this looks totally useful... and CHEAP!!

[electrip]

Or maybe someone can reprogram for GE leakage.....

Not neccessary, mine does measure and display GE-leakage:
ICE0 (leakage base floating) and ICEs (leakage base shorted to emitter)
MK-168 Firmware 1.10k

To compare different component analysers one has to know the test conditions of every parameter tested.
hfe is dependent on Ic (and Uce),
Uf is dependent on If
C_reverse is dependent on U_reverse

Example (OC75 @ 27.0°C)
DCA75pro Test Result:

PNP Germanium BJT
Red-C Green-B Blue-E
HFE=78 at Ic=5,01mA
Vbe=0,370V at Ib=5,00mA
IcLeak=0,377mA   

MK-168 fw1.10k Test Result:
PNP
B=69
Uf=177mV
ICE0=0.37mA
ICEs=0.02mA

Lower B/HFE  and Uf/Vbe readings indicate a lower Ic/If for measurement
Now one could check the R.G. Keen-method against it
which actually tests at different Ic's depending on the transistor HFE.

Documentation for the chinese-german-tester is here:
http://www.mikrocontroller.net/svnbrowser/transistortester/Doku/tags/english/?view=tar
You have to extract a tarball  (*.tar.gz). 7zip does it for you on MS-OS based computers.

electrip

[italianguy63]

Got my magic box yesterday in the mail.  This is an awesome thing!

Seems to be very accurate.  I already identified the values on some mystery components.  The Hfe seems pretty close too.  I had one germ I had labled at 57, and it read 67.  Most of the time when I measure them, they start out higher using the R.G. method and take a short while to stabilize with voltage run through them-- and the final Hfe is lower...  But, man.. HOW EASY.

MC

[pinkjimiphoton]

got mine a week ago, just tried it today. best 2o bux i ever spent. identifys everything i throw at it, shows all kinds of useful info. i don't know if it displays leakage, mine displays hfe and ua in miliamps which i'm guessing is the proper bias current or something?
cool toy, for sure...

without reading the whole thread (sin, i know) anybody figure out what the definitive diy one is? i'd love one that read leakage on ge's

[Luke51411]

I ordered this one http://pages.ebay.com/link/?nav=item.view&id=161002450834&globalID=EBAY-US
It was cheaper but it seems like the extra 4 bucks would have been worth it from the reports here. It seems to work well enough thoug I have some mp20a germs that measure as p channel JFETS in one direction and when rotated they read in the 400s for hfe though some of them read around 50 to 80 which would be expected range. For JFETS it seems to just give the pinout. Resistor is spelled resister haha. For mosfets it gives Vt reading, is that anything useful? It doesn't appear to give anything for leakage as far as germs. It gives a Uf reading in mv is that forward voltage?
Has anyone had similar experiences? Is the slightly more expensive unit as much better as it seems?
Also of note this one doesn't have a graphic interface for pinout, it just has for example GDS=312 and it is not backlit. All in all it still seems like a nice tool just not everything I hoped it would be.

[Ripdivot]

Or maybe someone can reprogram for GE leakage.....

Not neccessary, mine does measure and display GE-leakage:
ICE0 (leakage base floating) and ICEs (leakage base shorted to emitter)
MK-168 Firmware 1.10k

To compare different component analysers one has to know the test conditions of every parameter tested.
hfe is dependent on Ic (and Uce),
Uf is dependent on If
C_reverse is dependent on U_reverse

Example (OC75 @ 27.0°C)
DCA75pro Test Result:

PNP Germanium BJT
Red-C Green-B Blue-E
HFE=78 at Ic=5,01mA
Vbe=0,370V at Ib=5,00mA
IcLeak=0,377mA   

MK-168 fw1.10k Test Result:
PNP
B=69
Uf=177mV
ICE0=0.37mA
ICEs=0.02mA

Lower B/HFE  and Uf/Vbe readings indicate a lower Ic/If for measurement
Now one could check the R.G. Keen-method against it
which actually tests at different Ic's depending on the transistor HFE.

