ZTX109C replacement.

[G0HZU_JMR]

I rescued hundreds of old Zetex devices at work many years ago. I still have quite a few although I rarely use any of them.

Most of them are varactor diodes but I also have some NPN and PNP transistors in amongst them all. A quick rummage delivered a bag of 100 x ZTX300 (E line) still in the Farnell bag, another bag of 100 x ZTX500 and another bag of ZTX313. I think the ZTX313 is a switching transistor and this is the bag that I've dipped into over the years.

Is there anything special about these devices? The datasheet suggests they are just jellybean parts.

[Cruisin Marine]

I recall reading in a mag (maybe R&EW, can't remember) that these trannie's had better ft and current handling capabilities than the spec. sheet showed. Whether that was true, I have no idea. But, the one's I did buy and use, always worked very well.

[paulsherwin]

The ZTX300 is really just a jellybean transistor like the 2N2222. There's nothing wrong with it though. Check the ratings if you think the application may be stressing it.

[Bill]

Originally Posted by Valvepower

Hello,

Really sorry tangent coming up…. Was it Ferranti Zetex, who was behind the Practical Electronics Orion amplifier from 1975? I found one at a Swapmeet, jumble etc., a good while back – to be honest I really can’t remember exactly where…

Was this the Zetex take on the Texas Texan?

Sorry about the tangent!

Terry

The name Zetex did not appear until late 70s or early 80s. I have somewhere a book published by Ferranti Applications Lab containing design(s) for solid state Hi Fi amplifiers, I haven't seen it for years but the output devices were ZT3055s and it would date back to 1970 or so. The applications lab at Gem Mill was rather nice, a group of very bright people busy finding uses for the thousands of different Ferranti devices with no particular direction from management. The lab published all sorts of information on the products and not necessarily aimed at industry for sales.
I still have a few prototype very tiny LEDS (green,yellow , red) made in 1969 at Gem Mill, these may be the first leds made in UK?
I have mentioned before on here: once the E-line (ZTX) production got going , it cost 1p per device to make and the cheapest lowest spec transistor was sold for 4p, this was to a toy manufacturer (don't know who) that just needed a device that transisted!

[Diabolical Artificer]

I've found a few more duff ZTX109C's and tried replacing them with some BC107's I have. These were in a buffer circuit, see attached, where I think the hfe spec isn't an issue so much, it's an emitter follower soless than unity gain right?. As mentioned the BC107 has less gain, I'm wondering if I can pull good 109C's out of the buffer circuits, sub them with 107's, then use these pulled 109C's in circuits where gain is an issue, like mic preamps. Does hfe effect input impedance?

I also have some BC184's that may do if not, the hfe is a tad higher though the pinout is different.

Andy.

[paulsherwin]

These are just general purpose transistors, and there's no point in tying yourself in knots over replacements. The hfe won't be critical and anything over about 300 should be absolutely fine. Use your BC184s, or buy a bag of BC549s. A BC107 isn't low gain but isn't low noise either - it's a high voltage BC108.

It's odd that lots of the original transistors have failed though.

[Restoration73]

Versions of the E-line package included those with a heatsink tab.

I was given a bag of devices that looked and worked the same as ZTX300,
they are marked BT8616, so you know who the client was !

[Valvepower]

Hello,

Ooh, that’s and mid 70’s Allen and Heath desk… nice mixers those. From memory they were based around the Mullard circuits found in the Mullard Transistor Audio and Radio Circuits book.

I’ve found the circuit for an earlier (c. 1973?) Allen and Heath mixer which uses BC169 and BC168 transistors – see attached pictures.

What is the condition of the electrolytic capacitors as if they are the originals, they maybe becoming leaky and upsetting bias conditions.

I remember the electrolytic capacitors we used in the Canary desks, which we bought form George English Electronics back in the late 70’s, could have slightly high leakage even back then! However on the whole they were OK.

Terry

[Diabolical Artificer]

It's odd that lots of the original transistors have failed though

I've found a few but most seem to be ok.

That's interesting Terry, I have those older schematics but thanks for uploading anyway. The caps are Nichi branded "CE310-325" ? They are light blue, I'm sure I've seen them in other stuff of the same era. Most of them are in spec with an ESR similar to replacement caps I have. I've found a few that all though the ESR is ok, they have gone up in value. The only other components that are tired or damaged are some 10k skeleton presets and there are several damaged 1/4" jack sockets which is to be expected.

Andy.

[mhennessy]

It might be a problem with the E-line package - perhaps it's less resistant to moisture than it should be? I've never really used them much, but have replaced a fair few failed ones over the years - but without analysing the operating conditions, it wouldn't be fair to blame the transistors themselves at this point.

