LED Curve Tracing / FV knee with DCA75

[chromesphere]

Jez from Peak sent me a DCA75 to review and shoot videos on.  I've been playing around with it over the last couple of nights, lots of fun!

Here's some forward voltage graphs of diodes and leds i have made up tonight showing the diode knees for your curiosity.   If there's anything you want to see specifically please feel free to request and i will try to get it done for you.

geVSsi_CB_Junction: a 2n5088 collector base junction (silicon) vs a GT308V collector Base Junction (germanium)

leds.jpg: The DCA specifies that LED colors towards the red end of the spectrum have lower VF drops compared to those towards the blue spectrum.  You can see this clearly in this graph.  These LEDS are 5mm "economy" (i think from tayda).

[Lubdar]

This is quite informative!
Naively, I suppose a warp control would be 'most effective' blending a red and blue LED?

[drolo]

I'm surprised by the linear curve (can you say that?) of the 1n60

[EBRAddict]

I'm surprised by the linear curve (can you say that?) of the 1n60

Sure behaves like a resistor. I'm not sure that cutoff region is real maybe the DCA75 test current doesn't go below a certain threshold.

[stecykmi]

I'm surprised by the linear curve (can you say that?) of the 1n60

Sure behaves like a resistor. I'm not sure that cutoff region is real maybe the DCA75 test current doesn't go below a certain threshold.

it is weird, maybe another one needs to be tested?

[chromesphere]

The test circuit has a 1k resistor as load for the LED and sweeps voltages at 51 stages from 1v to 12v.  So the minimum current would be 1v / 1k = 1ma test current.

I tested the diode again this morning with the quick test function of the DCA, it showed a FV of .63v which is suspicious of course being a germanium diode.  I double checked with my DMM and it showed .32v.

I found some info in a datasheet, scroll down to page 14.  Unfortunately its a little cryptic...

http://www.datasheetarchive.com/dl/Scans-048/DSAGER000208.pdf

[chromesphere]

I should have said "0v to 12v" of course 

Anyway, i redid the test with 3 samples each of oa1160 and 1n60.  I think the DCA is reporting correctly.  It looks like the diode becomes forward biased around 180mv.  I'm very keen to compare these to something with a sharp knee in an effect now...!

It might be overkill but i think i'm going to add a curve trace image to diodes (used mainly for clipping) on my store, i find this stuff quite fascinating. 

[EBRAddict]

I'll try to run a few Ge diodes on my DCA this weekend and do a side-by side with waveform clipping on a scope.

[EBRAddict]

I ran a few tests with a waveform capture and DCA Vf plot side by side

1N5817
1N4148
5mm Red LED
1N34A (I each SBE and Pedalhacker)
1N270
D93

This should be a link to a PDF on Dropbox:
https://www.dropbox.com/s/z2pgor0srqq79o2/DiodeTest.pdf?dl=0

I also noted the 1N270 and both 1N34a didn't show a nice hard clip like the 1N4148. I'm guessing leakage or stray capacitance?

[blearyeyes]

I love this stuff, really facinating. Why the big difference between the Fv on Fluke and DCA75 ?

[chromesphere]

I love this stuff, really facinating. Why the big difference between the Fv on Fluke and DCA75 ?

I would presume this is because of the voltage both units are testing the diodes (higher with the DCA).

[blearyeyes]

I love this stuff, really facinating. Why the big difference between the Fv on Fluke and DCA75 ?

I would presume this is because of the voltage both units are testing the diodes (higher with the DCA).

So there's no standard?  I don't recall seeing the Fv quantified in any specs but I don't know.

[chromesphere]

Cant find it for the Fluke 287 but for my DMM, the BK Precision 2709B its 1.2ma
For the DCA its 5ma.  Much higher.  I dont think explains the curve of the 1n34a, 1n60 etc.  I can only imagine this is just how they behave and its a curve trace from 0 - 12v
EDIT: An example is the BAT41.  The peaks quick mesaurement shows .629v at 5ma.  My bk dmm shows 0.389v (at approx 1.2ma).  Take a look at the curve trace of the bat41, you can see these 2 corrospending measurements.  Curve tracing shows you the bigger picture.   I think its really valuable information, ive added a curve trace to each of the diodes used popularly for clipping on the store.

http://www.diyguitarpedals.com.au/shop/index.php?main_page=product_info&cPath=2&products_id=39

[EBRAddict]

I was playing around with R for the first time in years and thought these time series plots would be interesting to the forum. It's the same data from the plots attached in the PDF above. Something wonky is going on with the 1N34A diodes.

Si diodes:


Ge "normal" diodes:


These 1N34A have me scratching my head. They behave like voltage dividers. Diodes from three vendors (SBE, Pedal Hacker, Mammoth) act like this. I'm going back and testing some pedals I've used them in and make sure I'm actually getting clipping instead of attenuation. Some of them don't act right and I suspect this may be a reason.

[midwayfair]

I tested the diode again this morning with the quick test function of the DCA, it showed a FV of .63v which is suspicious of course being a germanium diode.

It's not suspicious, it's something about the Peak. Fv is also current-dependent, but despite my interest in diodes I lack the knowledge to explain real testing methods.