Storing Cookies (See : http://ec.europa.eu/ipg/basics/legal/cookies/index_en.htm ) help us to bring you our services at overunity.com . If you use this website and our services you declare yourself okay with using cookies .More Infos here:
https://overunity.com/5553/privacy-policy/
If you do not agree with storing cookies, please LEAVE this website now. From the 25th of May 2018, every existing user has to accept the GDPR agreement at first login. If a user is unwilling to accept the GDPR, he should email us and request to erase his account. Many thanks for your understanding

User Menu

Custom Search

Author Topic: Pulse Pendulum Projects  (Read 19182 times)

PhiChaser

  • Sr. Member
  • ****
  • Posts: 256
Pulse Pendulum Projects
« on: December 08, 2014, 05:55:10 AM »
Hi all,

I have been having a lot of fun fooling around with the 'perpendupetulum' (sp?) circuit that TK posted from Nuts and Bolts a couple weeks back. Some of the things I have tried:

I have a coil swinging over a magnet instead of the other way around... I just wanted to prove to myself that it would work either way.

I got a coil and a magnet to swing in opposite directions. Quite difficult to get it to swing exactly the same and required a second magnet on the base (under the swinging coil) pushing against the magnet hanging above the coil enough to just keep it off center. Without the magnet on the base the coil will swing under the suspended magnet. Without a decent way to test, I'm gonna guess that the weight of the magnet vs. the weight of the coil could get both to swing on the same pulse (without the base magnet)? Surely this has been figured out by someone smarter than I already...

I built the circuit on a wooden hobby stick using a button battery and was going to add the spinning 'lid' thing (sort of a TK/lidmotor tribute) but lidmotor totally owned that one so I just have the portable circuit-on-a-stick (ha ha)...

The current build that I have been watching is sort of interesting (to me anyways):
The only way my setup is different than TK's is that the pendulum is hanging about 10cm and so the pulse from the coil only pushes the magnet (and flickers the backwards LED) every other swing.

I have some questions to those of you who know how to actually use your equipment.
What is the best way to hook up a two channel scope to the circuit? In other words: Should I just look at voltage in and voltage out? Across the caps? ???
What type of waveform would you expect to see?
Does using a rechargeable battery matter to this thing?
Would removing the reversed LED improve the efficiency? Would that little flicker go back into the cap(s) or get lost to the winds? Am I gaining anything out of that flickerless swing?
Maybe I should hook up an old school ammeter on the positive (neg?) leg (in series with the coil) and watch which way the needle swings?
I can't see more than a 0.001V change daily with my cheap DMM so I'm curious how little juice this thing uses...

Put one of these in a small deep picture frame and you have a unique holiday gift! Hiding the coil and adding a Newton's Cradle would be very cool if it could be made to work...

Please feel free to post pulse pendulum pictures.
Thanks in advance for any advice on setup/testing is greatly appreciated.

Peace.
PC

The coil is wound on a regular thread bobbin, 1500 (+/-10) winds of 27?32? Dunno, I don't have any calipers...
The sphere magnet weighs quite a bit more than the coil does and is magnetically suspended from the tip of a square nut from an erector set that is attached to the string.

dvy1214

  • Newbie
  • *
  • Posts: 43
Re: Pulse Pendulum Projects
« Reply #1 on: December 08, 2014, 07:06:13 AM »
Do alllll that.

Watch the level on the caps vs the level of battery charge to get an idea of the power flow. Try pulsing once for every 2 passes and pulse LED off second pass just to see what happens! Neat little device, don't need to tell you why thats all it is. I'm sure you understand.

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Pulse Pendulum Projects
« Reply #2 on: December 08, 2014, 07:24:12 AM »
Neat!
Scoping very slow circuits like this is kind of a pain, I'd use single-shot trigger mode, and I think you'd see the most interesting information by using the common negative as the reference, and looking at the other side of the coil on one channel and the junction of the resistors and transistor on the other channel.
I'm not quite following your description on the rest of the post, do you mean that with the shorter suspension (mine is 20 cm tall with 10 cm apart at the top) you have such a short period that the cap doesn't charge enough to give you a pulse each passage, so you only get one every other passage?
The LED flicker indicates when a pulse happens, obviously, and you probably have already seen what happens when you have too much supply voltage.
I thought of swinging the coil but never tried it. Have you tried using a photovoltaic cell for the power? And does yours self-start easily?

