Substitution for germanium transistor

If the transistor you need is a DTG600, NTE179 is a substitute and is germanium. I have one in stock if you want it. This is a TO-3 style transistor right?

The posted schematic was a bit fuzy, so if I have the part number wrong, please post the correct one.

Dan

Because the current limiting is partially determined by the relationship between the voltage drops of the diode (0.3V) and the emitter / base junction of the transistor. If you change only the transistor to (0.6V) you will probably never develop enough voltage on the transistor to start it conducting.

Mikek

Without looking, I would say the diode is a form of temperature compensation and the material of the 2 devices should be the same. That would be true if they are mounted on the same heat sink.

On 1/5/2017 11:38 AM, amdx wrote: ...

Ah! That makes perfect sense. Thank you.

The transistor is a DLG600. I remember seeing that the NTE179 was a substitute, but the NTE179's that I found were just about as expensive. Especially compared to the $0.48 for the TIP32C.

Thanks just the same.

Bob

Oops - D _T_ G, as you said

Try eBay for an NTE179: Yep, expensive.

DLG600 is rated at 90V/25A, while the TIP32C at 100V/3A. That would seem to be a rather poor choice for a substitute.

If you substitute silicon, you'll need to change some component values to deal with the Vbe change from 0.3v to 0.6v. Probably doubling the resistance of the 8.2 ohm sense resistor, and changing D7 from germanium to silicon. Hopefully, the current ranges will remain the same. The silicon device will have a better gain-bandwidth product, so a few ferrite beads to keep the power xsistor from oscillating. If you choose to use a different package, you'll need to do some creative mounting and heat sinking. The analog series current limiting transistor will dissipate considerable heat at high loads.

Sounds like too much work. I'll see if can find one in my junk pile. I've been hoarding germanium devices for years for use in repairs.

TBH: I'd strip out the original current limiter circuit and rebuild with one from a more modern linear PSU.

Its usual practice to use a separate small signal transistor who's B/E threshold detects the volt drop across the current sense resistor. Its collector shunts away base current from the pass transistor. If you go that far - a Darlington pass transistor has a number of advantages.

The B/E threshold is different between silicon and germanium, so you need to re calculate the current sense resistors. There's various online archives of test gear schematics - you can probably find a circuit that you can pretty much copy.

It looks like a very simplistic version of the current limiting circuit. The diode Vf is pretty much a reference voltage generator that the sense resistor volt drop is compared to - the fact that the diode also provides dome degree of temp-comp, is probably incidental.

Too much. I could even live without the limiter, but for a couple of cheap parts, I'll fix it. But not anything more.

On 1/5/2017 1:04 PM, Jeff Liebermann wrote: ...

What you don't know, cause I didn't say, is that the supply is 50v, 1.5A.

There is no calibration in the circuit - the current limit is picked by rotating a pot until the current is limited where you want it. Hopefully this will be able to account for the different device. There was a poster who did this, so I'm probably OK.

I will mount it, isolated, to the steel frame. And at 1.5A that "should" be enough sinking.

My order is in, but thanks anyhow.

Bob

Yeah - the diode and transistor are very thermally independent.

I've knocked a complete current limiter together in an evening - basically just a scrap of sheet aluminium and a few components that were laying around.

You could use a Si transistor and fit the matching Si diode, but you'll probably have to recalculate all the sense resistors - which is most of the work.

Mikek, would replacing the germanium diode with a silicon one fix the voltage drop you speak of?

John :-#)#

There are twenty regulars here that could give you all the details with the math, I'm not one of them. If it were me, I'd change both diode and transistor to silicon and see how it works. I think it will. Mikek

It looks like the diode generates a reference voltage - that needs to be kept in step with the Vbe of the new transistor.

The silicon transistor still needs more Vbe to get it conducting, so I suspect the sense resistors will also need some adjustment.

I have not heard about germanium in years. When I played around with old tube stuff in the 60s - 70s, I remember germanium diodes were fairly common. I never ran across a germanium transistor.

Reading this thread made me question what germanium really is, and I read the following article (good article). I know it was used to make the FIRST semiconductors, I never knew much more about it. It appears it's a costly elemental material.

Here are a few clips from that article.

From:

Germanium "metal" (isolated germanium) is used as a semiconductor in transistors and various other electronic devices. Historically, the first decade of semiconductor electronics was based entirely on germanium. Today, the amount of germanium produced for semiconductor electronics is one fiftieth the amount of ultra-high purity silicon produced for the same. Presently, the major end uses are fibre-optic systems, infrared optics, solar cell applications, and light-emitting diodes (LEDs). Germanium compounds are also used for polymerization catalysts and have most recently found use in the production of nanowires. This element forms a large number of organometallic compounds, such as tetraethylgermane, useful in organometallic chemistry.

- Germanium differs from silicon in that the supply is limited by the availability of exploitable sources, while the supply of silicon is limited only by production capacity since silicon comes from ordinary sand and quartz. While silicon could be bought in 1998 for less than $10 per kg, the price of germanium was almost $800 per kg.

Apparently it was refined from flue dust where a specific type of coal was used.

Not difficult to see how it might be hard to come by nowadays.

Germanium transistors were very common in the 1960's. For example, I was selling all germanium Motorola pagers and partly germanium mobile radios at the time. All of the old vibrator power supply to transistor conversions used germanium power transistors. Most of the AM car radios were germanium.

Germanium is not dead today. There are SiGe devices that use the best properties of both materials to good advantage.

Germanium is also the preferred material for Infrared lenses.

Yep. Germanium is expensive. Price at the end of 2015 was $1,760/kg ($800/lb):