Documentation for the chinese-german-tester is here:
http://www.mikrocontroller.net/svnbrowser/transistortester/Doku/tags/english/?view=tar
You have to extract a tarball  (*.tar.gz). 7zip does it for you on MS-OS based computers.

electrip

electrip, would you mind taking a photo of your tester showing what the display looks like when testing a transistor? I think I may pay the extra bucks to get the one you got because it tests leakage. Also yours tests JFETs does it not? Much appreciated.

[Ripdivot]

Or maybe someone can reprogram for GE leakage.....

Not neccessary, mine does measure and display GE-leakage:
ICE0 (leakage base floating) and ICEs (leakage base shorted to emitter)
MK-168 Firmware 1.10k

To compare different component analysers one has to know the test conditions of every parameter tested.
hfe is dependent on Ic (and Uce),
Uf is dependent on If
C_reverse is dependent on U_reverse

Example (OC75 @ 27.0°C)
DCA75pro Test Result:

PNP Germanium BJT
Red-C Green-B Blue-E
HFE=78 at Ic=5,01mA
Vbe=0,370V at Ib=5,00mA
IcLeak=0,377mA   

MK-168 fw1.10k Test Result:
PNP
B=69
Uf=177mV
ICE0=0.37mA
ICEs=0.02mA

Lower B/HFE  and Uf/Vbe readings indicate a lower Ic/If for measurement
Now one could check the R.G. Keen-method against it
which actually tests at different Ic's depending on the transistor HFE.

Documentation for the chinese-german-tester is here:
http://www.mikrocontroller.net/svnbrowser/transistortester/Doku/tags/english/?view=tar
You have to extract a tarball  (*.tar.gz). 7zip does it for you on MS-OS based computers.

electrip

electrip, would you mind taking a photo of your tester showing what the display looks like when testing a transistor? I think I may pay the extra bucks to get the one you got because it tests leakage. Also yours tests JFETs does it not? Much appreciated.

Well this sucks, apparently the new M328 firmware doesn't show leakage.

http://www.ebay.de/itm/Transistor-Tester-Diode-Triode-Capacitance-ESR-Meter-MOS-PNP-NPN-Module-Schale/271498772909?_trksid=p2047675.c100011.m1850&_trkparms=aid%3D222007%26algo%3DSIC.MBE%26ao%3D1%26asc%3D24201%26meid%3D8900038664636516450%26pid%3D100011%26prg%3D10287%26rk%3D1%26rkt%3D10%26sd%3D271556264712

[electrip]

electrip, would you mind taking a photo of your tester showing what the display looks like when testing a transistor? I think I may pay the extra bucks to get the one you got because it tests leakage. Also yours tests JFETs does it not? Much appreciated.

Well this sucks, apparently the new M328 firmware doesn't show leakage.

http://www.ebay.de/itm/Transistor-Tester-Diode-Triode-Capacitance-ESR-Meter-MOS-PNP-NPN-Module-Schale/271498772909?_trksid=p2047675.c100011.m1850&_trkparms=aid%3D222007%26algo%3DSIC.MBE%26ao%3D1%26asc%3D24201%26meid%3D8900038664636516450%26pid%3D100011%26prg%3D10287%26rk%3D1%26rkt%3D10%26sd%3D271556264712

Hi Ripdivot,

I've got the one from your ebay link, same seller, same tester.

M328 is the processor/controller type, firmware (functionality) is 1.10k on mine which is up to date.

Mine does measure JFETs but I don't like the data format I @ Vg.

I prefer the Vgs(off), gfs, Id and Vgs(on) data from my Peak DCA 55/75.

Leakage is only displayed if the tested transistor has some leakage
so most of silicon bjt will show only HFE and  Uf data.