PhiChaser

  • Sr. Member
  • ****
  • Posts: 256
Re: Pulse Pendulum Projects
« Reply #3 on: December 08, 2014, 03:41:33 PM »
Neat!
Scoping very slow circuits like this is kind of a pain, I'd use single-shot trigger mode, and I think you'd see the most interesting information by using the common negative as the reference, and looking at the other side of the coil on one channel and the junction of the resistors and transistor on the other channel.
I'm not quite following your description on the rest of the post, do you mean that with the shorter suspension (mine is 20 cm tall with 10 cm apart at the top) you have such a short period that the cap doesn't charge enough to give you a pulse each passage, so you only get one every other passage?
The LED flicker indicates when a pulse happens, obviously, and you probably have already seen what happens when you have too much supply voltage.
I thought of swinging the coil but never tried it. Have you tried using a photovoltaic cell for the power? And does yours self-start easily?
Okay, single shot and common negative as a reference. Thanks!
Yes, exactly right about the shorter suspension (although now mine is shooting off every three out of four passes?)... That makes the most sense why I just see it every other pass; not having enough voltage each time to flicker...
I have not tried a photovoltaic cell for a power source. (I do have one that will work I think, I just haven't tried it yet...)
Every build I have made self starts easily. The coil starts a little slow because it is hanging from it's own supply wires and not thread which gives it a bit of unnecessary friction.
I will try and dig out my scope today and see if I can get some shots of the waveforms...
Thanks again TK!

PC



 

PhiChaser

  • Sr. Member
  • ****
  • Posts: 256
Re: Pulse Pendulum Projects
« Reply #4 on: December 08, 2014, 09:26:06 PM »
So I got out my scope and clipped on the probes and had to use the lowest voltage setting on my scope (0.01v/div) with the probes set to 10X to see the waveforms.
I took a bunch of pictures with my tablet but maybe I should have taken a video because my scope DIED... That is to say the power light is still on but nobody is home when I look in the window ya' know?
Ugh...
I opened up 'the box' and this thing was made in 1965...
I hope I live longer than this thing did...

So anyhow, I have four pictures which pretty much sum up what I got to see before the cool moving green lights went out.
-flicker: when the LED was flashing every other swing (only pulsing in one direction)
-noflicker: the magnet swinging over the coil without a flicker
NOTE: the center line is what I set as my top graticule marker for channel 2 on the bottom. Between these two pictures is the typical every other flicker pair.
-flickerflicker: when the LED is flashing every swing of the magnet it looks like the flicker is shorter still (edit: the cap doesn't fill up as far by the looks of it...)
-nonoflicker: when the LED is flashing after 'skipping' a swing; this is where I had a difficult time catching it with the camera on my tablet. I eventually just kept tapping the 'button' until I hit lucky... It looked like the magnet swinging over the coil kept the cap from discharging a second time (when it wanted to) so that it charged up about 1/3 (edit: looks more like 1/2 now that I take another look) of a gridspace more than the flicker trace. Not sure whether the top trace went higher when the cap discharged. Mph...
So what I think I am seeing is the swing of the magnet pushing the voltage down on the capacitor just before it is supposed to discharge? Why is it different sometimes? I'm sure there is a good explanation for it.
Is it more efficient in the hit and miss mode? I have seen it hit three and skip one, hit two and skip three etc... It tends to level out at at every other one (usually heh). Again, I'm sure there are formulae to describe the effects in minutae...
Looks like the LED flashes for a grand total 0.05 seconds? I didn't get a chance to look at the swinging coil or any other builds before my scope died.
Dead scope sucks...

Peace
PC

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Pulse Pendulum Projects
« Reply #5 on: December 09, 2014, 03:24:34 AM »
Oh no, we've got to get that classic Tek 422 working again if it's at all possible! Fortunately this scope seems to be easy to work on and the photos I'm seeing show that all the transistors are in sockets.
http://w140.com/tekwiki/wiki/422#Specifications

(My traces are quite a bit different from yours but that might be because my scope is faster; in your first shot on the top trace you seem to be seeing, almost, the train of oscillatory pulses-within-pulse that I am getting.)

