I am fairly new at this but liked the P3a because it seems so simple and easy to build. There must be a cloud over my head.
Having spent more hours than I care to admit checking, rechecking and then checking again, I can say with some confidence that the resistors are of correct value and the polarity of the capacitors and diode are correct. Until asked, I wont list all the various combinations of transistors tried, but there have been enough that at least one should have worked by now.
After completion, tests using a 12V power supply (and 20 ohm resistors) result in a big drop at input (to about 8.5) and a bigger drop at the other side. Whatever the reading at the amp side of the fuse is exactly the same reading at the output of the amp.
Any advice is welcome,
PR
Having spent more hours than I care to admit checking, rechecking and then checking again, I can say with some confidence that the resistors are of correct value and the polarity of the capacitors and diode are correct. Until asked, I wont list all the various combinations of transistors tried, but there have been enough that at least one should have worked by now.
After completion, tests using a 12V power supply (and 20 ohm resistors) result in a big drop at input (to about 8.5) and a bigger drop at the other side. Whatever the reading at the amp side of the fuse is exactly the same reading at the output of the amp.
Any advice is welcome,
PR
Check the small signal transistors. If you used something other than the BC546's parts that Rod uses (I used MPSA06), the pinouts are different. I experienced the exact same thing. Reverse the transistors (check pinout datasheets online!) and try again. Have fun - it's a good sounding amp when you're done.
PR,
What is your bias pot setting? I just powered up my P3A two days ago and it drew a ridiculous amount of current at low voltage because the bias pot was set all the way at one extreme. This will also be obvious from a fair amount of heat from the output transistors. Once biased per the instructions, the temperature rise is barely noticable to the touch. Checking the voltage across the collector resistors will help confirm this.
For what its worth, I built Rod's variable power supply and it saved my amp from turning into toast the first time I powered it up. I had misread the pinout on the small-signal transistors and had them in backwards. I had misinterpeted the diagrams on the data sheets as being from the top down, rather than the bottom up. Could have been a fatal mistake (for the amp that is).
Rory
What is your bias pot setting? I just powered up my P3A two days ago and it drew a ridiculous amount of current at low voltage because the bias pot was set all the way at one extreme. This will also be obvious from a fair amount of heat from the output transistors. Once biased per the instructions, the temperature rise is barely noticable to the touch. Checking the voltage across the collector resistors will help confirm this.
For what its worth, I built Rod's variable power supply and it saved my amp from turning into toast the first time I powered it up. I had misread the pinout on the small-signal transistors and had them in backwards. I had misinterpeted the diagrams on the data sheets as being from the top down, rather than the bottom up. Could have been a fatal mistake (for the amp that is).
Rory
hopefully
Envision- I am using Fairchild BC546 which have correct pins (got the diagram and made sure)
Rory- bias is at about 1.9K, I am using a very small 12v power supply, I can leave it hooked up for several minutes and nothing gets warm.
I haven't made an led light up in about a week.
Could someone confirm orientation of Q4, 5, & 6? Is it same as Q7, & 8 (b-c-e looking from front of board)
How reliable is testing a transistor using a diode test across b-c and b-e? I feel like I should have blown one by now, but every time I take one out I check it and it seems OK.
BTW I just got an email from Mr Elliot himself and he thinks it is component orientation, so you guys are on the right track,
Thanks for the replies
PR
Envision- I am using Fairchild BC546 which have correct pins (got the diagram and made sure)
Rory- bias is at about 1.9K, I am using a very small 12v power supply, I can leave it hooked up for several minutes and nothing gets warm.
I haven't made an led light up in about a week.
Could someone confirm orientation of Q4, 5, & 6? Is it same as Q7, & 8 (b-c-e looking from front of board)
How reliable is testing a transistor using a diode test across b-c and b-e? I feel like I should have blown one by now, but every time I take one out I check it and it seems OK.
BTW I just got an email from Mr Elliot himself and he thinks it is component orientation, so you guys are on the right track,
Thanks for the replies
PR
Envision- You are correct about the bias not being a factor in this case. I missed the "DC on the output" in the original post. Thanks for pointing that out.
PR- If you are getting a reading on the diode checker, your transistors are probably good. Usually a blown transistor will look like a short, or less commonly, and open circuit.
This may be a silly question, but are you using a dual rail (+12 and -12) supply?
The fact that the LED doesn't light is a bit mysterious. This could possibly be explained by two things. A) The LED may be installed backwards or B) Q3 was installed backwards and your LED is toast.
As I previously mentioned, I originally installed my transistors backwards and the result would have been a burnt out LED had I not been using a variable supply and realized the problem before I got past a couple volts.
I'm not using Rod's board, so I can't confirm the orientation of the transistors. It might be worth your while to track down the data sheets and verify the pinouts.
