T
T.M.Haja Sahib
Guest
Sorry.I am not able to continue in this thread,because the OP is disturbed by my technical deliberations...........
Flatened top sine wave, does it mean anything?
- Thread starter iwire
- Start date
- Status
Sorry.I am not able to continue in this thread,because the OP is disturbed by my technical deliberations...........
- Location
- Massachusetts
Gar,
I sincerely appreciate the help you are offering me. Unfortunately a lot of it is well beyond me, there are things I know and then there are things I don't. Most of what you post I do not know. :ashamed1:
That said I do find the discussions between those that do know interesting and I hope you continue.
Really I was hoping someone would be able look at the snap shots I posted and be able to say 'This is caused by X and is the reason they have failures'.
I know that was a lot to hope for but you never know till you try.
Tomorrow I will pick up the recording meter and by Saturday I should be able to post very detailed information.
Last edited:
I don't know if this will help or confuse, but I'll give it a go anyway.Most of it but that is a problem with me and not your explanations.
This is a transient over voltage.
You can see the voltage overshoot and oscillate a few times before it more or less settles at around 900V.
This a rather messy but periodic waveform.
It is common for those of us who regularly deal with such thins to say it is full of harmonics. But in a way, that's a bit misleading. The waveform is what it is. It can be described mathematically as a series of sine waves at, in this case, 50Hz, 150Hz, 250Hz and other multiples of the periodic supply frequency. Harmonics. It's a useful tool for analysing how the nasty waveform will react with circuit components and what impact it might have on the supply.
- Location
- Massachusetts
It just dawned on me that Gar and others might be interested in what meter is being used.
This is it here. http://dranetz.com/portable/powerxplorer-px5
This is it here. http://dranetz.com/portable/powerxplorer-px5
- Location
- Massachusetts
Pictures help me a lot, thank you.
have used Dranetz analysers when they were a bit less portable. I always thought it was good kit if a tad expensive.
These days I use a digital storage oscilloscope with the means of downloading data to a PC and drop it into a spreadsheet.
Then I can analyse the data for a good many things.
And make pretty presentations for customers.
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
120119-1120 EST
iwire:
Later I will try to illustrate what I have described with the RC discussion with some plots.
My guess is that your failure problem is not a result of the flattened top of the sine wave. I think I indicated this earlier, but i could be wrong on this assumption that the flat top is not causing the failures.
Hopefully your monitoring will provide some new useful information.
If there is any way to see what is failing in the ballasts that would be very useful information. Has the diode bridge shorted, has the filter capacitor shorted or opened, has a copper trace between the bridge output and the filter capacitor burned up, is there an MOV at the input that failed?
.
iwire:
Later I will try to illustrate what I have described with the RC discussion with some plots.
My guess is that your failure problem is not a result of the flattened top of the sine wave. I think I indicated this earlier, but i could be wrong on this assumption that the flat top is not causing the failures.
Hopefully your monitoring will provide some new useful information.
If there is any way to see what is failing in the ballasts that would be very useful information. Has the diode bridge shorted, has the filter capacitor shorted or opened, has a copper trace between the bridge output and the filter capacitor burned up, is there an MOV at the input that failed?
.
As far as capacitor failures that I have seen, they blow due to overheating. Ripples in the voltage will always heat the capacitor a bit (capacitor acts to prevent the voltage to change). If the ripples existing in the supply gets worse, you should expect lots of capacitors to blow/ get damaged. IDK if you see it that way but that's how I think happens with high harmonic content in the voltage is present.
- Location
- Massachusetts
Appreciated, sir.
As you may have gathered, I'm somewhat into the theory side of such things. As is gar.
But then they are topics that need that kind of approach if you have to look at them in depth.
That also makes it difficult to convey the ideas in reasonably intelligible form to those not so involved in the theoretical side.
I do admire is your openness in accepting that there are areas that you don't know about. That's big. And a refreshing approach that others here could learn from asking rather than claiming to know everything about everything. The beginning of understanding is the acknowledgement etc....
For me, the boot is most often on the other foot here and, as an ignorant Brit, I have to ask about fairly common terms familiar to everyone else. Ufer as an example. Raceways. Conductor sizes in real world units. NM wire. Stuff.
Others, yourself included, have had the patience to explain. When, and if, I can reciprocate in some small way I try.
wirenut1980
Senior Member
- Location
- Plainfield, IN
I just had a thought. I wonder if the flat tops are causing the DC bus to get low voltage. Do flourescent lights have a DC bus? What do they have that converts the AC-DC? They have third harmonics, so is it like a switch mode power supply? I wouldn't normally think this could cause a problem since the AC input range usually pretty large on flourescents, but the thought got me thinking.
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
120119-2131 EST
wirenut1980:
In post #31 I referenced the following location:
I found the following chart p 5 of 15 at http://www.southalabama.edu/engineering/ece/faculty/akhan/Courses/EE334-Fall07/EE 334-chapter-2.pdf
labeled Power Supply Circuits, Filters and Regulators, top right on p 5. This is for a half wave rectifier, but the current pulse is generally the same.
This has an idealized current pulse shown. In the real world the sharp points, and steep initial rise will be somewhat rounded, and reduced.
Note the shape of the current waveform. When you add a small amount of series resistance and inductance, then the current pulse shape changes and ends somewhat after the source voltage peak. The leading edge won't be so vertical and sharp peaked.
