# Inrush current limiting scenario

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.

#### rhythmsandy

I have a situation for the soft start where Im using dual relay and three 56ohms 5w ceramic resistor in parallel and these resistors are in parallel with second relay.

Transformer 800VA with large core. Primary 230V, secondary 38-0-38, Total capacitors 20000uF per rail.

When the power button turns on it first triggers the first relay which will send the power through the three 56ohm resistors. so for the duration of 1sec when turned on the actual connection is through the three 56ohm resistors in parallel.
After 1 sec the second relay triggers and shorts the resistors.

Now the problem is that i still see the lights dimming in the house when the soft start turns on so totally two times.

Now what I feel is that the current circuit is not controlling the inrush current and hence planning to use one in rush NTC thermistor inrush current limiter in series with the primary.

Example http://www.mouser.com/ds/2/600/SL3210015-237978.pdf
Ametherm 10ohm 15A

Are there any suggestions that you can tell that what is my current problem? Is the resistance too low like 56/3 = 18.6ohm?
So adding a NTC 10ohm 15A will solve my current problem as I dont want to see that inrush dimming my lights even little.

P

#### JonSnell Electronic

The use of an NTC instead of the 56R is a good idea. Don't let the contacts close until your secondary DC voltages reach at least 60% of their normal voltage. Try using smaller capacitors, that will further reduce the charge current. Charging 20.000uF with 54V takes a lot of Joules!

#### AndrewT

I'm charging 90mF to 50.5Vdc when I turn on a 2channel 100W into 8ohms amplifier.

I don't see lights dimming.
I can't even measure a drop in mains voltage at the adjacent socket outlet.
And I do use a soft start with resistors timed to bypass in <300ms.

Based on my 240Vac experience, I would use 8r to 10r as a soft start resistor on a 800VA 110/120Vac mains transformer and fit an T8A close rated fuse. and switch on the timed bypass in ~200ms.
One could use a 10r Power Thermistor and delay the timer to more than 1s.

Last edited:

#### DF96

Lamps dimming on 12A inrush suggests a fault in your house wiring. Maybe a loose screw somewhere?

#### AndrewT

But it's not 12Aac

He has three 56r in parallel giving 18r7
On a 115Vac supply that limits the initial start up current to ~6.1Aac.

Either he has wired it all wrong, or his house wiring is faulty. Either way there is a fault that needs a competent person to find and sort.

#### Mooly

Paid Member
I don't see lights dimming.

I think the light type (incandescent, LED, CFL etc) makes a big difference in how subjectively noticeable any volt drop is.

Connecting a 4.7 ohm 12000 watt resistor across a nominal 240vac supply certainly produces a noticeable dimming here, but only pulls the supply down by around 6 volts

The degaussing circuit of our Sony TV also seriously dims any (incandescent) light momentarily and yet is only on a T4A (I think) internal fuse.

#### AndrewT

rhythmsandy said:

AndrewT said:

I don't know why I thought he was on a 110/120Vac supply.
My mistake.
And for consistency I did the same in post3.

#### AndrewT

I'll change my recommendation to suit 230Vac mains supply.

800VA needs a T3.1A, or T4A, mains fuse.

That will not blow for an initial start up current of 8Aac.
Use a 27r, or 30r, to current limit and bypass that in ~200ms.
Don't use a CL60 (10ohms), nor a series pair of CL60 (20ohms).

#### rickmcinnis

AndrewT,

I would like to know why you suggested to rythymsandy not to use the thermistors? I know I should be able to figure it out but I can't.

You have in other threads I have read been a proponent for using these as long as they are switched out of the circuit. I thought that made sense so is this recommendation due to the UK mains voltage?

I am in the US.

Another thing I would like to know: is using a relay necessary? Could one not use two switches; using one to take the soft start stuff out of the circuit? AND if one does this is the timing critical? If one left the soft start in the circuit too long (say a second or two) is there any harm to the circuit? My assumption is heating of the soft start resistors depending upon how generous the power rating is. If one goes overboard it seems there should not be any concerns but I would like to know what you think.

