Oscilloscope Output & AVR

[Ray C]

Hi Folks...

I've been reading the threads about flickering lights from the output of ST heads.  Does anyone have some oscilloscope images of the output wave as well as the field windings?  I saw the graphs on the ST section of George's CD but, it doesn't quite show me what I'm looking for.  I'd especially like to see the waveform of the output of the rectifier.


One more thing...  Has anyone tried cleaning-up the output of the rectifier or, hanging a regulator off it?


Thanks


Ray

[dkwflight]

Hi
I'll be posting a link to an "O" scope picture of the output from my st head soon.
I spent part of this weekend redoing my cooling system with heavy duty hoses, etc.

Have a good Christmas everybody.
Dennis

[Jim Mc]

Learning to work the forum search function can be a wonderful thing...

http://listerengine.com/smf/index.php?topic=634.0

[1966Crew]

Onan figured out a way to combat the flickering lights syndrome on their single cylinder gensets such as the DJA3.0 and MDJA3.0. On the side of the engine timed to the cam is a device called "flicker points". These points are connected to a ceramic resistor and the field coil.

These points reduce the voltage to the field slightley at the proper point in the engines rotation thus reducing the flicker.

Of course this is a direct connected 1800 rpm generator\engine combo. Not sure if it would work with a belt drive system.

Jon

[mjn]

Here is a link to a page with oscilloscope pictures of an ST15 driven by a Changfa engine. http://lerch.no-ip.com/ChangFa_Gen/Day_07/page_01.htm

Of particular interest is this picture:


This a photo of the field voltage.  It appears that there is room for a great deal of improvement here. 

I really need to see if my field voltage looks just as weird.  However, I hate being in the same room as my Changfa hammering away at 1800rpm let alone trying to hook up a scope 2 feet away.

[Andre Blanchard]

This a photo of the field voltage.  It appears that there is room for a great deal of improvement here. 

I really need to see if my field voltage looks just as weird.  However, I hate being in the same room as my Changfa hammering away at 1800rpm let alone trying to hook up a scope 2 feet away.

The voltage ACROSS the field coil is meaningless it is the current going THRU the coil that matters, that is what makes the magnetic field.
The field coil is extremely inductive and like all inductors the voltage and change very fast but the current is slow to change, that is why you get the big spark when you disconnect a coil while the current is flowing.

Measure the current, forget about the voltage.

[SHIPCHIEF]

How to measure the current?
Put a shunt resistor in line and measure it's voltage drop on the O scope?
Any kind of Amp Clamp that can pick up variable DC and show on the O scope?
I would like to stop the flicker too.

[Jim Mc]

Put a shunt resistor in line and measure it's voltage drop on the O scope?

That's a good idea. 

There are clamp-on current probes  that could be used, but there are plenty of them that would not be suitable since this waveform has a DC component and harmonics way past 60 Hz.  So, yep, a 0.1 ohm resistor or similar in series with the field is a very good way to measure current.  I doubt it'll get you any closer to solving a flicker problem though, as the most likely cause of that is cyclic fluctuation in engine speed.

[Guy_Incognito]

I doubt it'll get you any closer to solving a flicker problem though, as the most likely cause of that is cyclic fluctuation in engine speed.

Which brings me to a question I've been meaning to ask :

I'm looking at getting a PM head, but it spins at 3000RPM. Too much wiggle in the frequency would be bad for my application. Gearing up to 3000RPM doubles the variation compared to the normal 4-pole/1500RPM (for 50Hz) systems.

For those of you with a digital o'scope, can you suss out the amount of frequency variation you get?
If you can capture ... 12 complete cycles (at 60Hz), that should be pretty much 2 engine revolutions.
Measuring the time of each cycle should then give an indicative amount of how much engine speed varies over the 4-stroke cycle.

For those with an analog scope, you should be able to set it to trigger at the front of a cycle, and set it so that one cycle takes up nearly the full display. Then you should at least be able to work out the max/min Hz by how much the waveform wiggles back and forth.

Any info would be appreciated.

[GuyFawkes]

see engine mounting v4 thread

[Jim Mc]

...Gearing up to 3000RPM doubles the variation compared to the normal 4-pole/1500RPM (for 50Hz) systems.

You might want to noodle on that a bit more.   The speed variation, expressed as a percentage will be the same regardless of whether you use a 2 pole or 4 pole machine.  Since a 1% variation in speed causes a 1% variation in frequency in either a 2 pole or 4 pole machine, the frequency variation is the same in either case.

[Bikerbob]

Ok, here comes a long post.

So far I have not hooked up the O-scope to the STC head, but plan to do that within a few days.
I have however tested several other heads over the past years. Some are presented here.
Nice thing to have a PC connected O-scope...

Short version:
You get what you pay for...

Longer version:
Please go ahead and read...

Sure, nothing of this applies to the Listeroids and ST heads, but still... Some sort of reference is always good.
Or as we say in Sweden: "Knowledge is never heavy to carry around with you"....


Note:
-Everything tested is 230V.
-Everything tested is 50 Hz (Europe).
-All the data, all the information and all the pictures are my own.
-No, I do not market this stuff for sales (anymore…).
-The black line in the diagrams (mains) are actual reading from the grid. They are there as reference only.
-I still have to master the "Word to jpg" process. Some of the smaller "ripple" in the diagrams are pixel conversion errors, not actual reading errors. Mainly applies to the "mains" line, since that was not a solid colour to start with. However, look at the pictures, and you will get the picture...   

First, a simple chinese 2-stroke unit:



Something with electronics in it. Looks impressive, but with chinese durability at it's finest...   



Japanese unit with Honda engine. Looks promising, but at this price, forget it! 
Not everything with an inverter in it is good to be sold, especially not if the asking price is 80% of a genuine Honda unit...



First diesel is a Chinese high revving Yanmar clone. Max 1800W. My favorite for field service use. Lightweight and quiet (well, relatively...)



Second diesel is pretty much the same as the first, but 4200W, head with AVR.



Third diesel is a sound-proofed version of the one above. However, 400V unit (phase/phase voltage).




And finally, a proper generator from MeccAlte. I have the specification... Somewhere. Or I will go out in the garage and read off the head.

[Guy_Incognito]

You might want to noodle on that a bit more.   The speed variation, expressed as a percentage will be the same regardless of whether you use a 2 pole or 4 pole machine.  Since a 1% variation in speed causes a 1% variation in frequency in either a 2 pole or 4 pole machine, the frequency variation is the same in either case.

Of course! Forget I said anything. I hadn't really thought about it that much.

Wait, no, I'd still like to know the overall variation in engine speed under load if anyone can suss it out.

[Doug]

I can only speak for what I have seen and done myself. Small machines don't have enough iron, copper or slots to make a clean wave form and i don't get worked up over it.

Harmonics are a fact of life, most modern electronics generate harmonics.

That much said you do get what you pay for, but nothing is ever pefect....

Harmonics on my square wave Sendon UPS are bad, keeps the house from flooding when needed....

Doug

[Bikerbob]

Sure Doug, and that is probably the reason why the MeccAlte has a better waveform. Properly engineered, twice the weith of the others and a real AVR to match it.

However, Honda makes a big deal about the "pure" waves on their inverter units, and that makes people think that they really NEED a pure sine. That is the reason I made these tests to start with, to convince people that even the chinese are not as bad as you may think!

The ST heads may be a different story, but I still need to find out myself.
If you want it done properly, do it yourself! It may not be cost efficient, accurate or even reasonable, but still, you have no one else to blame, and for that reason, it is the RIGHT way to do it... 

/Bikerbob