I'd be replacing them with the BC549 - and given what you've found so far, I'd be considering a blanket replacement for peace of mind. And I wouldn't worry about hFE.

Swapping them around probably isn't a good idea as the heat cycling might give misleading results. And if a transistor has gone noisy for whatever reason, it is degraded and if anything, it can only get worse, so even in a less critical position, I just wouldn't have long-term confidence in it.

Caps that read high are caps that are leaking. Whether that matters depends on the surrounding circuitry. Of course the true value of capacitance might be less than it should be, which might make a difference (again, depends on the environs). As with the transistors, given the age and the fact you've found some duffers, a blanket recap wouldn't be unreasonable.

BTW, a cap that is dead short will look perfect on an ESR meter. ESR is just one of several parameters. It's understandable that people focus on ESR in this age of switched-mode power supplies, but it's not the whole story.

That said, the last audio mixer I overhauled was made in the early 1990s, and nearly every electrolytic was shot - as in practically open-circuit. Fairly early on I stopped fault-finding and went into shotgun mode, but out of interest I did measure each one that I removed, and not one of them was even half-decent. And there were a good brand, 105 degrees as well. Luckily they hadn't leaked electrolyte...

[knobtwiddler]

>nearly every electrolytic was shot

Nichicon? If memory serves me right, there was a period where their caps had high failure rates a decade or two back, but they've managed to recover their reputation since.

It's not uncommon to hear of studios re-capping Neve VRs and SSL E/G consoles every 5 years or so. Quite a few of these mixers will be several sets of caps into their lives. The VR, in particular, bakes them (along with switches and other parts that get clogged from hot, dusty air currents).

[Valvepower]

Hi,

Allen and Heath used good quality parts back then, but they will have aged and seen better days. If you want to put reliability first, then replacing the capacitors is preferred. I've had studio/mastering engineers say desks sounded better after replacing the capacitors.

Some 40 years ago a friend had the earlier (as per the diagram) mixer and it sounded rather nice. The Mullard type circuits were widely used as I also remember them being used in the DJ Electronics Disco mixers in the early 1970’s.

If it helps, I have some old-school TI BC384B in the workshop, which I feel is a useable equivalent. I’ve just popped one on the Atlas analyser and the hFE is 329 and the pinout is EBC. I have a bag of these, and I could easily spare a Dozen FOC if they are of interest, I’ve more than I'll ever need of this part.

Terry

[bluepilot]

Originally Posted by paulsherwin

It's not clear why ZTX transistors would be used in a 70s mixing desk. Ferranti had a reputation for making high quality but expensive components for military projects. Maybe the desk was a homebrew 'foreigner' by a Ferranti engineer.

When I was at school around 1970 our physics lessons included electronics. We had a load of "modules" which could be plugged together to make various circuits. I found pictures of some here, here and here
They all used Ferranti transistors and the physics lab had a good supply of ZTX something or others as spares. Interesting that they are now museum pieces.

[G6ONEDave]

ZTX type transistor could be found in all manner of consumer goods, Ferguson is one brand that comes to mind. I think the ZTX range superceded the BFS range. Personally I would try BC109C or BC549C but check the voltages first, in terms of what the supplies are in the mixing desk.

The specs for the ZTX109C are; vcb = 45v, vce = 30v, veb = 5v, Ic max = 100ma, Tj max = 125c, p tot = 300mwf, Ft min = 150M, Cob max = 5p, Hfe = 450mn and Hfe bias = 2ma.

Dave

[paulsherwin]

The circuit fragment shows a 24V rail, so nothing to worry about there.

If you want to err on the side of caution, you could use a BBC550 which is both low noise and HV.

[Diabolical Artificer]

I've been giving each strip a clean and cleaning SW's and pots, then checking DC conditions on each channel strip/OP strip first against A&H's figures in the manual, IP'ing a 1khz 100mV sinewave to check continuity, re-capping each "section", re-checking DC conditions, then a final AC check. I'm getting quicker with each strip.

For now I'm trying to get it all working whilst familiarising myself with the desk then I'll go back probably and change the transistors. The few duff ZTX109C's I've found so far pass a DMM test but show as "part damaged" or "pair of diodes" which might equate to them being leaky.

Thanks for all your IP, Andy.

[Herald1360]

A transistor can be C-E short, terminally leaky, and still check out as a pair of diodes.

The "wet finger C-B" with meter connected C-E check works OK with a DMM as well as with an analogue meter.