Thanks,

#### AndrewT

AndrewT,

I would like to know why you suggested to rythymsandy not to use the thermistors? I know I should be able to figure it out but I can't.
You have in other threads I have read been a proponent for using these as long as they are switched out of the circuit.
The best in my view is to use Power Thermistors and to switch them out (bypass after a delay) to return the transformer to seeing a low impedance source.
I thought that made sense so is this recommendation due to the UK mains voltage?

I am in the US.

Another thing I would like to know: is using a relay necessary? Could one not use two switches; using one to take the soft start stuff out of the circuit? AND if one does this is the timing critical? If one left the soft start in the circuit too long (say a second or two) is there any harm to the circuit? My assumption is heating of the soft start resistors depending upon how generous the power rating is. If one goes overboard it seems there should not be any concerns but I would like to know what you think.

Thanks,
One could easily use two switches. That is what a switch+relay does, or a pair of cascaded relays does.
I would not place two manual switches on the front panel. Confusion? which first? how long?
What if the delayed switch was not switched?

Last edited:

#### CharlieLaub

Paid Member
I have a situation for the soft start where Im using dual relay and three 56ohms 5w ceramic resistor in parallel and these resistors are in parallel with second relay.

Transformer 800VA with large core. Primary 230V, secondary 38-0-38, Total capacitors 20000uF per rail.

When the power button turns on it first triggers the first relay which will send the power through the three 56ohm resistors. so for the duration of 1sec when turned on the actual connection is through the three 56ohm resistors in parallel.
After 1 sec the second relay triggers and shorts the resistors.

Now the problem is that i still see the lights dimming in the house when the soft start turns on so totally two times.

Now what I feel is that the current circuit is not controlling the inrush current and hence planning to use one in rush NTC thermistor inrush current limiter in series with the primary.

Example http://www.mouser.com/ds/2/600/SL3210015-237978.pdf
Ametherm 10ohm 15A

Are there any suggestions that you can tell that what is my current problem? Is the resistance too low like 56/3 = 18.6ohm?
So adding a NTC 10ohm 15A will solve my current problem as I dont want to see that inrush dimming my lights even little.

P

In the USA wiring is typically connected (in the walls) using twisted connections of solid core wire made using wire nuts. After many years these can become slightly oxidized and offer some contact resistance. If you are at the end of the branch there might be several connections like this before the line reached the wall outlet. If the home was built before 1975 or 1980 or so, there could be smaller gauge wiring in the walls (like 14 gauge) if the builder assumed only a few lights would be connected to the branch. Both of these can cause voltage drops in the walls when lots of current is flowing, like at start up, leading to "lights dimming" type phenomena.

Unless the breaker is tripping or there are other problems caused by the short term high current draw I would not worry about it. You could hire an electrician to check the wiring and/or install a dedicated circuit leading directly from the main panel to the wall outlet. This can even be done temporarily by running Romex thru the interior of the home to check. Also, since nothing besides the stereo system would be connected to that branch it would not dim any other circuit (unless you REALLY have some lame power delivery to your home!!!).

#### rhythmsandy

Primary: 230V
i think what Andrew says makes sense as using a NTC in series even with primary reduces the strong transient current delivery. The actual problem is just startup inrush so using till then and then shorting out the NTC is actually a better idea.

Im currently using 5ohm resistor and now I see reduced dimming of lights
Added a series NTC to the primary and now the dimming is almost gone.
But i havent tried using NTC of 10ohm instead of the resistors in the soft start would try that and let know.

#### nigelwright7557

I use a 8 pin PIC micro, opto isolator and a triac to control the mains phase angle on power up.
It works very well and no relay contacts to burn.
The only down side is the triac needs quite a big heat sink for high power amps.