In my opinion it is a 'good to have' thing with its pros an cons.
I once zapped my PEAK LCR 40 and bought a new one (quite expensive)
but i wouldn't do it again because the MK-168 does it all for less money.
And it does diode_capacity, ESR and both poti-resistances at once.
Easy for a quick check of any part at your workbench.

The testers with bigger display will not measure more or better data it is just the presentation.
I was going for the one with the enclosure for practical reasons.

For serious research on specific semiconductors nothing beats the ATLAS PEAK DCA75 pro
with its curve tracer functions and data export to spreadsheet analysis (-> spice model).
And you can download updates from the manufacturer.

Hope this helps a little,

electrip

[Ripdivot]

That is very helpful, thank you electrip!  There are versions with the nicer display that come in a case like this:

http://www.ebay.com/itm/12864-LCD-Transistor-Tester-Diode-Triode-Capacitance-ESR-Meter-MOS-PNP-NPN-LCR-/121342332277?pt=LH_DefaultDomain_0&hash=item1c40910d75

Hey karbomusic, have you tried measuring a germanium transistor in your tester to see if it will show leakage when there is leakage present?

[karbomusic]

That is very helpful, thank you electrip!  There are versions with the nicer display that come in a case like this:

http://www.ebay.com/itm/12864-LCD-Transistor-Tester-Diode-Triode-Capacitance-ESR-Meter-MOS-PNP-NPN-LCR-/121342332277?pt=LH_DefaultDomain_0&hash=item1c40910d75

Hey karbomusic, have you tried measuring a germanium transistor in your tester to see if it will show leakage when there is leakage present?

I did and it didn't display it but I only tried one due to being in a hurry. Will try again tonight. I'll tell you one unexpected benefit. This morning was cleanup day and bet I had 200 various parts lying around in little paper cups, baggies and on my workbenches. I had been putting off, putting them back in their right places due to all the differing values and so on. This made it a breeze in comparison, sometimes it's the little things.

[electrip]

That is very helpful, thank you electrip!  There are versions with the nicer display that come in a case like this:

http://www.ebay.com/itm/12864-LCD-Transistor-Tester-Diode-Triode-Capacitance-ESR-Meter-MOS-PNP-NPN-LCR-/121342332277?pt=LH_DefaultDomain_0&hash=item1c40910d75

Hey karbomusic, have you tried measuring a germanium transistor in your tester to see if it will show leakage when there is leakage present?

That's a nice one but the socket is soldered on, no leads and clips.
If someone sells this within the EU maybe I buy a second one. ;-)

Back to the leakage:

I forgot to mention that on the small display
leakage is shown only for some seconds
and then the display shows HFE and Uf until shutdown.
So you have look while testing to get the leakage data.

Maybe it is shown all together on the big displays but I don't know!

electrip

P.S.:
This one looks very nice:http://www.ebay.de/itm/111395885356
Built in LiIon-battery charged via micro-usb.

This does zenerdiodes up to 50 V. Small processor AT168 although:
http://www.ebay.de/itm/111253189174

[JohnForeman]

i just got the one from the original post and was wondering if anyone has been able to access any sub functions by long pressing?  If i long press, all i get is a display contrast control

[italianguy63]

That's what I get too -- contrast control.

[italianguy63]

I just measured a bunch of leakage on some Germaniums.  Compared the RG method to the gadget.  Values all over the place...  Some measure substantially higher, some measure substantially lower.     On the average, the batch I did (16 transistors) were 122 Hfe (RG), and 142 (gadget).  I'm going to do some more...  MC

[electrip]

I just measured a bunch of leakage on some Germaniums.  Compared the RG method to the gadget. 
Values all over the place...  Some measure substantially higher, some measure substantially lower.    
On the average, the batch I did (16 transistors) were 122 Hfe (RG), and 142 (gadget).  I'm going to do some more...  MC

Sorry for quoting myself:
http://www.diystompboxes.com/smfforum/index.php?topic=108294.msg986737#msg986737
Hope this helps a bit.

electrip

[italianguy63]

I just measured a bunch of leakage on some Germaniums.  Compared the RG method to the gadget.  
Values all over the place...  Some measure substantially higher, some measure substantially lower.    
On the average, the batch I did (16 transistors) were 122 Hfe (RG), and 142 (gadget).  I'm going to do some more...  MC

Sorry for quoting myself:
http://www.diystompboxes.com/smfforum/index.php?topic=108294.msg986737#msg986737
Hope this helps a bit.

electrip

Hey, thanks for that!  I am doing the RG vs. MK-168.

