EEVblog Electronics Community Forum

Electronics => Beginners => Topic started by: billbyrd1945 on November 26, 2018, 02:06:26 pm

Title: Controversy on Transistor Testing: Can't be both!
Post by: billbyrd1945 on November 26, 2018, 02:06:26 pm
Here's a drawing that shows how I understand the test to be done. Several videos show it done this way. Conflicting videos show that the last transistor in the drawing should be OL so that the middle and right pins would pass no voltage in either direction. One method condemns my transistors while the other says they're good. What do you guys say? I have no schematic. The markings are all I have.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: ataradov on November 26, 2018, 03:34:10 pm
According to a quick search, 13005D appears to have a built-in diode. That's what you are measuring in the last two pictures.

And the pinout makes sense, it is B-C-E (left to right).
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: nsrmagazin on November 27, 2018, 12:18:00 am
According to a quick search, 13005D appears to have a built-in diode. That's what you are measuring in the last two pictures.

And the pinout makes sense, it is B-C-E (left to right).

Also what you are looking for is that both video options are true. You have to check for your transistor and see which pin is what. Its not a guarantee that it will always be the same.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: billbyrd1945 on November 27, 2018, 01:10:32 am
Okay guys. I get what you're saying. This is what I was just about to post:

In this video at 11:50, Lorton shows a convincing diagram that inhibits flow from C to E in either direction for NPN or PNP. Are you saying this does not apply in my application?  (I think that's what the two of you are telling me.) Just trying to learn. Thanks!

(Please see image below from video)

https://www.youtube.com/watch?v=_3G8t-cV1d8 (https://www.youtube.com/watch?v=_3G8t-cV1d8)
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: nsrmagazin on November 27, 2018, 01:17:19 am
What my post told you is that not always the pins are "B, C, E" = pin 1, pin 2, pin 3. You have to check the datasheet for your current transistor. The other post told you that there is a diode connected with the transistor and when you measure you are not actually measuring the "NPN" transition of the transistor, but you are measuring the "PN" transition of the diode. A transitor is a "double diode" a diode is a "half transitor" in terms of how they are built.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: billbyrd1945 on November 27, 2018, 01:32:54 am
Thank you nsrmagazin. I had looked at every 13005 datasheet I could find, but didn't understand the circuitry. So I resorted to YouTube. Here's an image of the circuitry on one datasheet. I've added red to what I suspect you're saying explains why E will flow to C, but not vice-versa. I must learn more about how these circuits work. Thanks to both of you.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: nsrmagazin on December 04, 2018, 12:56:50 am
I am sorry to offend you, but truth hurts and it does not change!

If you don't understand these basic principles, you are a long way from bulding and measuring circuits.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: billbyrd1945 on December 04, 2018, 03:34:50 am
"If you don't understand these basic principles, you are a long way from bulding and measuring circuits."
I don't use forums very often. I always feel so stupid. Sooner or later, I get a message such as yours. Maybe some day you'll need to know about my area of expertise.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: Wimberleytech on December 04, 2018, 03:43:59 am
"If you don't understand these basic principles, you are a long way from bulding and measuring circuits."
I don't use forums very often. I always feel so stupid. Sooner or later, I get a message such as yours. Maybe some day you'll need to know about my area of expertise.

Don't worry or feel intimidated.  This is the forum area for "beginners" so we expect beginner questions.  Not a problem...keep the questions coming.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: billbyrd1945 on December 04, 2018, 03:49:43 am
Thank you sir!
I was beginning to think I had mistakenly walked in to a phD class room.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: Benta on December 04, 2018, 04:08:12 am
I had looked at every 13005 datasheet I could find, but didn't understand the circuitry. So I resorted to YouTube. Here's an image of the circuitry on one datasheet. I've added red to what I suspect you're saying explains why E will flow to C, but not vice-versa.

Whoever made that schematic must have been on prohibited substances.

The (MJE)13005 is a completely normal high voltage NPN transistor (400 V). The 13005D is the same, with the addition of a diode from emitter to collector. This is often seen in switching transistors.

Ignore the derogatory comments, this is the "Beginners" section.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: billbyrd1945 on December 04, 2018, 04:53:09 am
Thank you for your kindness sir.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: nsrmagazin on December 05, 2018, 12:17:19 am
If you are working on circuits without knowing the basics you risk hurting yourself. The very least you can burn your circuit.

Its better to start with the principles that you don't know, instead of just getting a ready answer.

