5 Things to Know Before Buying Low-Voltage Capacitor Bank

KG4MXV

Member

• May 9,

• #1

I got my hands on 4 banks of 7 300uf 350V AC mylar capacitors and would like to use them for power factor correct on my home/shop.

When ever my HVAC start up or when I use my arc welder in the shop it makes the lights flicker in the house.
And it also affects my neighbor that is on the same tap of the transformer.

My question is how to connect them to my service.
SInce I do not want to use bleeder resistors (waists energy) I am going to use a contactor to separate them from the mains in the event of power failure using a under voltage relay to open the contactor if any of the phases drops below 90V.

Now the real question is to connect the neutral of the cap bank to ground or to leave it isolated? Last edited: May 9,

Grossel

Well-Known Member

• May 9,

• #2

Hi.

A capacitor bank will probably not help for the flicker. Caps will help only if you have a problem with reactive currents and want to compensate for that.

What you need is is a rotating net stabilizer. An syncron motor with high inertia (or whatever you call huge rotation mass) on it's shaft. That way, the syncron motor can provide just enough power to prevent flash. Or at least minimize the flash effect.

That said, if you and your neighbour's houses is connected to the same transformer outpur I'd say it must be a poor transformer. So I think a better idea would be to talk about the issue to your local electrical provider.

Just for beinc curious, How many caps did you use to get that capacity?

tronitech

Active Member

• May 9,

• #3

I don't know were you are based but home and small businesses are rarely penalised for poor power factor which is totally unrelated to your problem.

The HVAC and arc welder are pulling very high currents at switch on and I'm not sure your capacitor banks will help that much? Last edited: May 9,

Reloadron

Well-Known Member

• May 9,

• #4

You do not use a bleeder resistor normally with AC mains power and capacitors used for power factor correction.

You do not mention the mains voltage? However, I suggest you

I have never seen power factor correction capacitor banks used for a residence. Generally for a residence, even a residence using the inductive loads you mention, it isn't all that practical. Also, I assume that the wire runs to the loads are adequate gauge wire? Meaning for example if you are in the US the wire gauge is in compliance with NEC code.

The link shows how it would be done for 240 VAC 60 Hz. power in the US. It also covers the formulas used.

Ron

KG4MXV

Member

• May 9,

• #5

Actually each bank has 6 300uf and 1 100uF for a total of uF
View attachment
I plan on paralleling two banks per phase for uf on each phase.
View attachment

Hi.

A capacitor bank will probably not help for the flicker. Caps will help only if you have a problem with reactive currents and want to compensate for that.

What you need is is a rotating net stabilizer. An syncron motor with high inertia (or whatever you call huge rotation mass) on it's shaft. That way, the syncron motor can provide just enough power to prevent flash. Or at least minimize the flash effect.

That said, if you and your neighbour's houses is connected to the same transformer outpur I'd say it must be a poor transformer. So I think a better idea would be to talk about the issue to your local electrical provider.

Just for beinc curious, How many caps did you use to get that capacity?

KG4MXV

Member

• May 9,

• #6

110V phase to N and 220 phase to phase.

KG4MXV

Member

• May 9,

• #7

100A service to my shop a 60 foot run 0awg for each phase, ground and neutral.
I have called my utility provider to check the transformer and they said all was ok.
I know it was over kill but I rather be safe than sorry.
Plus the wire was free. ( left over from another job.) Last edited: May 9,

Reloadron

Well-Known Member

• May 9,

• #8

If you want to place capacitors across the power coming in it is 240 VAC single phase or commonly called split phase in the US. It is actually single phase so there is no phase to phase. Anyway, if you want to place caps across it I see nothing wrong with doing it. For loads like AC or compressors running on 240 VAC the neutral is not used as a return. Not on my home system anyway. So yeah, place caps across the 240 VAC mains and between each leg of the split phase and neutral. The latter will only cover the 120 volt loads. While I do not see a gain it can't hurt I guess.

Ron

ccurtis

Well-Known Member

• May 9,

• #9

When ever my HVAC start up or when I use my arc welder in the shop it makes the lights flicker in the house.

