inrush current

The component you are referring to is sometimes called an Inrush Current Suppressor. It is actually a low-ohmic NTC resistor,

RS Components in Australia have them as shown here

It is highly likely that RS in UK/Europe will also have them.

First the good news - after hours of searching I found the document, its a Shindengen MJ2400. Now for the bad news! - The chip is based on a modified bridge rectifier with integral thyristor to shunt the inrush limit resistor after start up. Sorry I made a mistake, this chip is only suitable for SMPSU circuits that have a mains bridge rectifier and smoothing cap input - it won't work with a transformer.

My hoard of data sheets has likewise been downloaded on an as needed basis, I usually get any other sheets on the page while I'm at it and often phone chip manufacturers and scrounge their current data book CD.

There are 2 common types of degauss thermistor, the most common of these has

2 PTC elements one of which is in series with the degauss coil the other is connected across the mains input - this is constantly heated to ensure that the degauss PTC proper is heated above its cut off temperature. This ensures there is no magnetic ripple on the picture in normal operation.

The less common type contains both NTC & PTC elements, the current drawn by the equipment flows through the NTC to keep the PTC element warm during normal operation.

Hours of searching? You sure are willing to invest time in helping people. You have my thanks, even though I can't use the chip.

Anyway, Instead, I'm gonna make a time delayed relay driver, which bypasses a block of resistors (which are in series with the primary coil) a short time after power up. I'll put a thermal fuse on the resistors to make sure it disconnects should the transformer develop a short. A normal fuse, slow blow, in series with the resistors, rated at lower value than what the resistors would use without the transformer in series with them might also be a good idea. If I time it right, I should be able to use a fast blow fuse as main fuse then (as noted by the PDF Ross Herbert pointed to). Experimention is in order.

An easy way to make a time delay is with a long tailed pair of transistors like a differential amplifier stage and add a high value cross coupling resistor so it acts as a Schmidt trigger and avoids relay chatter. Connect one base to a zener about half the voltage of the secondary rail used and the other base to a potential divider shunted by a time delay capacitor - don't forget a discharge diode from the capacitor to Vcc, this ensures that as Vcc falls below 1/2 value at switch off it discharges the cap. A relay driver transistor tapped off the collector resistor of the Schmidt output transistor completes the line up of only 3 transistors, be sure to remember also the back emf diode on the relay coil, the relay of course must provide adequate isolation between primary & secondary supplies!

Sorry, that url was not correct. This one appears to work (very long url watch wrap)

The RS Components UK url is

I have a circuit in use which is sensitive to relay chatter, so anything better is welcome.

My circuit design skills are somewhat limited (I really need to work on it), so I can't completely follow what you mean. Especially the LTP, I never really understood how they work. Could you draw up a schematic? You can post images to

should you be unable to send attachments. Thanks in advance.

On Thursday 13 April 2006 08:23, Ross Herbert wrote:>

This URL contains a session which now longer exists. But, I found it by clicking through from the first link you gave me.

Anyway, I have two of them in use now, but I'm gonna use a softstarter circuit, because the NTC's don't reduce resistance... I guess the idle current is too low.

The NTC will remain high resistance (if you call 5 or 10 ohms max high) during the first few cycles of the inrush current period at switch-on no matter what the load is. It's temperature will increase and the resistance will fall due to the high surge current but if the load on the DC supply is negligible or non existent then the NTC resistance will rise again to almost its nominal cold value. However, one would imagine that a 600VA transformer would normally have an idle load current of several hundred milliamps (depends on the application though) which would be sufficient to keep the NTC in its low resistance range providing you choose the right NTC. The fact that the idle current is low and the NTC resistance not at its minimum during low idle current is not really a problem since it has already done its job in protecting the fuse. When the normal load is subsequently applied the current drawn will definitely cause the NTC to go to its minimum value and that is what you want to happen.

If the NTC method isn't to your liking then you will have to use a delayed switch-on circuit which uses a relay contact to short out a resistor in the primary winding a couple of seconds after switch-on.

A typical delay circuit such as shown here

could be used where the relay contact will short circuit say a 10 ohm 10W resistor in the primary circuit of the transformer shortly after switch-on thus preventing the surge current from damaging anything.

A better alternative is to re-post on alt.binaries.schematics.electronic as I have sketched a basic circuit, but as I find math highly tedious I usually do just enough to get by and determine most of the values by experiment. There are some great engineers on a.b.s.e so you might get some even better circuits to try.

I'm going with this one:

If all goes right, the MOSFET switch should prevent chatter, because of it's fixed turn on voltage.