I just did 10 more, with better results.  This time, across the board, all but two were measured 35% to 40% higher by the MK-168 vs. the Keen Method.  By RG They averaged 126Hfe and the gadget got 169Hfe. on the average.  I am going to go do 10 more that I have to do, and call it quits.

MC

Edit-- Did another 9.  1 was bad.  Anyway.. similar results, but the MK-168 read between 30% to 70% high.  Avg. was 43% higher.

[karbomusic]

I just measured a bunch of leakage on some Germaniums.  Compared the RG method to the gadget.  
Values all over the place...  Some measure substantially higher, some measure substantially lower.    
On the average, the batch I did (16 transistors) were 122 Hfe (RG), and 142 (gadget).  I'm going to do some more...  MC

Sorry for quoting myself:
http://www.diystompboxes.com/smfforum/index.php?topic=108294.msg986737#msg986737
Hope this helps a bit.

electrip

Hey, thanks for that!  I am doing the RG vs. MK-168.

I just did 10 more, with better results.  This time, across the board, all but two were measured 35% to 40% higher by the MK-168 vs. the Keen Method.  By RG They averaged 126Hfe and the gadget got 169Hfe. on the average.  I am going to go do 10 more that I have to do, and call it quits.

MC

Edit-- Did another 9.  1 was bad.  Anyway.. similar results, but the MK-168 read between 30% to 70% high.  Avg. was 43% higher.

Pardon the noob ignorance but I've never seen leakage displayed on my gadget. If that is the case could the discrepancy be due to the gadget not taking leakage into account? And if the gadget does show leakage, how do I make mine do that? 

[electrip]

Edit-- Did another 9.  1 was bad.  Anyway.. similar results, but the MK-168 read between 30% to 70% high.  Avg. was 43% higher.

What specific types of transitors were tested?

@karbomusic:
Which tester do you have? (ebay-link?)

electrip

[italianguy63]

Mine was the one from post #1 in this thread.

EDIT-- Karbo, it does NOT appear to be related to the leakage.. Both high and low leakage are off by a similar %.  But, there is still a good amount of randomness in it.

I tested some GT402B and GT402G Russians.

I also tested some NKT275s just now, and they read like 80% higher.

Working on some AC128's as I write this...

Mine doesn't show leakage.  I do it with a spreadsheet, DMM, breadboard rig, and a laptop.

I'm actually just trying to determine if I can use the gadget instead of the rig and save gobs of time and effort.  The answer is MAYBE.  If I take the reading and deduct off like 40%, I will probably be in the ballpark....  for germs anyway.

MC

[tonyharker]

Don't forget that the values of germanium leakage and Hfe, are affected by temperature. The action of measuring will heat up the device due to the current flowing through it.  If one measurement is made then the device must be allowed to return to room temperature before a further measurement is made. 
Also the different test methods will use different voltages (Vce) This will also affect values of leakage and Hfe.

[italianguy63]

Thanks Tony.. doing all the measurements the same.  Doing the 2nd measurement while the device is still "hot" as the 1st is after the measurement takes time to stablize...

Here is a sample of what the data looks like.  This is from a sample of (10) AC128's I just did:

Hfe (RG)   Leak   Hfe (MK168)   % Diff.
54   158   85   57%
108   608   211   95%
95   488   192   102%
80   375   145   81%
83   325   154   86%
113   758   251   122%
56   242   103   84%
57   313   105   84%
102   467   218   114%
75   308   136   81%

It is typical of what I have been seeing.