transistor letters

[Zorac]

i have an amp i am currently repairing, and with a lot of help from a fellow forum member, the problem seems to be one of the transistors in the control circuit (power supply). since there aren't a lot of transistors in that circuit the plan is to just replace them all, they are:

2sa949(o,y) quantity 1
2sa970(gr,bl) quantity 1
2sc1815(o,y) quantity 2
2sc2240(gr,bl) quantity 1
tlp508(v) quantity 1 (unlikely this is the culprit, but might as well replace it anyway if i can find it(not available at newark))

my question is the only store that i can find them all at online (ie www.newark.com) have transistors that are close, but not exact for example, i need a 2SC2240(gr,bl) but and only find a 2SC2240 or, i need a 2sc1815(o,y) but can only find a 2SC1815-Y. can anyone help me out with these letters after the transistor number? i have been told they are somewhat interchangeable as they deal with how the leads are positions? google hasn't been any help to me on this one...

thanks!!

[poynton]

The letters after the transistor name are gain bandings.

gr=green
bl=blue
o=orange
y=yellow

this may give you some idea :-

http://www.rfparts.com/tranmatch.html

Otherwise they are the same transistor.

Andy

[jacco vermeulen]

If the schematic mentions two hFe classes it means you can use either one.
Also means that you are not supposed to use a device with an hFe class other than those mentioned.
So, either O or Y grade, and either GR or BL grade.
The O and Y hFe grades overlap eachother, so do the highest GR and BL grade.

O is up to 140, Y lower than 240, GR can go as high as 400, BL devices are the best and can have hFe values as high as 700.
TLP508 reads as a LED optocoupler, those come in diferent grades too.

[DreadPirate]

Why not test them and only replace the bad one? This little guy won't set you back much and does the job.

http://www.harborfreight.com/cpi/cta...emnumber=90899

[Zorac]

Originally posted by DreadPirate
Why not test them and only replace the bad one? This little guy won't set you back much and does the job.

http://www.harborfreight.com/cpi/cta...emnumber=90899

actually i have fluke 87 that i bought for fixing this amp, and is how i have narrowed it down this far. ill pull them all off the circuit board and test them. it seems to be a relativly simply procedure to measure the resistance over various leads.

then we can sort out the correct class. considering the price of the transistors, ill still probably replace what i can, but it would be nice to confirm this is the problem.

what i dont get is when i find the transistors listed online, they dont list a hFe class for them, if they dont list a class, does it mean they dont have them and they are wrong? or would i have to track down the manufuactors specs for it to find the correct class?

[AndrewT]

Hi,
very few vendors tell you what hFE class their stock is.
The exception are the BC5xx series that are commonly specified with suffix b or c.

The datasheet for each transistor will tell if the transistor is available with pre-selected hFE grades and what letter codes are used.
Many amplifiers will work with any grade in any position, but some specifications can be improved if a more precise gain is specified.

[DreadPirate]

Originally posted by Zorac


actually i have fluke 87 that i bought for fixing this amp, and is how i have narrowed it down this far. ill pull them all off the circuit board and test them. it seems to be a relativly simply procedure to measure the resistance over various leads.

then we can sort out the correct class. considering the price of the transistors, ill still probably replace what i can, but it would be nice to confirm this is the problem.

what i dont get is when i find the transistors listed online, they dont list a hFe class for them, if they dont list a class, does it mean they dont have them and they are wrong? or would i have to track down the manufuactors specs for it to find the correct class?

I also have a Fluke 77, but this little nifty meter for $2.99 allows you to instantly test a transistor, just stick the pins in the holes. You have to know if it is pnp or npn ahead of time. Alot quicker than using a plain diode tester or ohmeter.

[Zorac]

Originally posted by DreadPirate
I also have a Fluke 77, but this little nifty meter for $2.99 allows you to instantly test a transistor, just stick the pins in the holes. You have to know if it is pnp or npn ahead of time. Alot quicker than using a plain diode tester or ohmeter.

i followed this procedure to test the transistors:

With the meter set to measure ohms, clip one meter lead to the base connection of the transistor.
Touch the other lead first onto the collector lead and then onto the emitter lead.
The readings should both be the same, either both high resistance or both low resistance.

Now reverse the leads and repeat the procedure.
The results should be the opposite of those obtained before.
If they were both high before they should now be both low.
If they were both low before they should now both be high.

Now measure the resistance between emitter and collector.
It should read high resistance in both directions.

all of them seemed to check out. resistance on the leads that provided one were in the 3 to 6 M ohm range readings from the c and the e leads were within 0.8 M ohm on any given transistor.

i though for sure this was the problem, oh well back to trouble shooting (and soldering all these transistors back in! )

[poynton]

Before you do that, post a circuit diagram and the symptoms.
Someone may be able to help.

Andy

[AndrewT]

Hi,
Don't solder them back in yet.
Have you checked all six combinations of the two leadout test?
C-E, E-C, C-B, B-C, B-E, E-B.
Have you got a 10k to 100k resistor and a 1k0 (or nearby)?
You can check the transistor gain with these and a 9V battery.