How to measure germanium Hfe TUTORIAL!!! (pcb & spreadsheet)

[Carlos Best]

Guys, I searched a lot for something like this and it seems nobody did this for dummies. So being one myselff I decided to post something  

First of all here is the theory behind this:
http://www.geofex.com/Article_Folders/ffselect.htm

Here you have the pcb and layout for the circuit noted in the above link
https://mega.co.nz/#!gt5hibYI!9vbFakCUUPgjxATjofUDmreCp7Lp6QfwLrRtNXxaFEw

Download and burn. You have to hand pick a resistor that is closest to 2.2M, and with a muti turn pot, dial 2,472 (or 2,47 if your DMM goes off range)
Populate the board.

After that hook up everything as per the diagram in the first link. The board silkscreen tells you were you have to connect everything anyway  .

When your done, pinch your Germanium Tranny on the socket, paying attention to the pinout.
Set your DMM to DC Voltage in the 2v range and read the value. Write it down
Press the switch at the board. Write that down

Load that values in this spreadsheet . (idea taken from here http://www.diystompboxes.com/smfforum/index.php?topic=102014.msg910364#msg910364)
https://mega.co.nz/#!Jxp1DSQJ!P0kSoyHGug41b3j-6kr3oYj9TxempQMUc4q9Jbjw7es

First column first reading, 2nd column 2nd reading.
The leakage current and hfe will appear at 3rd and 4th columns respectively

Thats all.

Tell me if there is any error in the steps and I'll correct it.

Regards!

[midwayfair]

Awesome! I'll give it a go when I can get to a computer that can actually run the spreadsheet.

If the voltage supply is different from 9V, can that be input and will it compensate?

[Carlos Best]

I think if you mess with the 9v power supply that will mess with the 4uA calculations. That needs to be exact 9v for accuracy.
I just copied the formula from the topic above, (http://www.diystompboxes.com/smfforum/index.php?topic=102014.msg910364#msg910364)
Maybe DiscoVlad could clarify it.

If you are asking about the spreadsheet, it is made dinamicaly so if you change the 9v at the Psupply section it will change the other values.

[slacker]

You don't really need to use a pot to get exactly 2,472 Ohms, R.G only uses that value because it works out that you just take the voltage you measure and multiply it by 100 to get the gain, which is easy to do in your head. The formula you use in your spreadsheet calculates the gain in a different way so there's no need to use the exact value you can just use a 2.4k resistor.

[Carlos Best]

Hi Slacker!

Great news. I was just re reading Rg article and I can't realize why he picked that resistor value
In fact, that resistor is just an excuse to use the voltimeter at the DMM and then calculating the current with the Ohms law.
There you can just stick the DMM in series and use the current measurement function setted in a lower multiplier...

[Arcane Analog]

There is a tutorial on FSB that is very good for beginers.

I would link to it but apparently that is forbidden on this site. Search for "RG's Tester for Idiots" and you should find it.

[LucifersTrip]

http://tinyurl.com/k5j5f5t

[davent]

You don't really need to use a pot to get exactly 2,472 Ohms, R.G only uses that value because it works out that you just take the voltage you measure and multiply it by 100 to get the gain, which is easy to do in your head. The formula you use in your spreadsheet calculates the gain in a different way so there's no need to use the exact value you can just use a 2.4k resistor.

Grab a few 30k & a few 2k7 resistors and bingo...

[Carlos Best]

There is a tutorial on FSB that is very good for beginers.

I would link to it but apparently that is forbidden on this site. Search for "RG's Tester for Idiots" and you should find it.

Yep. I read that post.
you have to use a breadboard, and lift one end with each transistor. Definitely not practical for a batch of 70 germaniums as I tested
That and the risk that lifting things the resistor could disconnect something else, so you have to recheck, and a large etc.

Also, the leakeage is not calculated. You have to compare your results with a table.

This way you build the circuit, measure, load the spreadshet and voila. It goes as accurate as your election of components and power supply.

[Arcane Analog]

There is a tutorial on FSB that is very good for beginers.

I would link to it but apparently that is forbidden on this site. Search for "RG's Tester for Idiots" and you should find it.

