TIP142 and TIP147 available with 20 Amp continuous collector current rating?

[Chris Wilson]

I see some Ebay adverts for TIP142 and 147 with a rating of 20 Amps continuous collector current, Is this legit, I see some TIP142's from Farnell or RS Components with a 20A rating, but they are dear and uncommon, but no 147's. Thanks.

20A TIP147's : http://www.ebay.co.uk/itm/1-X-1-piece-TIP147-Darlington-Transistor-PNP-20A-100VTO-247-L3113-/251466543058?hash=item3a8c92efd2


20A TIP142's : http://www.ebay.co.uk/itm/5-x-STMicroelectronics-TIP142-NPN-Darlington-Transistor-20A-100V-TO-247-/321818076280?hash=item4aedda2478

[Kleinstein]

The TIP142/147 are specified for 20 A peak current, at best.  So don't expect to get something better than the datasheets says.

With such offers there even is a chance to get fake parts.

So for such standard parts, and not even a good price, why not use a regular source for components.

[mikeselectricstuff]

At 20 amps you really want to be using MOSFETs not darlingtons, due to the voltage drop.

[Chris Wilson]

I would only buy branded from a respected reseller, it's just I didn't know if a 20A TIP147 even existed, it seems 20 Amp TIp142's do :

http://uk.rs-online.com/web/p/products/4860294/

Thanks  for the reply.

[MagicSmoker]

If you think a transistor spec is dodgy then the first thing you should do is look at the datasheet from a reputable manufacturer. In the case of bipolar transistors I would check OnSemi (formerly Motorola) and Fairchild (formerly... actually, still Fairchild).

And from what I see, there ain't no such thing as a 20A TIP147.

[mikeselectricstuff]

I would only buy branded from a respected reseller, it's just I didn't know if a 20A TIP147 even existed, it seems 20 Amp TIp142's do :

http://uk.rs-online.com/web/p/products/4860294/

Thanks  for the reply.

If you look at the datasheet, it say a 20A abs max peak,  10A abs max.

[Chris Wilson]

Ah, right, a typo perhaps, thanks for that Mike. As an aside I discovered your superbly written article on making your own PCB's yesterday. Quite the best resume I have come across!

Are you still a fan of Ferric Chloride as an etchant, or has anything changed? There's some other "stuff" that's become popular, that's supposed to be better, faster and far less messy, but I can't remember the name of the chemical.

Thanks again.

[Kleinstein]

Wether a MOSFET is better suited really depends on the application. MOSFETs also have a few difficulties:
 - often poor SOA
 - SOA tested part are rare and expensive
 - large varations in threashhold
 - difficult to paralel in linear applicaions
 - gradual onset at low currents

So a MOSFET is good when it comes to switching, but often a poor choice in a linear power application.

[mikeselectricstuff]

Ah, right, a typo perhaps, thanks for that Mike. As an aside I discovered your superbly written article on making your own PCB's yesterday. Quite the best resume I have come across!

Are you still a fan of Ferric Chloride as an etchant, or has anything changed? There's some other "stuff" that's become popular, that's supposed to be better, faster and far less messy, but I can't remember the name of the chemical.

Thanks again.

Never seen a reason to change, but as it's in a dedicated fixed tank, the mess issue isn't a problem, so not investigated alternatives. It's cheap, and has infinite shelf life.
I very rarely do homebrew PCBs these days.

[mikeselectricstuff]

Wether a MOSFET is better suited really depends on the application. MOSFETs also have a few difficulties:
 - often poor SOA
 - SOA tested part are rare and expensive
 - large varations in threashhold
 - difficult to paralel in linear applicaions
 - gradual onset at low currents

So a MOSFET is good when it comes to switching, but often a poor choice in a linear power application.

People still do linear power stuff ?

[MagicSmoker]

...
Are you still a fan of Ferric Chloride as an etchant, or has anything changed? There's some other "stuff" that's become popular, that's supposed to be better, faster and far less messy, but I can't remember the name of the chemical.
...