Is there a "Beam find" button hidden away somewhere? I guess not...   :-[

You may find this link helpful, manuals are linked from there.
http://w140.com/tekwiki/wiki/422#Specifications

Step Zero: confirm that there is a problem:
Disconnect the probes from the scope. Then set all the Red knobs fully clockwise into their "Cal" detents. (Or to the "cal" midrange position if no Cal detents present). Set the display controls on the left of the CRT like you had them in these photos. Look on Page 13 of the manual here:
http://w140.com/tek_422.pdf
and set the controls as specified in Step 1 of the "first time operation" section.

Now power on.... what happens? Do you see a trace at all?

If not...
1st thing to try is to clean all the control contacts with a good brand of electro contact cleaner. I prefer LPS brand but you might not be able to find it where you are. Get a can, power off the scope and follow the directions on the can, spray every switch contact and inside every rotary potentiometer and work them (gently but firmly) through their full ranges several times. Remember where the "gain" and "step atten balance" pots are positioned and set them back to there once you've sprayed and worked them thru their full ranges. Gently pull out the transistors and the 8056 Nuvistor tube (in the earlier models)  one at a time and reseat them carefully. Don't mess with the CRT connections at this point. The Nuvistor, if present but nonfunctional, can be replaced with a common MPF102 FET, according to the link.

Repeat Step Zero, do you see a trace?

Do you have a function generator that you can use to provide a signal for further testing?

PhiChaser

  • Sr. Member
  • ****
  • Posts: 256
Re: Pulse Pendulum Projects
« Reply #6 on: December 09, 2014, 05:29:28 AM »
@TK
I had the ground clip on both probes attached to the negative leg between the capacitors so I think it was just showing the smaller (1000uF) cap? Positive leads were between the resistors and between the transistors. I should have shown how I had the probes hooked up, sorry.
Beam find button LOL!!! No, that would be a nice feature though. Gotta wiggle knobs to find them on this one.
I tried the setup procedure; the trace is gone.
I haven't really tried to take it apart yet, the power supply (backside) seemed really clean inside when I took it off. There were some socketed transistors (?) in that part...
The power light comes on so the power supply appears to be still working. It connects to the rest of the unit through an amphenol (sp?) type of connector where the two parts come together.
I guess I will have to find some contact cleaner. One of the channel knobs occasionally gave the trace 'twitches' when it was being adjusted (and sometimes set the trace too high on the field) so maybe taking it apart and cleaning all the contacts will fix that (and whatever else might be the problem).
I will take pictures (and notes if I think I need them) when I take it apart.
There can't be that many socketed parts to test right?
It seems a daunting task but I'm game...
Thanks for the help TK, I will keep you (and whoever is reading this thread) updated.
Looks like I have some reading to do, thanks for the links!

Peace,
PC





TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Pulse Pendulum Projects
« Reply #7 on: December 09, 2014, 06:34:42 AM »
Bummer.
The "beam find" button is a nice feature that is found on many oscilloscopes, it tests the HV supply by taking the horiz and vert amps out of the circuit so you just get a spot  more or less centered on the screen, as long as the scope is capable of making a beam at all.

But you have no traces on either channel... so it is not likely to be a blown input FET/Nuvistor or the tunnel diodes. It could be that the "unblanking" circuit is bad or out of adjustment... or that there is no HV.

I had a problem with one of my scopes that turned out to be a DC offset swung all the way over, so the trace could not be moved onto the screen, this turned out to be a blown input FET and the associated diode. But since both your traces are affected this is probably not your problem.

Here's where I would get out the DMM and start checking power supply voltages as indicated at the test points on the schematics in the manual. The fact that the power light comes on doesn't necessarily mean the power supply is good or on the correct voltages ... it just tests the line cord, switch and bulb!

Peer into the neck of the CRT to see if the filament is glowing. If the CRT is bad, it's probably not worth fixing, you can get another working 422 for the cost and trouble of replacing the CRT. Sometimes the filament is hard to see even if it's good, so don't give up at first glance even if you can't see it.

Page 44 of the manual shows a troubleshooting flowchart that will help you to narrow down the problem. I'd just check the power supply voltages first no matter what, the procedures for this start in the Calibration section on page 81, section 6-6 of the manual linked above. Checking the HV supply is important, be careful here, you may need to find a suitable HV meter or make a voltage divider to do this part.

Once you know that the power supplies are within tolerance you can start checking the voltages at the test points indicated on the schematics. Look close and you'll see voltages indicated at various places on the schematics, and there are photos of the boards so you can identify where things are at. This will help you to find a bad part in one of the subsections.

Suspect the "blanking" circuit first, you may be able to fix the thing just by tweaking the blanking center adjust pot, see the manual in the Calibration section.