Yep. I read that post.
you have to use a breadboard, and lift one end with each transistor. Definitely not practical for a batch of 70 germaniums as I tested
That and the risk that lifting things the resistor could disconnect something else, so you have to recheck, and a large etc.

Also, the leakeage is not calculated. You have to compare your results with a table.

This way you build the circuit, measure, load the spreadshet and voila. It goes as accurate as your election of components and power supply.

No - you do not need a breadboard to do that method - you can usealigator clips and make the setup very easy - almost like an Atlas. I would say your setup is actually harder to fit transistor legs into a tiny PCB. That is simply agravating. On the physical end of things, breadboarding and aligator clips would be far easier on the head and faster.

I am not sure you have addressed it or not but your setup does not ensure that you are getting the proper voltage to the transistor either.

Leakage is extremely easy to calculate when you use the components listed. The chart he posted is for reference.

[Arcane Analog]

If you are going to test a lot of germanium without an atlas, one of these or something similar is much better than a breadboard or a PCB with tiny sockets.

I am sure the 'experts' use their vastly superior Atlas tester but I highly recommend making something like this if you aim to test any decent amount of germanium:



You can watch the transistor fluctuate and/or settle which gives you indications on the device's quality that an Atlas cannot. I own an Atlas and I rarely use it other than for rough sorting.

Calculating leakage is easy.

I can try to offer a simple response.

First, if using the RG method, you must regulate your power supply to an even 9V. A simple pot works well for that as it can dump the excess juice from your adapter or battery and make it easy to hit the 9V that RG specifies - along with the 2.472K and 2.2M resistors.

Measure your transistor for leakage with the tester and the old multimeter:
0.345V

As a side note, a lot of people use uA to note/discuss leakage: uA = microamps - mA = milliamps - 1000 uA = 1 mA
0.345V divided by 0.002472 = ~139.6uA of leakage

Flip Switch for the 'total' gain reading:
1.666V

So take the 'total' gain reading and subtract the leakage:
1.666 - 0.345 = 1.321V

Take that number and multiply by 100 for your 'true' gain.
1.321 x 100 = 132

Hope that helps.

To quote RG...

Openoffice spreadsheets will save as an excell (.XLS) spreadsheet. Do "Save as" and pick the appropriate file format.

However, I still philosophically object to using a nuclear spreadsheet to squash a fly. It's very nearly the level of arithmetic that can be quickly done in your head.

[davent]

I found some small spring actuated terminal blocks with a .10" spacing, added a second  Emitter connection on the other end so no awkward lead re-arrangements to accomodate the different pinouts.

Also added an adjustable regulator, feed in 12vdc and trim to 9v.



dave

[Carlos Best]

Arcane,
I didn't want to start a controversy

My setup works, and is virtually hasle free for a beginner. No calculations, and they obviusly can ommit the pcb and use a breadboard, or alligator clips or they bare hands to twist the wires.
There are a lot of ways to do it better, but this works very well.

Here are two test rigs, one RG Keen, the other from Smallbear. There's a slight modification you can do, adding a spdt with the comon to the base, to connect it to the emitter for pure leakage measurement, and the other to the 1M or 2.2 M resistor to ground.

[Arcane Analog]

No controversy. I am simply stating your setup with the floating PCB and small sockets look terribly awkward and cumberson for measuring quantity.

Edit: Are you sure your calculations are compensating for supply voltage?

[Arcane Analog]

This layout is more friendly and provides a switch to avoid moving those clips around. Add a trimmer to trim the DC to 9V exactly. Slap it all in a box with proper srping sockets to avoid fighting with the transistor leads and you are in business.

http://tagboardeffects.blogspot.ca/2012/08/germanium-transistor-tester.html

[slacker]

Edit: Are you sure your calculations are compensating for supply voltage?

His calculations take the supply voltage into account, a different supply voltage gives different operating conditions eg; the base current will be different so the calculations give the correct results for those operating conditions.
R.G's method will give the wrong result in the same conditions because it assumes 4uA base current.
 Using a different supply voltage might not give the same answer as using exactly 9 volts, depends how much the gain varies with different base currents. The calculations don't attempt to adjust the answer to compensate for this if that's what you meant.
Whatever method you use so long as you use about 9Volts, about 2M2 and about 2K47 the result is probably close enough for rock 'n' roll.

