At the Location where I have 250Kvar fixed bank, I found the internal fuse blown, bad resistor on 2 of the 5 50Kvar sections and TWO of the 600 amp fuses blown. None of the Capacitors test bad. The 600 amp fuse is located in the Main disconnect for the ENTIRE unit, it feeds directly from the Main 480V buss.
Failed Capacitor/resistors
- Thread starter raiderUM
- Start date
- Status
At the Location where I have 250Kvar fixed bank, I found the internal fuse blown, bad resistor on 2 of the 5 50Kvar sections and TWO of the 600 amp fuses blown. None of the Capacitors test bad. The 600 amp fuse is located in the Main disconnect for the ENTIRE unit, it feeds directly from the Main 480V buss.
I have no idea... it doesnt make any sense. I found another location where nothing was wrong with the capacitors, no bad resistor, no blown internal fuses, nothing wrong. This location was feed from a bucket in the Main gear, I had Two of the Three 200 amp fuses blown. After testing the capacitor I replaced the fuses feeding the Cap bank. They are still working to this day! So I dont know what is going on.
On a side note: When fusing Kvar banks is this correct? Kvar/480*1.732=amps ; Amps*135%=fuses needed for bank???
Not inside the capacitor can.
I have a notion that the Maharajah thinks there is a fusible element internal to the capacitor. Even if there was, it wouldn't be 100A for a 40uF capacitor.
As a suggestion, maybe look at the quality of the supply voltage. Over the years there has been a huge increase in the proliferation of non-linear equipment connected to the supply.
In short, harmonic voltage distortion may be the problem or at least part of it.
Eaton has a reasonable technical note on this. Technical Data SA02607001E.
I agree with the calculation of current (assuming a clean sine wave) but don't for get to factor in the 1,000 for the k in kVAr.
As far as the fuse rating goes, we'd usually go for a factor of about 1.7 times the nominal current.
The Eaton document referred to above gives 100A for your 50kVAr - nominal current is 60A - so a factor of 1.67.
The 150kVAr fuse is 300A
newservice
Senior Member
old out of tolerence 150k ohm half-watt resistor blew because old electrolytic capacitors all dried out? just a guess
All capacitors were checked and found to be good. See post #30.
SG-1
Senior Member
- Location
- Ware Shoals, South Carolina
If the capacitors & resistors are 30+ years old, I would expect they are near the end of their life cycle.
The capacitors test ok with a meter, but would they if they were charged to full voltage, then discharged, while measuring the time it takes ?
The capacitors test ok with a meter, but would they if they were charged to full voltage, then discharged, while measuring the time it takes ?
electroman00
Member
- Location
- Orlando FL
Those are 180K... 5 watt carbon resistors based on the referenced terminal size they are connected to.
They are bleed resistors and simply failed due to age and usage.
Their primary purpose is to discharge caps after power removal (safety).
Since there are two in parallel per bank, the circuit resistance is approx. 90K.
You can use 220K which will take a little longer to discharge but will last a little longer.
Parallel circuit resistance will be approx 110K.
I would not use wire wound or carbon film, as those types will act as inductors in AC circuits.
As far as the 5% or 10% goes...this circuit is not what we call resistance critical by any means.
Pretty much anything goes in this application.
IOW what is the desired bleed time, 30sec-1min...no big deal...we just need to get them discharged today.
If you need those resistors I have them in stock, just let me know.
They are bleed resistors and simply failed due to age and usage.
Their primary purpose is to discharge caps after power removal (safety).
Since there are two in parallel per bank, the circuit resistance is approx. 90K.
You can use 220K which will take a little longer to discharge but will last a little longer.
Parallel circuit resistance will be approx 110K.
I would not use wire wound or carbon film, as those types will act as inductors in AC circuits.
As far as the 5% or 10% goes...this circuit is not what we call resistance critical by any means.
Pretty much anything goes in this application.
IOW what is the desired bleed time, 30sec-1min...no big deal...we just need to get them discharged today.
If you need those resistors I have them in stock, just let me know.
petersonra
Senior Member
- Location
- Northern illinois
- Occupation
- engineer
Yes, I am very aged..... And blessed and cursed with a fairly good memory for particularly for numbers. And a lot of it of no use to me whatsoever. I know an acre is 4840 square yards and that there are eight furlongs in a mile and stuff such as that I'll never use. Is it occupying brain capacity that could be put to better use I sometimes wonder.....
Back on topic - I remember two sayings for the colour code. One is far too coarse to be permitted here.
But I don't actually use them. I just know or somehow remember what the colours are. Then the resistors are specific values in what we usually encounter as either the E12 or E24 series.....10, 12, 15, 18, 22, 27, 33 for example. You get to recognise patterns and can identify a value without going through the band, band multiplier thought process. At least that's how it seems to me.
Looks like silver - 10%.
I remember Avagadro's number for some reason. 6.023X10^23
Not to mention 1.6*10-19
Keeping it electrical.
I do remember some useful things too.
Like the conversion factors from metric (SI) to Imperial for weights, measures, capacities and some useful approximations.
