Unit7

Noob question about measuring transistors
« on: October 12, 2013, 11:22:54 AM »
Here's yet another noob question… (Please see question in bold below if you want to skip all the babble)

I just ordered everything for David's NV73 preamp build (for 51X/24V). Re the discontinued BC184c transistors I decided to go for a pack of 100pc from an eBay seller.

I've read all I could find about the build and noticed that the 184c are good for this design because of the higher HFE. I don't know what HFE is, and I'm not sure I'd understand even if someone tried to explain, but as I understand higher HFE values helps keeping noise down.

I also read that I should pick transistors with a HFE value of 600 or higher for certain positions. I'll ask David eventually about where those are in his design.

The bottom line is that I will have to measure my BC184c transistors before completing the build.

I googled a little and found the basics about transistors - collector, emitter and base. I still don't understand much what they really do in a design but perhaps enough to be able to measure.

One article I found describe a way to measure a transistor with a multimeter set to measure resistance, to see if the transistor is healthy or not. This brings to my question: Is a multimeter enough to measure the HFE, or do I need another tool?

I've seen special measuring tools for transistors but would prefer to find a cheaper way to do it because this is something I won't do very often.

Grateful for any advice
Paul


JohnRoberts

Re: Noob question about measuring transistors
« Reply #1 on: October 12, 2013, 12:18:25 PM »
HFE or beta is the current gain of a transistor. In other words, how much current flows in the emitter in response to a given base current.

A simple VOM curent measurement in series with say collector or emitter in response to a base current set by a resistor and known voltage, would allow for simple beta measurements (emitter current divided by base current).

For most useful measurement, I would try to replicate as much of the topology and voltages as the device will be used at in circuit, so the measurements will be most representative.

If the base current is 100uA as used in the circuit don't measure the beta at 2 mA, etc.

JR

Visit https://circularscience.com to hear what properly "cleared" drums sound like.

Unit7

Re: Noob question about measuring transistors
« Reply #2 on: October 12, 2013, 12:39:06 PM »
Thanks John!

I believe I understand. I do understand the part that I should replicate the conditions of the application, I guess in this case the specific position in the NV73 preaamp.

But to feed the proper voltage/current into the transistor I guess I'd need some kind of bench lab?

JohnRoberts

Re: Noob question about measuring transistors
« Reply #3 on: October 12, 2013, 12:54:35 PM »
for something like this you don't need a fancy precision lab power supply. A 9V battery could work as long as it's voltage doesn't change too much over the course of the measurements.

A few resistors could be used to set base current. Even this does not need to be precise, If matching beta you just want them to be the same, to grade for beta you want it to be high enough.

Sorry I realize this is still complicated if just starting out

the BC 184C is an NPN. It looks like beta is highest around 2 mA of collector (emitter current), falling lower at higher and lower collector current.

2 mA of collector current calculates out to around 8 uA of base current.

So with emitter connected to ground or 0 terminal of an 9V battery, and 1M resistor from top of 9V battery to the base you can measure the current with a VOM between +9v and collector lead.

Divide the collector current by 8.5uA to determine actual device beta.

JR

Visit https://circularscience.com to hear what properly "cleared" drums sound like.

Unit7

Re: Noob question about measuring transistors
« Reply #4 on: October 12, 2013, 01:16:18 PM »
John, thanks again!!

Even if I won't do the measuring like you describe I will study your post to see if I can learn something eventually. It will keep me buzy for a few hours!.. :)

I just found that transistor HFE testers are included in some cheap DMMs. I thought a tester was from like $100 and upwards, but it seems I could get one of those DMMs for like $30 which is totally ok for me.

Do you think that would be enough for this purpose?

JohnRoberts

Re: Noob question about measuring transistors
« Reply #5 on: October 12, 2013, 06:07:57 PM »
As I mentioned ideally you want to test the device close to how it will be used. If the cheap DMM HFE tester allows you to test the device at different base currents it could be OK.

Years ago I owned a heath kit curve tracer which was used for testing and matching transistors. It was used in combination with an oscilloscope and  gave a pretty display of device behavior, but I ended up giving away to an audio-phool (since RIP) who wanted to match and select transistors for some innocuous buffer inside a old CD player... I was glad to give it to somebody who found it useful.

If I had a HFE tester I'd give it to you.  8)

JR
Visit https://circularscience.com to hear what properly "cleared" drums sound like.

Unit7

Re: Noob question about measuring transistors
« Reply #6 on: October 13, 2013, 07:51:46 AM »
John, wow thanks for all the advice! It reminded me to finally make a donation to GDIY!