EDIT: and I did not say you are stupid! What I meant is you should start with the beginning!
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: billbyrd1945 on December 07, 2018, 10:33:14 am
Finally found two things that I could actually measure, an opamp and what I think is a transformer. I replaced the opamp and all the caps. I have 110vac going to the transformer and about 1 volt coming out. Out of circuit, the primary and secondary (if that's what they are) test open. The number on the transformer is UM-007/UF 16-45 and seems to exist no where except on my boards. So I don't have any way of knowing what to order. The boards ultimately produce 12vdc. But I don't know if that means the transformer in question should be replaced with 110vac/12vdc or something else. Any help with be appreciated.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: rhb on December 07, 2018, 12:35:17 pm
The forward voltage drop across the base emitter junction is higher than the drop across the base-collector junction.  With older meters you can use the resistance range.  The lower resistance is the base collector junction because it has larger area.  Newer meters you'll need to use the diode test function to get enough voltage to get the junction to conduct.

If it's not just a transistor I can't say what to expect.The BSIDE ESR02 Pro is $20, has a nice case and seems to do a good job.  But I used the method in the first paragraph for 40 years.  I learned it from a semiconductor physicist who was teaching an intro electronics class in the building next to where I was looking at rocks through a microscope all day.  I didn't take the course, I just went to the lectures for a break.

In the 40 ears since, I have encountered two people who knew how to do this.  Everyone knows how to find the base, but knowing how to identify the collector and emitter is is uncommon in my experience.  I find that sad as it is so simple and logical.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: billbyrd1945 on December 07, 2018, 02:36:39 pm
Thank you sir! Back to the drawing board.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: rhb on December 07, 2018, 03:30:38 pm
Around 1979 I had built a number of simple things using a package of  Radio Shack 2N2222s that did not work.  I became convinced that either the transistors were all bad or the pin out on the package was wrong.  I went around asking everyone I knew how to identify C & E.  No one knew and this included the ex-Navy techs who did all the exotic lab  instrument repairs such as X-ray diffraction machines and such.

When I heard the explanation from the physicist.  I didn't run home because it was over a mile and up and down hill, but I went as quickly as I could walk which is 4-5 miles per hour.  I pulled one of the remaining transistors out of the package and tested it.  C & E were opposite the package labeling.

After that I had great fun with anyone I ran into who was seriously good at electronics. I'd ask them.  I  have only encountered a couple of people who knew, and only in the last 5-10 years.

I look forward to a post from you saying the problem is resolved. If not, I look forward to resolving it. Can't always get it right the first time.  I've never played with big transistors, just small ones.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: Wimberleytech on December 08, 2018, 01:41:50 am
I look forward to a post from you saying the problem is resolved. If not, I look forward to resolving it. Can't always get it right the first time.  I've never played with big transistors, just small ones.

BE reverse breakdown will be relatively low (5-10 volts).
BC reverse breakdown will be relatively high (>20 volts).

Using a simple single-trace curve tracer, you can test this readily, or with a sufficient power supply and limiting resistor.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: Wimberleytech on December 08, 2018, 01:55:57 am
Here is my junction tester (curve tracer) that I built in 1974 or '75.  It is currently connected to my analog scope (465B) in a dedicated test setup.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: rhb on December 08, 2018, 01:52:10 pm
That's just an octopus!  That's not a curve tracer for determining transistor parameters.  The octopus is a great diagnostic tool, but it will not give you they kind of results a Tek 577 gives.  The OP is wanting a cheapie and small 577.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: Wimberleytech on December 08, 2018, 03:01:27 pm
That's just an octopus!  That's not a curve tracer for determining transistor parameters.  The octopus is a great diagnostic tool, but it will not give you they kind of results a Tek 577 gives.  The OP is wanting a cheapie and small 577.

Hey rhb!  You are brilliant.  Thanks for sharing.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: rhb on December 08, 2018, 11:26:28 pm
Sorry.  I conflated this with another thread.  One glass of wine to many.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: Wimberleytech on December 09, 2018, 01:07:53 am
Sorry.  I conflated this with another thread.  One glass of wine to many.

No harm...do I win the prize for the C-E answer?
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: rhb on December 09, 2018, 09:20:36 am
You're the 3rd person to know how to tell them apart, but using the reverse breakdown is a new one on me.  So yes. You certainly do.   I learned another way to do it.