One metre, for example is about a yard or a yard plus 10% if you want to be a bit more accurate or better is 39.37 inches.
One kg is about two pounds in weight plus 10% or 2.205 pounds.
We use pretty much all metric in the electrical business.
Conductors are given by their cross-sectional area.
For example, 2.5mm2 is 2.5mm2. And 4mm2 is thicker than 2.5mm2
I can convert to approximate AWG sizes - there are no exact equivalents - but that I can't do in my head.
ptonsparky
Tom
- Occupation
- EC - retired
I think Rick is the closest here. The bleed resisitor failed prior to the blowing of the fuse. With no bleeder resistor to drain the current from that bank of capacitors, it remained charged when disconnected. Then when power is restored, at a very precise cycle of time, you essentialy have a short circuit, blowing the fuses.
I don't think so.
Each bank of capacitors had two 180k resistors and, as far as we know, one of them failed so there would still be a discharge path through the remaining bleed resistor.
And essentially you have a short circuit anyway on restoration of power unless the supply voltage and capacitor voltage are equal at that time. Long odds against that regardless of the state of charge on the capacitor. So there will almost always be a transient. If you happened to be unlucky and, at the instant of restoration, the voltages were opposite polarity the transient would be greatest but still a transient of short duration. I've checked the time current curves of a few 100A BS88 fuses and they can take typically 2,000-3,000A for 0.1 seconds. About six cycles of the 60Hz supply.
kwired
Electron manager
- Location
- NE Nebraska
Harmonics in the supply can and will cause what appears to be short circuit conditions across the capacitor causing the fuse to blow. Same conditions will apply excess voltage across the bleed resistor eventually leading to failure of the resistor.
I like the two ideas I have quoted in this post as being great possibilities of what has happened to the OP.
Add: Well after Bes' last reply maybe the harmonic issue climbs even higher on the possibility list.
ptonsparky
Tom
- Occupation
- EC - retired
Increased use of NL loads causes failure of the Caps or the fuses? In this case the fuses blew. All Caps good.
PFC causes the voltage to rise, increased voltage causes more current to flow in the caps. Would they be better off reducing the PFC to match the needs?
PFC causes the voltage to rise, increased voltage causes more current to flow in the caps. Would they be better off reducing the PFC to match the needs?
ptonsparky
Tom
- Occupation
- EC - retired
No, the resistors would have to be in series with the caps. The do make something similar to what you are talking about it is not the case here. These are bleeder resistors.
kwired
Electron manager
- Location
- NE Nebraska
You are going to either make me do some reasearch or hopefully someone with more harmonics experience will step in and explain exactly what happens. It has been a while since I experienced this problem and don't recall off the top of my head exactly what happens, but if harmonic levels are high enough you get additional current in capacitors - enough that it will take out fuses. Probably been 20 years ago the EC I was working for was putting in PFC's in a plant. We were putting individual PFC's at each motor starter. Spent time mounting units, running raceways, in some cases where larger group of starters maybe even aux gutters to route conductors through to caps. One day production was down for cleaning and / or maintenance so we had the opportunity to make final connections to motor starters on quite a bit of production loads that otherwise ran 24/7 when production was going on. Come back next day and had blown fuse indicators lit up on almost every single capacitor that was in use. I was not directly involved in reasearch of finding the reason for this but it was eventually determined that the harmonics from a 250 Hp VFD was the reason for this. Even was blowing fuses on far side of plant away from where that drive was located, but was all fed from same POCO transformer was all that mattered. We ended up disconnecting all the capacitors and installing larger capacitor banks, with what appears to be line reactors in series with them - supposably these helped take care of the harmonics issue, I was pretty green in the trade at time so I didn't fully understand everything at the time but tried to absorb what information I could.I really have not run into such a situation since then so it is only experience I have to go with as well as other things I have either read about or was involved with in a more second handed manner.
I agree with all of that. Seen it. Done it.
We always fit in line chokes with power factor correction capacitors. It's common here to tune to 225Hz, just below 5th harmonic.
kwired
Electron manager
- Location
- NE Nebraska
Can you possibly explain to us exactly what happens in an instance like this that causes the fuses to essentially see a short circuit. Might not really be a short circuit but I remember it being explained as a very brief overcurrent situation through only part of each cycle and to the fuse is kind of no different than seeing a short circuit. It has been a while since I have had anything to do with something like this and I suspect it is something you are more familiar with.
For circuits where the voltage isn't a pure sine wave it can be analysed as a series of sine waves that are multiples of the supply (or fundamental) frequency which is 60Hz for you.
These multiples of the supply frequency are referred to as harmonics. Typically for a three phase VSD they are at (6n ? 1) times the fundamental frequency where n represents the counting numbers 1, 2, 3, etc
so you get components at 300Hz, 420Hz, 660Hz, 780Hz etc as well as the 60Hz component.
No doubt you know that the impedance of a capacitor goes down with increasing frequency so the higher frequency components can add significantly/considerably to the normal 60 Hz component. This increased current can, and sometimes does, blow fuses.
- Status