Looking for a NEC 3P6MH thyrister?

[danrogers]

Hi all, I need to get hold of a NEC 3P6MH but can't seem to find one around?

I have found a datasheet but that is about all I can come up with?

Thanks in advance!

 

[vtech]

Probably available on ebay but it is not anything special. Should be able to use any general purpose SCR 3A ~700v TO202 package substitute.

 

[danrogers]

I did check ebay and could find one, but it is very expensive at £6.25, I was hoping to find one on RS or farnell but not having much luck. Both of the aforementioned suppliers don't seem to do any in a T0202 package?

Thanks

 

[Mr RB]

Or just use a TO-220 package and check the pinout.

 

[danrogers]

https://uk.rs-online.com/web/p/thyristors/6562548/ this one seems fairly close, but I am unable to match up some of the ratings because of the differning terminology between the two datasheets.

I don't want to get it wrong because I want it to be reliable, could anyone lend some knowledge?

Thanks

 

[Mr RB]

Good substitution would require knowing what the board does, and preferably a decent photo of the whole board.

Substituting just based on some datasheet specs is iffy. It's quite likely that once we know what it does then it might be perfectly happy with a device that has higher specs and will be more reliable. Also, do you know why the original part failed?

 

[danrogers]

The board is from some GHD hair straighteners. I'm not 100% but I think this PCB senses current to regulate the heat, switched by the thyrister. Hope the photos help?

many thanks

 

[Mr RB]

I repaired a temp controlled soldering iron that has very similar control electronics. The opamp IC on mine was Lm358 (your opamp is a 2903) and did the zero cross detect and temperature setpoint comparison.

It's a simple phase angle controller so the type of TRIAC or SCR does not matter that much. Mine was an AC triac. Yours seems be half-wave only, as it runs the power through big diode D9 before the thyristor.

You should have no problems using a thyristor of higher rated current and voltage, provided it has equally sensitive gate current Ilatch (as the opamp will have a limit to how much gate current it will provide).

Also you didn't say as to the cause of failure, but that thyristor looks very much like it has seen massive current overload, ie a short circuit on the heater wires. It would be good insurance to put a fuse in series with the heater wires in case this might happen again.

 

[danrogers]

Thanks Roman

I am just looking over the datasheets again (original part and the part I linked from RS) the gate current seems to be 0.2mA on the original part and 5uA on the RS part which is quite different?

the more I look at datasheets the more confused I'm getting

 

[Mr RB]

Something is wrong with that 5uA!

There should be a latching current listed or gate trigger current, normally it can be high as 30mA for a larger device and down to about 0.2mA or so for the smaller devices.

If your original part has Ilatch specced at 0.2mA you need a "sensitive gate SCR". It should be easy enough to find something. I would look for any larger part ie >=800v and >=5A that has a sensitive gate under 1mA. It may not need to be as low as 0.2mA.

 

[danrogers]

I'm struggling to find one, they all seem to have far higher gate currents?

The RS one I linked could be ok, with the low gate current surely?

Many thanks for the help!

 

[Mr RB]

Sorry that RS web page crashes my browser.

Like I said you may not need one as low as 0.2mA, as long as it is a "sensitive gate SCR" you should probably be ok.

I just checked a C106D, possibly the most common small SCR ever made; 400v, 2.5A av, 4A RMS, 0.17mA Ilatch. It's pin compatible.

The C106M is same part but rated for 600v. (https://www.electro-tech-online.com/custompdfs/2013/06/C106_DPDF.pdf)

 

[danrogers]

Thanks very much Roman. Great help. Do you think I really need a part rated to 800v? Surely with a 240V supply the board will never have a voltage that large on it?

Thanks again

 

[Mr RB]

What was the original part's voltage spec? If you match that as a starting point it's likely to be ok.

I like lots of voltage overhead for safety, but it's probably not that critical. A 600v part would probably be ok on a 240v supply and a resistive load.

You still have not said what the original fault was that blew the first part?

 

[danrogers]

the original part was 600v i think from the datasheet "The 3P4MH and 3P6MH are P-gate fully diffused mold
SCRs with an average on-current of 3 A. The repeat peak offvoltages
(and reverse voltages) are 400 V and 600 V."

Should be ok then? I'm not sure how the fault occurred to be honest? I guess it looks like a short?

 

[Mr RB]

It's probably ok, and worth a try. You never said what the wattage of the heating element was, but I guess it is FAR below 240v @ 2.5A that the C106M is specced at, that is 600W! Hair straightener clamps likely use something like a 50W or 60W element.

And as I said before if you can fit a fuse in series with the heating element, at the PCB end of the wires, that will protect the PCB if the heaters get some wiring short condition. A 240v 2.5A slow blow fuse should be fine.

 

[()blivion]

Never forget that 240v is volts RMS. The peek to peek is much higher, from +340 volts to -340 volts.

 

[danrogers]

Thanks very much. The power of the heating elements are only stated at 35w I think, so I guess this part should be fine https://uk.rs-online.com/web/p/thyristors/6562548/ Datasheet - https://www.electro-tech-online.com/custompdfs/2013/06/0900766b80d3f729.pdf

thanks for all the help

 

[Mr RB]

That TIC106M part looks good! It's got the 0.2mA sensitive gate current of the 106 series but a larger package and 5A avg, 30A peak. Better than your original part, good work!