I had this problem after my home A/C compressor was installed. It turned out that the connections where the wires from the pole transformer meet the lead-ins to the house were deteriorated. After repairing them, no flicker. That doesn't explain the issue with the neighbor though.

KG4MXV

Member

• May 9,

• #10

Well My trans former is a ground unit and not on a pole. so I can't see the connections.
I do have a infra red imager but I can't open the transformer can to shoot the connections.

I will have the utility do another inspection since the last one was over a year ago Last edited: May 9,

tcmtech

Banned

• May 10,

• #11

I plan on paralleling two banks per phase for uf on each phase.

I want to be there when you hook that up!

uf at 60 Hz will approximately equivalent to a .7 ohm short circuit per leg!

If light flicker from current draw is an issue this definitely wont help the situation! Odds are that high of pure capacitance load will put your main utility transformer way past its designed KVA capacity limits and burn it up in short order!

You need to do some serious reading and learning about AC power systems and how inductance and capacitance work before going any further.

JimB

Super Moderator

• May 10,

• #12

If my quick calculations are correct, 380 μF across a 120volt 60hz supply will result in a current of 170amps flowing.

Why would you want to do this?
Do you have 170amps of inductive current to balance out? Probably not.

The only result that I can see is that there will be extra load on the line and transformer giving more volt drop due to the resistive losses of the line and transformer.

I hope that you are not thinking that this capacitor bank will act in the same way as the reservoir capacitor on a DC power supply?
It will not.
There is no way that a capacitor on an AC power line will provide any buffer against varying loads.

JimB

On edit:
It look like TCM and I are in agreement here. Last edited: May 10,

KG4MXV

Member

• May 10,

• #14

Ok, I will conduct a small scale experiment tomorrow.
I will connect only one capacitor @ 300uF with a 10A breaker.
And use a 100A contactor to close the circuit. safely from a distance.

if you are correct it will pop the breaker and no harm done.
and I will abandon the project.

I will admit my lack of knowledge of AC power circuits and will
take the above advice. Last edited: May 10,

tcmtech

Banned

• May 10,

• #15

300 uf at 60 hz will be equivalent to about 8.8 ohms. At 120 volts it should pull about 13.6 amps.

If you have no real uses for them I would interested in buying them or trading for other things if the price is right.

Need any multi hundred amp 600 volt dual transistor blocks or - uf 350 VDC electrolytic capacitors? Last edited: May 10,

KG4MXV

Member

• May 10,

• #16

Then the question is what are you going to use them for.
I have 6 of the above banks but one of the capacitors has a large dent.
I have several IGBT's
But I Would love to get my hands an a decent sized UPS 10kva or larger minus the batteries (I have lots of batteries) To do this hack https://imgur.com/a/FJ3X9
Or a ATS switch.
Or a good weather station.
Name a price and I will see.
I will include the hardware. each post has a flat washer, split washer and 17mm nut.
The flat washer and nut are chrome plated brass.

Shipping is going to be kind of expensive.
I am in Raleigh NC.
Each bank weighs about 15 libs each.
I would have to put them on a pallet and freight it.



The above caps are from a retrofit kit for a 500KVA UPS that had shipping damage. (where the dent came from)_ Last edited: May 10,

tcmtech

Banned

• May 11,

• #18

I build rotary phase converters with them.

https://www.electro-tech-online.com/threads/3-phase-converter-schematic-miller-system./

I did some checking and I think they are worth $6 a piece plus shipping to me and I dont need all the big base frameworks. Just the capacitors plus connection hardware.

One or two big boxs UPS or FedEx would be good enough for me.

Sorry I dont have any big UPS units. I do have a bunch of non functioning Trace and Heart Freedom inverters though. The boards are bad in them and not worth my time to repair.

https://www.electro-tech-online.com/threads/tcmtech-is-cleaning-out-the-shop./

Just a thought.

KG4MXV

Member

• May 11,

• #19

I don't know where you got that quote.
but I did a search on the part number and this is the link,
https://www.findchips.com/avail?part=b-a-J030

at the quantity pricing of 25 they are $208.39 ea

Given that I am not greedy and I offer no warranty even though these are new and unused.
I will offer $20.00 each. that is less than 10% of the retail.
If you only want to buy a few that is ok.