This scope does not look like it would be that hard to repair, and the information you need is probably all there in the manual.

PhiChaser

  • Sr. Member
  • ****
  • Posts: 256
Re: Pulse Pendulum Projects
« Reply #8 on: December 09, 2014, 05:39:10 PM »
It looks like the HV supply is gone. I get about 6+ volts instead of 1385+/-69.5volts at the back of the CRT socket.
I figured my meter would at least 'peg' to the 1. (overload) if there was HV there but there wasn't...
Looks like you were right about the power light TK.
Now how much is it worth my trouble to hunt down the ass end of a 422?
I definitely need more coffee for this type of thing...

Peace,
PC



Kator01

  • Hero Member
  • *****
  • Posts: 898
Re: Pulse Pendulum Projects
« Reply #9 on: December 09, 2014, 07:16:14 PM »
Hello PC,

you may find this here interesting. This guy is serious and has a lot of experience. He has only posted
a few experiments but the quality is excellent

https://www.youtube.com/watch?v=Tm1LmkmYkjQ&list=UUVZmmMOhzhWEUamhGLPSTFQ

Regards

Kator01

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Pulse Pendulum Projects
« Reply #10 on: December 09, 2014, 07:54:51 PM »
It looks like the HV supply is gone. I get about 6+ volts instead of 1385+/-69.5volts at the back of the CRT socket.
I figured my meter would at least 'peg' to the 1. (overload) if there was HV there but there wasn't...
Looks like you were right about the power light TK.
Now how much is it worth my trouble to hunt down the ass end of a 422?
I definitely need more coffee for this type of thing...

Peace,
PC
Fortunately the HV supply is ridiculously simple in the 422, basically consisting of a Cockroft-Walton voltage multiplier and some regulator components. The first thing I'd check, after the connector P701 and the transformer T801 is the 51 volt Zener diode D839. Just a hunch. Do a good visual check of the HV multiplier board and the HV regulator board with a magnifying glass, you may be able to see a burned component visually. Check the connections and contacts at the Centronix-type connector P701 (has the inputs to the primary of the HV supply's transformer),  and check the voltage outputs from the HV transformer T801, you'll want to unsolder the secondary output terminals 7,8,9, 10 to isolate the transformer from the voltage multiplier and filament supply to see if the transformer is putting out anything for the CW bank to multiply. If it is, then check that Zener, and the 10 diodes in the CW bank, and the electrolytic caps C829 and C849, etc.  Schematic on page 181 of the pdf manual. If the transformer is shot you are probably "s.o.l." as they say but if it's good you can probably fix the problem easily enough. Be careful poking around in there, even the CRT filament which is a low-voltage filament, is elevated by the supply to a high (negative) voltage to aid in creating the beam.
To avoid ruining your meter you can connect the scope's AC line cord to a Variac and check the behaviour of the HV system at low input voltages, you should at least be able to tell if it is working before it exceeds the range of the meter that way.

My old Tek RM503 had a fault in the HV system that caused it to blow fuses (by the way, check to see if the line fuse is the correct value and _don't_ sub with a higher current value than specified). The filament supply in the big mains transformer was internally leaking the elevated HV in the 6 volt filament circuit. The HV supply was good but because of this leak in the mains transformer it was drooping severely and also blowing the line fuse. I finally fixed it by installing a completely separate 12-volt center-tapped RadioShack transformer for the filament circuit so the HV had nowhere to leak to; the scope is 100 percent functional now. The hard part, once I figured out the problem, was finding room physically for the "extra" transformer inside the case.
In your case the filament supply comes from the HV transformer itself, a low voltage winding, which is then elevated by a section of the C-W bank. You may actually have the same problem I had and might be able to fix it in a similar manner. Since fil. supply is not from the mains transformer maybe you aren't blowing the line fuse like I was. The way to check this would be to disconnect the filament LV supply at the T801 terminals 9 and 10 and then re-test for HV at the test point. If the HV comes up, with the filament supply disconnected from the T801, then you may be able to fix it by the same method I used. Although the diode FWB D849 should probably isolate the filament windings from the HV... but you never know.
Just thinking semi-randomly...

Right, more coffee.