Learning something new always makes my day.  At least if it isn't a "do do that again" something new.

Clearly I'm going to have to build an octopus.  I've known about them since the mid-70's but never got around to making one.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: Wimberleytech on December 09, 2018, 09:30:41 am
You're the 3rd person to know how to tell them apart, but using the reverse breakdown is a new one on me.  So yes. You certainly do.   I learned another way to do it.

Learning something new always makes my day.  At least if it isn't a "do do that again" something new.

Clearly I'm going to have to build an octopus.  I've known about them since the mid-70's but never got around to making one.

Last night I was having dinner with a guy who used to work for me.  I asked him the question.  He had the same answer as I.  Well...we are both IC designers for 40 years, so we think beyond the schematic (junctions, doping level, parasitics, etc.).  So, it is not surprising that we would get the same answer.

I was in high school when I built the octopus I have now from the schematic in the earlier post.  That pretty much gives away my age.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: rhb on December 09, 2018, 09:49:43 am
I think that you're correct.  If you design ICs you have to know the semiconductor physics.  I'm sure that the prof I got the method I use from knew because those details were what he specialized in.

And then, of course, there was Bob Widlar, who not only knew the physics, but could come up with ways to calculate things no one else could such as the resistance of an irregular patch of silicon using a sheet of resistive paper.

It bothers me that every intro book on transistors tells you how to identify the base, but not C & E.

Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: Jwillis on December 10, 2018, 11:24:41 am
Correct me if I'm wrong but I believe that is a common mode choke in the first stage filtering.
(https://www.eevblog.com/forum/beginners/controversy-on-transistor-testing-cant-be-both!/?action=dlattach;attach=589993;image)
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: nsrmagazin on December 23, 2018, 09:14:14 am
Age doesn't need to be bad, but it can be.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: spec on December 26, 2018, 10:02:08 pm
Hi billbyrd1945,

Just read your thread. :)

As a general point. The classic transistor test only applies to a single bipolar junction transitor (BJT), not a transistor with another speed up transistor built in, a Darlington transistor, logic transistor or an IGBJT. So before you do a test it is best to check the device datasheet to establish what the device is. And, as has been stated, the pin-out can vary even for devices with the same part number.

But there is another gotcha that you may like to know about: leakage current. Some BJTs have high leakage current which can make them appear to be faulty on an an elementary DMM test. Devices with high leakage are germanium transistors, schottky transistors, and high power transistors.

There is also a little trick. If you have a small signal NPN transistor, say a BC546, or BC337, and you want to check that it is operating as a transistor, you can set your multimeter to the high ohms range, then connect the positive lead to the collector and the negative lead to the emitter and you should get a high resistance. But if you then touch the positive lead (collector) with with your index finger on one hand and the transistor base lead with your other hand the DMM will show a lower resistance reading, and the harder you press with your index fingers,  the lower the reading will be. This is quite a good test because it demonstrated that the transistor actually has current gain (it works because dry index finger to dry index finger is around a 100k Ohms for most males).

This test is best done with analoue multimeters, but remember that with analogue multimeters on the ohms ranges, the positive lead has a negative voltage while the negative lead has a positive voltage.

Of course, to test a PNP BJT in the same way, you you just reverse the multimeter leads.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: glarsson on December 26, 2018, 10:38:30 pm
This test is best done with analoue multimeters, but remember that with analogue multimeters on the ohms ranges, the positive lead has a negative voltage while the negative lead has a positive voltage.
This "reverse polarity" was/is very common, but not all analogue meters are like this.
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: spec on December 26, 2018, 11:16:53 pm
This test is best done with analoue multimeters, but remember that with analogue multimeters on the ohms ranges, the positive lead has a negative voltage while the negative lead has a positive voltage.
This "reverse polarity" was/is very common, but not all analogue meters are like this.
I didn't know that- thanks for the information. :-+
Title: Re: Controversy on Transistor Testing: Can't be both!
Post by: 6PTsocket on January 02, 2019, 02:09:33 am
This test is best done with analoue multimeters, but remember that with analogue multimeters on the ohms ranges, the positive lead has a negative voltage while the negative lead has a positive voltage.
This "reverse polarity" was/is very common, but not all analogue meters are like this.
I didn't know that- thanks for the information. :-+
Now I have to go check out my old Simpson 260. I always liked it for checking transistors. The forward conduction was always around 10-15 ohms.

Sent from my SM-G900V using Tapatalk