After studying your posts and googling transistors for hours I think I've decided to order a 9V battery socket, a three way socket for the transistor and a 1M resistor and build myself a little BC184c tester! Good for future NV73 builds and also for my AMS/Neve reissue modules in my studio.

I'm curious about the value (1M) you suggest on the resistor. You mentioned that the value didn't matter, so... did you calculate the division of 8.5 (of the collector current) as a result of using a 1M resistor? And/or did you calculate 1M to create 2mA at the base of the transistor?

Also, what kind of resistor would you choose for this? Metal film? Watts? Tolerance?

JohnRoberts

Re: Noob question about measuring transistors
« Reply #7 on: October 13, 2013, 11:20:25 AM »
John, wow thanks for all the advice! It reminded me to finally make a donation to GDIY!

After studying your posts and googling transistors for hours I think I've decided to order a 9V battery socket, a three way socket for the transistor and a 1M resistor and build myself a little BC184c tester! Good for future NV73 builds and also for my AMS/Neve reissue modules in my studio.

I'm curious about the value (1M) you suggest on the resistor. You mentioned that the value didn't matter, so... did you calculate the division of 8.5 (of the collector current) as a result of using a 1M resistor? And/or did you calculate 1M to create 2mA at the base of the transistor?
I was just pulling numbers off the top of my head.

You can calculate the base current using Ohms law.  With the emitter connected to 0V, the base will be roughly 0.5V, so 1 meg resistor from 9V supply will have 9V-0.5V drop across it. 8.5V/1,000,000 = 8.5uA.  Since the nominal beta or HFE of that transistor is  250x, 8.5uA x 250= 2.1 mA collector current or close enough for government work. to the 2 mA  ideal for the device.

You can make this better than I can if the devices are operating at something other than 2mA in circuit, make the 1 meg resistor higher or lower to target that different nominal operating current.

My comment that this doesn't need to be precision, is because to match devices you just want to make them the same, so who cares if you HFE measurement is off a few % if the devices measure the same they will be the same.

Quote


Also, what kind of resistor would you choose for this? Metal film? Watts? Tolerance?
Something cheap...

JR
Visit https://circularscience.com to hear what properly "cleared" drums sound like.

Dr Gris

Re: Noob question about measuring transistors
« Reply #8 on: October 13, 2013, 01:31:42 PM »
Hi fellow Swede,

I also bought some BC184C transistors from eBay some time ago and got a cheap DMM to measure them. Is it just me, or is it easier to find hfe tester on cheaper DMM's?
Anyway, I thought that a cheap DMM with hfe tester might not be so accurate,
but hopefully it would be consistently inaccurate  :D

I wrote the measurement for each one on a little piece of tejp and attached to the transistors,
it took some time, but now it's done...

I guess you know where hi hfe matters the most, if not it's easily found searching this forum.
By that I mean that a BC184C with less than a hfe of 600 is stil usable.

Best
Magnus

Unit7

Re: Noob question about measuring transistors
« Reply #9 on: October 13, 2013, 02:16:02 PM »
Awesome info John! Can't thank you enough. I'll check the current the transistors get in David's NV73 and try to get the tester together accordingly.

And Magnus, thanks! Yes, that's exactly what I'm planning to do too. And yes, I've seen the recommendation of HFE 600+ at the first transistor in all three stages in the 73 design. I've studied David's NV73 schematic and will also double check with David before completing the build


Unit7

Re: Noob question about measuring transistors
« Reply #10 on: October 24, 2013, 04:40:18 PM »
I just got the few parts to make my simple BC184C tester. Found a pic that shows which leg is which on the BC184C. But I'm not sure if it's shows the transistor with legs pointing at you or away from you. Grateful for info about this little detail!

Edit: Just realized that I'm not 100% sure what I should measure?.. Ampere?
Edit 2: Ha! Found it on the NV73 PCB! It's marked C B E. And pic below shows transistor with legs pointing at you

Cheers
Paul
« Last Edit: October 24, 2013, 05:39:32 PM by Unit7 »

Unit7

Re: Noob question about measuring transistors New
« Reply #11 on: October 24, 2013, 06:00:25 PM »
So here's my BC184C tester  :) 
Set my DMM to mA and it started out at 6.19mA slowly rising after a while to 6.27mA (care to explain why?).
I divided 627 with 0.85 and got 737.6
Am I doing this right? Would this mean this transistor is 737.6HFE?

btw, I got word from David (NV73) that the transistors in his design is getting 'in the low mAmperes' so I guess this setup, feeding 2.1mA, would be close enough.
« Last Edit: October 24, 2013, 06:09:18 PM by Unit7 »