel Gajahtoto: giphy.com/channel/togel1x
0 #4 ntc33 2018-09-16 06:58
I was able to find good advice from your articles.
0 #3 HermelindaChief 2018-03-22 15:00
I have checked your site and i've found some duplicate content, that's why you don't rank high in google's search results, but there is a
tool that can help you to create 100% unique articles,
search for; boorfe's tips unlimited content
+3 #2 Robert Greenyer 2014-04-04 13:37

Hence our qualification about not being able to assess the power from this component in their system. The Mizuno work was far more methodical and Stoyans understanding of the formation of Rydberg state Hydrogen appears sound and since Leif Holmlid, somewhat an expert in the Rydberg state believes it to be a pre-cursor to successful fusion, there is good 3rd party evidence that this approach may yield important results quite apart from DGTs rumours.
+10 #1 Ecco 2014-04-04 12:43
I wouldn't rely too much on what DGT is doing, to be honest. There are persistent rumors from credible people of shady things going on during that demo and private ones occurred a couple days before.

Here is your generous contributions so far towards our $500,000 target, thanks everyone! : $45,020   Please Donate
See the current state of our booked costs here