With a flat topped input voltage the peak capacitor voltage will be somewhat less. But for iwire's waveform this won't be much. Project where the sine wave peak should be and take the difference between the ideal waveform and the actual, and that amount is about the amount lower the peak capacitor voltage would be.
Whether the flat topped waveform iwire showed in post #1 or a true sinewave is the input, the current pulse to the capacitor will be about the same, and a well designed device will have a capacitor that can tolerate the repetitive energy from this pulse.
.
wirenut1980:
In post #31 I referenced the following location:
I found the following chart p 5 of 15 at http://www.southalabama.edu/engineering/ece/faculty/akhan/Courses/EE334-Fall07/EE 334-chapter-2.pdf
labeled Power Supply Circuits, Filters and Regulators, top right on p 5. This is for a half wave rectifier, but the current pulse is generally the same.
This has an idealized current pulse shown. In the real world the sharp points, and steep initial rise will be somewhat rounded, and reduced.
Note the shape of the current waveform. When you add a small amount of series resistance and inductance, then the current pulse shape changes and ends somewhat after the source voltage peak. The leading edge won't be so vertical and sharp peaked.
With a flat topped input voltage the peak capacitor voltage will be somewhat less. But for iwire's waveform this won't be much. Project where the sine wave peak should be and take the difference between the ideal waveform and the actual, and that amount is about the amount lower the peak capacitor voltage would be.
Whether the flat topped waveform iwire showed in post #1 or a true sinewave is the input, the current pulse to the capacitor will be about the same, and a well designed device will have a capacitor that can tolerate the repetitive energy from this pulse.
.
I feel quite sure that your erudite, instructive, constructive, and informative contributions demonstrating the breadth and depth of your profound knowledge, prowess, and practical experience on these matters will be sorely missed here.
Not sure you'd call it a DC bus. CFLs have an input rectifier - as do may other pieces of electronic kit. Gar has kindly provided a link to such input circuits.
Joethemechanic
Senior Member
- Location
- Hazleton Pa
- Occupation
- Electro-Mechanical Technician. Industrial machinery
PetrosA
Senior Member
- Location
- Lancaster, Pennsylvania
I also would like to thank you guys for this thread and discussion, especially iwire, Gar and Besoeker. It's well over my head mathematically (and equipment-wise as well ), but I try to understand what I can. I do some simple beta testing of multimeters and software for one of the large manufacturers, and threads like this present concepts that may be useful for me at some point, however remotely related to what I'm doing. One of my dreams would be to have a chance to work alongside an expert for a week or ten, but that's unlikely
), but I try to understand what I can. I do some simple beta testing of multimeters and software for one of the large manufacturers, and threads like this present concepts that may be useful for me at some point, however remotely related to what I'm doing. One of my dreams would be to have a chance to work alongside an expert for a week or ten, but that's unlikely Kind words, thank you.
However, if we haven't made our contributions comprehensible to others in the field, that's our failure. Not yours.
That's not to say they are easy topics But, if we are well versed in our respective fields, we ought to be able to present it in such a way that others without specific knowledge in those areas can gain an understanding of it.
Maybe a mainly text base forum isn't the easiest means of conveying this information and none of us is a mind reader. We don't know how much another poster knows and not one of us knows everything. The one piece of advice I'd offer is that, if you don't know something and want to know, just ask. There is a huge base of knowledge here on this forum. Tap into it.
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
120120-1501 EST
Do ask questions, and when I make a mistake let me know.
A few important points:
Equations are not the law, they are simply a mathematical model of a simplified electrical circuit, and therefore only represent an approximation to the real world system. But they can be quite good.
In the real world do not assume something when an incorrect assumption could cause a major problem, and you could do better than make that assumption. For example: the radar information at Pearl Harbor about 7 AM on 7 Dec 1941. An assumption was made that they were our bombers even though the location was probably incorrect.
Sometimes in a study you need to make assumptions, then test the assumptions for validity.
Try to figure out how to ask the correct question when you have no idea what is the correct question.
Use statistics as a last resort to solve a problem, but use statistics to judge where to look for the problem.
Suppose you had a simple 5 tube radio and it does not work. Would you try to imagine all possible combinations of failures that might cause a failure? Or would you open it and make some visual and likely electrical measurements? Do the filaments light? Is there B+ voltage? Can you touch various points and get some speaker noise? Etc.
.
Do ask questions, and when I make a mistake let me know.
A few important points:
Equations are not the law, they are simply a mathematical model of a simplified electrical circuit, and therefore only represent an approximation to the real world system. But they can be quite good.
In the real world do not assume something when an incorrect assumption could cause a major problem, and you could do better than make that assumption. For example: the radar information at Pearl Harbor about 7 AM on 7 Dec 1941. An assumption was made that they were our bombers even though the location was probably incorrect.
Sometimes in a study you need to make assumptions, then test the assumptions for validity.
Try to figure out how to ask the correct question when you have no idea what is the correct question.
Use statistics as a last resort to solve a problem, but use statistics to judge where to look for the problem.
Suppose you had a simple 5 tube radio and it does not work. Would you try to imagine all possible combinations of failures that might cause a failure? Or would you open it and make some visual and likely electrical measurements? Do the filaments light? Is there B+ voltage? Can you touch various points and get some speaker noise? Etc.
.
Joethemechanic
Senior Member
- Location
- Hazleton Pa
- Occupation
- Electro-Mechanical Technician. Industrial machinery
- Status