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Pulse Pendulum Projects
« Reply #11 on: December 09, 2014, 08:15:40 PM »
Hello PC,

you may find this here interesting. This guy is serious and has a lot of experience. He has only posted
a few experiments but the quality is excellent

https://www.youtube.com/watch?v=Tm1LmkmYkjQ&list=UUVZmmMOhzhWEUamhGLPSTFQ

Regards

Kator01
LOL! I am afraid I strongly disagree with you! Do you define "overunity" as voltage amplification?

Voltage multiplication is not _energy_ multiplication. There is no "overunity" being demonstrated in that system. But certainly there are the required number of colored clipleads! A small oscillating input voltage is driving the pendulum, and as the coil swings through the strong magnetic field of the magnets a higher-voltage, very brief pulse is generated which flashes the LEDs. The system is a stripped down version of the perpendupetulum, without the "self looping" aspect where the higher-voltage pulse is triggering the drive pulse. He could add the loop by connecting the LED output to the trigger input of the FG with some tiny phase delay, so that the drive pulse from the FG to the coil happened just after the output pulse from the coil to the LEDs/FG.

But what is "overunity" about that? The PerPenduPetulum does essentially the same thing; the higher voltage pulse from the magnet passage is triggering the transistor which discharges the capacitor to produce the drive pulse in the coil, instead of lighting up some LEDs outside the drive system.

An experienced and serious researcher would never be claiming OU from such a system without a lot more solid evidence than he is presenting. And of course such evidence won't be coming, since there isn't any OU there.

PhiChaser

  • Sr. Member
  • ****
  • Posts: 256
Re: Pulse Pendulum Projects
« Reply #12 on: December 09, 2014, 09:17:20 PM »
@TK I will take it out again and take a peek. It looks pretty straightforward (referring to the PDF), I just wish I had a better soldering iron...
Time for a real disassembly.
I will look for what you suggested.
Thanks again for the help!
The worst that could happen is I don't fix it right?

PC

@Kator Yeah, I saw that one. Not OU, sorry buddy, gonna have to go with TK on that one...

PhiChaser

  • Sr. Member
  • ****
  • Posts: 256
Re: Pulse Pendulum Projects
« Reply #13 on: December 09, 2014, 09:56:03 PM »
Still looking...

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Pulse Pendulum Projects
« Reply #14 on: December 09, 2014, 11:47:08 PM »
Ah, beauty! I love that particular era of circuit board layout. Double sided but with room enough to work. That's the HV regulator board, with the filament supply FWB is the black thing on the top right, and the associated electrolytic cap C849 is the 150 uF, 15 volt unit right there in the top center.  When you unsolder stuff from those boards, then you can heat up the pads and use a wooden toothpick to "clean" or rather open up the holes without damaging the pads on either side of the board after gently pulling the component lead out of the hole. There isn't much that can go wrong on that particular board except the Zener and the electrolytic cap and possibly the FWB. The Zener D839 is given in the parts list and the schematic as a 51 volt zener but I can't make sense of the part number on that diode, a 51 volt zener is 1n3806 but that part number on the diode looks like  it could be 1n3789 which is a 10 volt zener. Weird.

What's the serial number of your scope, maybe there is a change with different s/n.

When the scope is together and powered up, does that neon in the middle glow at all?

If you can find the wire that connects from the Anode of the D839 zener over to the HV multiplier board itself, disconnect that wire from the regulator board and then test the wire (ie the bottom end of the CW bank) for HV. Again, I'd do this by connecting the mains AC line cord to a Variac and slowly bring up the supply voltage from zero, with the DMM connected to this wire from the HV multiplier board. If the basic HV supply is working (transformer and CW bank) you should see the voltage here come up (or rather down, since it is negative wrt the chassis ground.) If the voltage does come up with the HV regulator board disconnected like this then the fault is on the regulator board, which would be good. If the voltage doesn't come up, then the fault is probably in the basic HV supply, either the transformer or the diodes. I'd do this testing with the filament supply disconnected at the transformer, wires at both pins 9 and 10 unsoldered.

If you don't have a variac then you can try doing it with full power from the mains, but I'd suggest here disconnecting one side of the main fuse and putting an ammeter in series with the fuse and the circuit to see what the input current draw is and if it exceeds or even comes really close to the specified fuse rating, stop!
Or you could also try putting a 40 Watt or lower incandescent light bulb in series with the AC mains input to reduce the voltage supplied to the scope so that your DMM isn't stressed when you go to measure the HV. You won't hurt anything in the scope by supplying it with less voltage. A Variac is a very valuable tool and has many uses around the lab.