Mooly,
I installed them after observing the following excerpts from relevant datasheets, first the type (NPN/PNP) and leg configuration. I understand that C1124 is NPN type, thus replaced by BD139, while A706 and TIP32C is PNP type, hence the replacement being BD140.
Before removal, I observe the existing components, and the board has marking for 'E' leg, so that way I know where the emitter is. Comparing to the datasheet I can then find the C, which I wrote down on the board with some marker, so installing the new ones became easier.
I hope this makes sense, please bear with me as I have no formal training in electronics, but generally quite sensible and willing to learn. Thanks so much for taking the time.
I installed them after observing the following excerpts from relevant datasheets, first the type (NPN/PNP) and leg configuration. I understand that C1124 is NPN type, thus replaced by BD139, while A706 and TIP32C is PNP type, hence the replacement being BD140.
Before removal, I observe the existing components, and the board has marking for 'E' leg, so that way I know where the emitter is. Comparing to the datasheet I can then find the C, which I wrote down on the board with some marker, so installing the new ones became easier.
I hope this makes sense, please bear with me as I have no formal training in electronics, but generally quite sensible and willing to learn. Thanks so much for taking the time.
I'm just looking in an old Towers transistor book and it confirms what you show above, so it looks like you are good to go 🙂
I think you installed wrong the right most NPN transistor. Check it again.
As you see, same type transistors are rotated 180 degrees between them. The left NPN is properly placed. Make sure the right NPN is 180 degrees as the left NPN. I think you need to switch C-E pins between them on the right NPN. Don't take my word, but check that please.
As you see, same type transistors are rotated 180 degrees between them. The left NPN is properly placed. Make sure the right NPN is 180 degrees as the left NPN. I think you need to switch C-E pins between them on the right NPN. Don't take my word, but check that please.
Last edited:
Bob, I have the same suspicion, I think the board mislabeled the Emitter position here. Checking the existing piece, it was installed as follows, and I wrote down their leg designation. C leg is in 'E' hole according to the board. So I guess this is one of the case where the PCB is actually wrong?
If that's the case then switching C and E leg would be the right thing to do, as you suggested. Did you get that from reading the schematic?
If that's the case then switching C and E leg would be the right thing to do, as you suggested. Did you get that from reading the schematic?
I'll check the schematic as well later on, but I'm pretty sure that symmetry + the fact that it worked with the original transistors in that position say that you mush switch C-E pins.
That looks good now. Did you make the bulb tester? If not, then keep your fingers crossed, and test the amp.
I'm building a bulb tester. Got all the parts, haven't had the chance to put them together. I'll make it a habit to run every new gear/after mod thru a bulb tester.
Forgot to ask. Did you replace the burnt/wrong value resistors that were on the back side of the pcb?
Havent done that, got tied up with work so progress has been really slow. Should I do that first before firing up the amp?
It's a low-power amp, so it shouldn't matter much, but for future reference, you are using way too much thermal compound. When doing TO-220 transistors such as these, just put a minimal dab of compound on the middle of the transistor, then install. By minimal dab, I mean one with about the size of the screw hole, and a height of less than 1 mm. Usually, just squeezing the dispenser so that it becomes white on the tip, and then pressing it against the transistor will leave enough compound on it.
When you have screwed the transistor on the heat sink, nothing should seep out on the sides, except for some very tiny marbles (so that you can tell that there is compound used).
When you have screwed the transistor on the heat sink, nothing should seep out on the sides, except for some very tiny marbles (so that you can tell that there is compound used).
Last edited:
Thanks for the tip. I'll keep that in mind.
Did you mean low power amp should use little compound and higher power should use more compound?
This is the DBT that I made, for the purpose of testing this Sony. Hopefully it'll test a'many more amps to come. 🙂
https://www.youtube.com/watch?v=_SuNIQc1WbY
https://www.youtube.com/watch?v=_SuNIQc1WbY
What material and what wattage rating for this application, please?
I tried powering the amp up thru the DBT and it shows that the amp is working normally.
More compound will act as a thermal isolator, rather than conductor. This means the transisors will run hotter. On a low power amp such as this, it's not as destructive, because it will never run all that hot to begin with.
What I have locally is:
Metal film 1% 0.5W 390 Ohm and 560 Ohm
Carbon film 5% 2W 470 Ohm
Metal oxide 5% 2W 470 Ohm
Any suggestion? 😀
Also how to determine what wattage rating for a certain application?
Thanks so much.
Get them from ebay.
You have to apply Ohm's Law to determine the power. You need to know two out of four characteristics to determine the other two : voltage, current, resistance, power. Knowing the resistance (470 ohms) you only need the current through the resistor or the voltage drop across it to determine the dissipated power.
You can get these values from the schematic, or run the LTSpice simulation I gave you for worst case scenario. Double/triple the power rating and you're good to go. I think I remember that 0.5W is already overkill so get something that phisically fits there.
Ebay should have most values for cheap.
You have to apply Ohm's Law to determine the power. You need to know two out of four characteristics to determine the other two : voltage, current, resistance, power. Knowing the resistance (470 ohms) you only need the current through the resistor or the voltage drop across it to determine the dissipated power.
You can get these values from the schematic, or run the LTSpice simulation I gave you for worst case scenario. Double/triple the power rating and you're good to go. I think I remember that 0.5W is already overkill so get something that phisically fits there.
Ebay should have most values for cheap.
- Status
- Not open for further replies.