[mac]

In my gallery there are WIn, OSX and HTML helpers to automate calculations that include different PS voltages and bias resistors.
As you know hfe is not constant but a function of the collector current.
For example, some of my power germs like AD1xx have 2x more gain if base current is 50% higher than 4ua as in RG test.

This is the HTML script, wiring is missing but you can download it from the 2nd link and put both the page and gif in the same folder,

http://www.aronnelson.com/gallery/main.php?g2_view=core.DownloadItem&g2_itemId=43773&g2_GALLERYSID=1d31f3ccf9bb60f827f1ad6a04067a56

http://www.aronnelson.com/gallery/main.php?g2_view=core.DownloadItem&g2_itemId=43770&g2_serialNumber=1

mac

[DiscoVlad]

My Spidey sense is tingling...

I think if you mess with the 9v power supply that will mess with the 4uA calculations. That needs to be exact 9v for accuracy.
I just copied the formula from the topic above, (http://www.diystompboxes.com/smfforum/index.php?topic=102014.msg910364#msg910364)
Maybe DiscoVlad could clarify it.

Do I have to?

As I recall, the original reasoning was that 4µA is the approximate base current of FuzzFace transistors, and that the testing circuit is only for an EASY quick and dirty way of finding transistors that sound "good" in a FuzzFace. There's really no benefit to being completely anal about component values here (±1% tolerance on a 2M resistor is a ±20kΩ range).

Remember that Ge devices are incredibly temperature sensitive compared to Si transistors! The current gain, and base/collector currents all change with temperature. These effects have been ignored in the calculations -> They complicate things unnecessarily and a simple device selection circuit sitting on a test bench in no way reflects conditions found on a stage. Be aware that the effect exists though.

The Power supply only needs to be 9.0V to make the calculations with a 2472Ω resistor easier to do mentally (Use a bench supply, dial in 9V). Obviously if you're trying to test at eg. 18V it's going to throw things out of whack, but ±0.5V from a battery doesn't/won't make that much difference.

You don't really need to use a pot to get exactly 2,472 Ohms, R.G only uses that value because it works out that you just take the voltage you measure and multiply it by 100 to get the gain, which is easy to do in your head. The formula you use in your spreadsheet calculates the gain in a different way so there's no need to use the exact value you can just use a 2.4k resistor.

Regarding the spreadsheet formula: No it doesn't, it calculates it the same way as R.G. did, but in a general way that uses the measured value of any 2.4k resistor instead of a specially selected 2472Ω one.

Whatever method you use so long as you use about 9Volts, about 2M2 and about 2K47 the result is probably close enough for rock 'n' roll.

Exactly!

Sorry if my tone seems a bit annoyed, people seem to get really hung up about component selection when in a majority of cases (for 9V effect pedals at least) it doesn't matter anywhere near as much as people think it does.  

[mac]

You can get a perfect 9v supply and perfect resistors... but you can't get a stable temperature
And another thing to keep in mind is that Vbe is not constant, it also affects the *100* factor

The general formula for any set of resistors and power supplies is,

hFE = [Vnc - V]*Rb/[Vcc - Vbe]/Rc

where Vnc is the collector voltage when Rb is Not Connected,
and V is the value when Rb is connected.

RG tester set Rb/[Vcc - Vbe]/Rc = 100, supplying a base current similar to a FF.

As leakage goes to zero, silicons, Vnc tends to Vcc. You can use the formula to measure silicons as well.

mac

[Arcane Analog]

Sorry if my tone seems a bit annoyed, people seem to get really hung up about component selection when in a majority of cases (for 9V effect pedals at least) it doesn't matter anywhere near as much as people think it does.  

I have a powersupply that has nearly 10V and i suspect that there are many supplies similar out there. I would prefer to use 9V and stick with the 2M/2K4 as opposed to fudging it. Finding a 2M and 2K4 resistor and making the supply 9V is too easy. Testing at or close to room temperature is a no brainer.

If you are going to bother testing germanium you might as well put the effort in a do it right.