Rory
PR- If you are getting a reading on the diode checker, your transistors are probably good. Usually a blown transistor will look like a short, or less commonly, and open circuit.
This may be a silly question, but are you using a dual rail (+12 and -12) supply?
The fact that the LED doesn't light is a bit mysterious. This could possibly be explained by two things. A) The LED may be installed backwards or B) Q3 was installed backwards and your LED is toast.
As I previously mentioned, I originally installed my transistors backwards and the result would have been a burnt out LED had I not been using a variable supply and realized the problem before I got past a couple volts.
I'm not using Rod's board, so I can't confirm the orientation of the transistors. It might be worth your while to track down the data sheets and verify the pinouts.
Rory
It's perfect for Bipolar junction transistors! It also is ok (with practice) for some MosFets for checking D-S/G-S shorts or open flyback diodes.
Pictured below is my 3A amp with the Q1, 2, 3 and 9 transistors reversed. The TO-127 units are all facing towards the outputs as shown. You should see the back of the transistors Q 4, 5 and 6 as pictured below.
An externally hosted image should be here but it was not working when we last tested it.
Rory, yes I am using a +/- 12v supply, not sure of the amperage and assume it does not matter much at this stage. It's connected properly, too.
I too wonder about the led, the very first time I powered up, I definitely had something backwards, but the led came on for several seconds before it burned out. since that day, nothing seems to burn out or even get warm. Hmmm...I pulled the led...tested OK and correct polarity. Also re-checked pinouts of Q3- they are correct.
PR
I too wonder about the led, the very first time I powered up, I definitely had something backwards, but the led came on for several seconds before it burned out. since that day, nothing seems to burn out or even get warm. Hmmm...I pulled the led...tested OK and correct polarity. Also re-checked pinouts of Q3- they are correct.
PR
You might want to try measuring the voltage across the LED. If it is oriented correctly along with Q3, I would expect it to light up. The fact that it doesn't might point to a problem with your V- supply.
Something has to give here. The LED is essentially connected between V- and ground through R8. As long as Q3 isn't backwards or blown (shorted), you should have light.
Something has to give here. The LED is essentially connected between V- and ground through R8. As long as Q3 isn't backwards or blown (shorted), you should have light.
Just a thought....
I know this could be a possibility that I have seen other constructors miss. Are resistors R9 and R10 2.7k ohms? I have seen a few people read 2k7 as 2.7 ohms!! (Should be red, purple, red, gold (or whatever tolerance you chose). Heck, I used 2.2k and have no problems, though that's only a 500 ohms difference. This would certainly make Q9 get hot (or explode!), Q4 to get warm and pull down the entire bottom half of the VA stage, causing a large DC offset at the output.
I guess it's possible that since the LED is driven from the negative 12 volt (at this time) supply, a high current draw through Q9 would pull the supply down considerably causing the LED to appear dim. Check all the color codes for all the resistors. I'm not saying you're stupid, I think we've all seen brown bands that look red on resistors, especially five banders. If in doubt, pull one resistor lead out of the board and remeasure with your meter.
Give it a shot - something's gotta give!
I know this could be a possibility that I have seen other constructors miss. Are resistors R9 and R10 2.7k ohms? I have seen a few people read 2k7 as 2.7 ohms!! (Should be red, purple, red, gold (or whatever tolerance you chose). Heck, I used 2.2k and have no problems, though that's only a 500 ohms difference. This would certainly make Q9 get hot (or explode!), Q4 to get warm and pull down the entire bottom half of the VA stage, causing a large DC offset at the output.
I guess it's possible that since the LED is driven from the negative 12 volt (at this time) supply, a high current draw through Q9 would pull the supply down considerably causing the LED to appear dim. Check all the color codes for all the resistors. I'm not saying you're stupid, I think we've all seen brown bands that look red on resistors, especially five banders. If in doubt, pull one resistor lead out of the board and remeasure with your meter.
Give it a shot - something's gotta give!
well I'll be...
Envision, you hit the nail on the head, I misread the number and did it more than once. Boy, do I feel stupid. The voltages are now correct, led lights, about 300mv at output. The closest I thing had on hand was 2.2K, I will change them out when the 2.7k get here.
Hey! UPS just brought my toroidal transformer!
I cant tell you how much I appreciate everyones' help and for sticking with me on this. Happy Holidays,
PR
Envision, you hit the nail on the head, I misread the number and did it more than once. Boy, do I feel stupid. The voltages are now correct, led lights, about 300mv at output. The closest I thing had on hand was 2.2K, I will change them out when the 2.7k get here.
Hey! UPS just brought my toroidal transformer!
I cant tell you how much I appreciate everyones' help and for sticking with me on this. Happy Holidays,
PR
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.