I will take paypal.
at the moment I am willing to let go of 5 300uF and 1 100uF I will sell for $80.00 including shipping.
as a good will gesture. Last edited: May 11,

Here is my 64uF @ v capacitor bank. It has 3 options on top for firing: a cylindrical coil for can crushing, a pancake coil for ring launching, and two clips for attaching wires to be exploded (with blast shield). There is a switch on top to select the desired option. It is discharged using a high voltage relay that I added tungsten contacts to so they wouldn't ablate themselves away. There is a hv diode acting as a half wave rectifier for the charging circuit. There is also a 5 megaohm hv resistor across the caps acting as a bleeder for safety, and an Incandescent bulb in series with the charging primary acting as an indicator of level of charge. It crushes cans ok, and launches rings pretty well. It will also explode wires fairly well if they are small enough. Nice build, I like how the terminals on top kind of look like some lego bricks. 

The obvious solution for getting the state of charge is to just measure the voltage through a voltage divider (made from HV resistors).

You could also measure the capacitor voltage relative to the supply to get the difference between them.

If you don't want to deal with any high voltages you could also put a current shunt on the low side of the charching voltage.
This way you can measure the current which should go to 0 when the caps are fully charged.
Being in the lowside, you can simply hook up a multimeter directly across the shunt.

A more theoretical approach might also be to monitor the charging voltage and current and integrate the energy that has been transportet into the caps.

I hope this gives you some ideas on what might be interesting to try. 
Greetings,
Michael
I like the idea of the current measurement of the low end of the charging circuit. I guess that is what the bulb currently indicates (pun intended) but it would be great to have some numbers to go along with that. I did some YouTube learning about shunts and ammeters and how all that works. That'll be the next addition.

Another question: I currently have the half wave rectifier on the hv end of the charging circuit. Would it make any difference or be better if I put it on the low voltage end? And would the fact that it is half wave rectified with no smoothing cap affect the current reading?

On a different note, I'm trying to figure out more things to use this capacitor bank for. So far I've crushed cans, exploded wires, launched rings. What else could I use this for? Thanks! I built a pulsed power generator several years ago. It can store up to 24,000 joules. I use it to shrink coins, crush cans, and launch discs. I use a NST to charge the capacitor. To monitor the charge level, I use a microammeter with a 25 microamp scale. I place a 1 gigaohm HV resistor in series with the microammeter. In doing so, the meter now reads in kilovolts instead of microamps. I just watch the meter as I apply charge.

I use two meters. One meter is for monitoring the voltage level of the capacitor during charging. The other meter is for monitoring the capacitor voltage level during bleed-down. The bleed-down meter is bipolar to allow for the possibility of positive or negative residual charge on the capacitor due to voltage reversal. The charging meter is disconnected during firing and bleed-down. The bleed-down meter is only connected during bleed-down.

Quote from: DrewScott on December 20, , 07:26:29 PM
Another question: I currently have the half wave rectifier on the hv end of the charging circuit. Would it make any difference or be better if I put it on the low voltage end? And would the fact that it is half wave rectified with no smoothing cap affect the current reading?

It should not make any difference on which end you place your diode. The capacitor bank acts as its own smoothing capacitor anyway.
Just make sure to place the shunt on the capacitor side of the diode.

Quote from: DrewScott on December 20, , 07:26:29 PM
On a different note, I'm trying to figure out more things to use this capacitor bank for. So far I've crushed cans, exploded wires, launched rings. What else could I use this for? Thanks!

You could build a coilgun and shoot "normal" projectiles.
You could build a railgun, however the energy of your capacitorbank might be a little bit on the small side for that.
You could build a Ruby laser if you connect a flashtube to it.
You could do a lot of plasma physics, which would however require a vacuum chamber.


Greetings,
Michael

Quote from: DrewScott on December 21, , 01:46:19 PM
Could you expand on the plasma physics? I actually am building a vacuum chamber right now for a demo fusor. It isn't quite finished yet, but it should work once it's done.

Sounds great, how do you plan to get your hands on Deuterium, can you just buy it where you are?