The other etchant that can be made and used at home is a 2:1 solution of 3% hydrogen peroxide (disinfectant for minor wounds) and 31.5% hydrochloric acid (aka "muriatic acid" used for swimming pools). It is very fast, doesn't need to be heated and is transparent.

[Circlotron]

MOSFETs also have a few difficulties:
 - often poor SOA
So a MOSFET is good when it comes to switching, but often a poor choice in a linear power application.

Are you kidding??? They have no second breakdown so they can be great for linear service e.g. Series pass element in a regulated power supply.

[T3sl4co1l]

Not sure how many times this needs repeating on one forum, but:

Modern MOSFETs have high enough integration density that they do exhibit second breakdown.  "Freedom from second breakdown" was a myth from the early days of low density and lateral MOS.

That said, at low voltages, both BJT and MOS tend not to exhibit second breakdown, and "linear" types are still available today (in both families) which exhibit a large (or even full) SOA.

Please RTFDS, people.  That is all.

Tim

[SeanB]

Second that Tim, no real issue with SOA provided it is operated as a fast saturated switch. Run with large part of the allowable current and not fully saturated, and switch slowly and it will go boom. There is a reason Mosfet gate drivers can source and sink over 2A on many of them, so the device spends as little time in the linear region as possible.

[Chris Wilson]

Second that Tim, no real issue with SOA provided it is operated as a fast saturated switch. Run with large part of the allowable current and not fully saturated, and switch slowly and it will go boom. There is a reason Mosfet gate drivers can source and sink over 2A on many of them, so the device spends as little time in the linear region as possible.

Which, as an aside, is a potential issue I am investigating running a 1kW 136kHz amp in modes that stop and start... It all seemed so simple when I started making it, but gremlins lie in the shadows, just as people who rebuild complex engines at home never see the potential problems a pro engine man like myself might foresee

But just like the kind folks here help newbies like myself, I try and give back where my own speciality knowledge can help Thanks everyone, great forum! Let's see how long these 4 spare MOSFET's last me

[PChi]

Using MOSFETs is a good idea. Power transistor maximum current ratings can be misleading because they aren't necessarily usable at the headline number. According to the Texas Instruments TIP147 data sheet (TI Plastic Power which I presume is the original) the current gain at 10 A is 500 minimum which is half that at 5 A of 1000 and according to the typical characteristics graph the current gain is falling rapidly with current so at the peak rated transistor current of 15 A it will be considerably lower.

Warning rant: Rogue parts are now getting bought in volume and fitted to PCBs causing mayhem. I don't recall this being a problem a couple of decades ago. What sort of low life sells fake LDOs at a few cents each?

[T3sl4co1l]

The OP hasn't stated if this is for a linear or switching application, or what the actual application voltages and currents are.

A TIP142/7 may be wholly unsuitable for switching because of storage time.

A single BJT or MOSFET may be wholly unsuitable because of power dissipation (linear or otherwise).  Alternative designs might be required.

A MOSFET may be less desirable than a BJT, for inrush/current limiting reasons.

There are many reasons to choose one over another, not just what comes to mind first.

Tim

[Chris Wilson]

An acquaintance sketched a schematic mod to use the TIP125 instead of a switch or relay to turn on a 12V rail in an amplifier when the exciter went to TX. I was just looking at TIP125's on Ebay and on various re-sellers sites and saw some were seemingly rated 20 Amp continuous. For completion this was the sketch. Thanks.

[Circlotron]

MOSFETs also have a few difficulties:
 - often poor SOA
So a MOSFET is good when it comes to switching, but often a poor choice in a linear power application.

Are you kidding??? They have no second breakdown so they can be great for linear service e.g. Series pass element in a regulated power supply.

My apologies to Kleinstein.
It seems that under some conditions when operating in linear mode a mosfet die may develop hot spots. This lowers the local threshold voltage and so that hot area digs in further until 

This is a totally separate issue to when a mosfet is fully turned on and any localised die heating causes the drain-source resistance in that hot area to increase, causing the hot spot to back off.