OC29 series pass transistor modern sub?

[Al (astral highway)]

I'm restoring an late 60s power supply and need a decent TO3 sub for OC29 as a series pass transistor ...

Most data sheets I can find lead to items I can't source. Any ideas, please?

Thanks folks.

[MrBungle]

On semi MJ15016G looks to do the job - disclaimer applies that this was a quick search.

RS part 774-3215 - 176 in stock.

Check pinouts obviously.

The OC29 has quite a bit higher hFE than a MJ2955 which is the goto PNP pass for these sorts of things but that might work anyway.

MJ2955 isn't stocked any more by the looks in TO-3 but you can use a TO-220 MJE2955 for the job if you don't mind it appearing different.

Depends on the design as well but might need a damping capacitor to kill off oscillations between collector and base (increase miller effect) as the transition frequency is a lot higher.

[Al (astral highway)]

Thank you! Detailed and v helpful answer!

[Refugee]

Beware of the biasing.
If current limiting is deployed it will almost be quadrupled if a germanium part is replaced with a silicon part.

[AC/HL]

The genuine article is available from trusted sources on eBay. Not exactly cheap, but fit and forget. Do you know how the original died?

[Kala_12]

Hi,
A quick search took me here. Any use?

http://alltransistors.com/crsearch.p...fe=45&caps=TO3

Stuart

[MrBungle]

Watch out for the data on alltransistors.com. Some of it was badly copied out from Towers' international transistor selector by the looks. Some 0's are 8's and some 8's are 0's and the figures matches. That got me a couple of times in the last few years.

[Nymrod121]

Al, would an OC28 suffice? - I have one here (ex-BBC PS2/49) which I'd be happy to send you FoC.
Cheers
Guy

[Al (astral highway)]

Originally Posted by AC/HL

... Do you know how the original died?

It's not clear if they're actually dead, technically. But there is irremediable corrosion on the pins and on the cases as the unit has clearly been stored in a damp place for a long time.

Many of the other semiconductors on board test badly.

[Craig Sawyers]

I've seen non-abused germanium power transistors which measure exceptionally badly - way below what they would have done originally, and leads me to wonder if there is a degradation mechanism somewhere.

[MrBungle]

They do indeed degrade. Leakage current goes up with usage/time. Also they suffer from tin whiskers and brittle contact bonding apparently.

I wouldn't go near a germanium transistor these days unless it's selling them to guitar pedal builders. You can categorise them with hFE and leakage current and if they are under expected values, they are worth a few quid

I actually built a fully automated test rig with an arduino, relay board and a logging multimeter to calculate this. This was after I inherited a bag of 300 of them

[Al (astral highway)]

Originally Posted by MrBungle

I wouldn't go near a germanium transistor these days unless it's selling them to guitar pedal builders...

Ah, so I may also have a problem with the ACY20's on board! I'll try to test them in circuit, although not easy!

[MrBungle]

If you can get them out without blowing them up, you can use this method to get the leakage and hFE. This was what my test rig was based on:http://www.geofex.com/article_folders/ffselect.htm

I used a couple of resistance substitution boxes for the resistors and a bench supply but you can just wire up a few in series until you get the right values.

[Al (astral highway)]

Originally Posted by MrBungle

If you can get them out without blowing them up, you can use this method to get the leakage and hFE. This was what my test rig was based on:http://www.geofex.com/article_folders/ffselect.htm

Very neat idea indeed. I also like your incredibly clear description, illustrations and lucid writing. Kudos to you and thanks!

It may not be possible to dig these out without damage - the circuit board tracks show severe corrosion from sustained moisture damage, very likely with saltwater hallmarks, to make things worth. Let's see!

[Nicklyons2]

I've always viewed OC28,OC29, AD140 and AD149 to be largely interchangeable; can't say I've ever had any problems with that 'philosophy'.

[G6Tanuki]

Add NKT404 to that list... !

[julie_m]

According to my Seeburg parts list, there's also a GC-114 that can be substituted by an AD149.

[Restoration73]

MJ15004 is TO3 and still current. If you need a higher hFE,consider using a PNP Darlington
transistor bearing in mind the different drive requirements e.g BDX62 and others.

[Al (astral highway)]

Originally Posted by Restoration73

MJ15004 is TO3 and still current.

Lots of good ideas and too many people to thank individually but I appreciate all of these contributions. Wrt to the ACY20's, one of these didn't survive desoldering (wire pull-out) and the other two failed on leakage current.

I've ordered a batch of ex-Soviet equivalents that I'll test individually before subbing them. I may eventually start another thread on that in case people are interested in the viability of old PNP germanium transistors?

Cheers folks!

[MrBungle]

Originally Posted by astral highway

Very neat idea indeed. I also like your incredibly clear description, illustrations and lucid writing. Kudos to you and thanks!

I can't claim I wrote that,merely followed the instructions