Regarding the capacitorbank, I thought about accelerating and / or compressing plasma.
You can use an electromagnet to "compress" the plasma to reach incredible energy densities, however you will generate some pretty serious EMP emissions, so be careful.
Another thing is "pulsed nuclear fusion" which is some state of the art plasma physics.

With the help of a vacuum chamber, you might also be able to do some plasma coating and metal vapour coatings.

Quote from: MRMILSTAR on December 21, , 04:23:12 PM
Exploding fruit is another entertaining "use". Your capacitor bank may not be able to store enough energy though. Maybe a grape?

I like the idea of an exploding grape, it seems pretty silly but incredibly funny. 
Greetings,
Michael

Quote from: DrewScott on December 20, , 07:26:29 PM
I like the idea of the current measurement of the low end of the charging circuit. I guess that is what the bulb currently indicates (pun intended) but it would be great to have some numbers to go along with that. I did some YouTube learning about shunts and ammeters and how all that works. That'll be the next addition.

Another question: I currently have the half wave rectifier on the hv end of the charging circuit. Would it make any difference or be better if I put it on the low voltage end? And would the fact that it is half wave rectified with no smoothing cap affect the current reading?

On a different note, I'm trying to figure out more things to use this capacitor bank for. So far I've crushed cans, exploded wires, launched rings. What else could I use this for? Thanks!

Building a large flash lamp would be something I would try if I had such capacitors. Be careful - eye damage can occur both due to intense light/UV radiation as well as glass shrapnel in case the lamp explodes.

As for charging, I would use full-wave rectification instead of half-wave if possible. Is there a reason why you are using half-wave rectification? Multiplier circuit or something? I am using half-wave only for simplicity. You are correct, I should probably be using full-wave. I am thinking of switching to full-wave and moving it to the low voltage side of the charging circuit. The HV transformer would still operate since it would be pulsed DC after rectification.

On a different note, I was reading this page here: https://www.stevehv.4hv.org/MOCs.htm

His bank is just over 0.57kJ. Mine is over 0.7kJ and his seems to outperform mine substantially. I am wondering what I can do to increase the performance of mine? Should I add beefier conductors to the discharge side of mine? Is my work coil not as efficient? I am using 14AWG wire for my work coil at 6 turns. He uses magnet wire with more turns. What should I work on to try and improve it?

johnnyzoo, what kind of bulb should I use for the flash lamp? Looks good, Drew. 

On what kind of load is the competing machine outperforming yours?
Results depend on much more than stored energy, even if capacitance and voltage are identical.

Pulse timing depends on the circuit inductance, and there's probably some optimum value for each load and charge energy.  For my 52 uF can crusher I tried coils with 3, 4, 5, and 6 turns.  Also experimented with turn to turn spacing, since a longer coil induces voltage in a wider (lower-resistance)  shorted turn.

Losses from interconnect and switch resistance are probably relatively small,
compared to the effects of inductance in your capacitor bank and interconnecting wires.

Quote from: DrewScott on March 30, , 07:05:36 PM
I am using half-wave only for simplicity. You are correct, I should probably be using full-wave. I am thinking of switching to full-wave and moving it to the low voltage side of the charging circuit. The HV transformer would still operate since it would be pulsed DC after rectification.

On a different note, I was reading this page here: https://www.stevehv.4hv.org/MOCs.htm

His bank is just over 0.57kJ. Mine is over 0.7kJ and his seems to outperform mine substantially. I am wondering what I can do to increase the performance of mine? Should I add beefier conductors to the discharge side of mine? Is my work coil not as efficient? I am using 14AWG wire for my work coil at 6 turns. He uses magnet wire with more turns. What should I work on to try and improve it?

johnnyzoo, what kind of bulb should I use for the flash lamp?

Its been quite some years since I built a MOC bank as Steve's. It was only holding 0.33kJ since I did not charge them at higher voltage and the internal bleeding resistors would also quickly bleed off the higher voltage. https://kaizerpowerelectronics.dk/pulse-power/333-joule-moc-capacitor-bank/

I think his performance is either from lower ESR/ESL capacitors and that he seems to use "sacrificial" coils compared to the more sturdy ones we have in our setups. So his magnetic field is a lot tighter with those thin wires closer together. I never ripped a can apart either Maybe US cans are thinner than in Europe?