#### AndrewT

Primary: 230V
i think what Andrew says makes sense as using a NTC in series even with primary reduces the strong transient current delivery. The actual problem is just startup inrush so using till then and then shorting out the NTC is actually a better idea.

Im currently using 5ohm resistor and now I see reduced dimming of lights
Added a series NTC to the primary and now the dimming is almost gone.
But i havent tried using NTC of 10ohm instead of the resistors in the soft start would try that and let know.
for the conditions you supplied, I showed that a 27r to 30r added resistor would prevent a close rated fuse blowing at start up on a 230Vac mains supply.
5r and 10r are way short of my suggested values.

Assume for this example your primary resistance is 1r0.
The total resistance seen by the mains at start up (after the current pulse has saturated the transformer core) will be 28r.
The AC current peak during this first half cycle will be ~ 230Vac/28r = ~8.2Aac
I have found that a close rated fuse of half that value does not rupture when it is subjected to repeated cold starts.
So In my view a T4A fuse will work. That allows for ~920VA, or any 800VA to 1000VA 220/240Vac mains transformer.

If you are on a 220Vac supply and use a 30r added resistance, you may find that a T3.1A mains fuse will survive repeated cold starting of an 800VA mains transformer.

Last edited:

#### rhythmsandy

for the conditions you supplied, I showed that a 27r to 30r added resistor would prevent a close rated fuse blowing at start up on a 230Vac mains supply.
5r and 10r are way short of my suggested values.

Assume for this example your primary resistance is 1r0.
The total resistance seen by the mains at start up (after the current pulse has saturated the transformer core) will be 28r.
The AC current peak during this first half cycle will be ~ 230Vac/28r = ~8.2Aac
I have found that a close rated fuse of half that value does not rupture when it is subjected to repeated cold starts.
So In my view a T4A fuse will work. That allows for ~920VA, or any 800VA to 1000VA 220/240Vac mains transformer.

If you are on a 220Vac supply and use a 30r added resistance, you may find that a T3.1A mains fuse will survive repeated cold starting of an 800VA mains transformer.

yes what you said is right but the problem is observed when the second relay is triggered.

The first relay is just to provide the contact for main trafo. Second relay has the resistor and that resistor value. Initial usage is about 56ohm three resistors in parallel so 18.6ohm but the step i mean the dimming of light is quite observable but when I changed it to 5ohm the dimming is reduced dramatically. When used a series NTC with the soft start the dimming almost gone.

Now i want to try the NTC in place of resistor as it would eliminate the requirement of the resistor and also suppresses the inrush what do you say?

#### trobbins

The first check when using an NTC is that its Joule rating exceeds the capacitor charging requirement in your circuit.

Your application needs an energy throughput of ½C(√2.VAC)2 = 29J for 38VAC and 0.02F. You need some margin above that for mains voltage and capacitance tolerances, and for transformer inrush.

In the CL range, a single CL50 to CL90 can allow up to 72 joule throughput for reliable repeated operation. Further checking of max continuous current and temperature derating are needed to choose a suitable NTC.
http://www.farnell.com/datasheets/85914.pdf

Using just an NTC (no relays etc) on the mains input offers excellent simplicity. You could use PSUD2 to look further in to how the extra NTC series resistance could possibly affect any amplifier operation - but I'd be suggesting it's effect would be negligible for most applications.

#### Michael Chua

I use a 8 pin PIC micro, opto isolator and a triac to control the mains phase angle on power up.
It works very well and no relay contacts to burn.
The only down side is the triac needs quite a big heat sink for high power amps.

You may want to check out the MOC3040. Very effective. Zero current surge.

#### nigelwright7557

You may want to check out the MOC3040. Very effective. Zero current surge.

I used the MOC3020.
I simply used a 390K resistor into a PIC I/o pin to detect zero crossing.

#### Michael Chua

Do you need to program the PIC micro?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.