That's good... very good.
Next step then is to measure across one (either) of the 0.47 ohm resistors and turn the pot to give around 8 to 10 millivolts across the resistor.
Lets see if it will do that.
Next step then is to measure across one (either) of the 0.47 ohm resistors and turn the pot to give around 8 to 10 millivolts across the resistor.
Lets see if it will do that.
and it's done! 8mv exactly.
I don't know if it's important but, The light bulb is still connected? (maybe volt drops or something)
The thing I don't get is how it's possible the bias pot was at the wrong side, I thought I measured it right.. (why did you let me set it to max resistance actually?)
I don't know if it's important but, The light bulb is still connected? (maybe volt drops or something)
The thing I don't get is how it's possible the bias pot was at the wrong side, I thought I measured it right.. (why did you let me set it to max resistance actually?)
Brilliant 🙂
Yes it is important (about the bulb) and why around 8mv seemed enough. That's because the current now drawn by the output stage will be pulling the supplies down.
So... I think the next step is to try that channel for real with a speaker.
If you test with a speaker and the bulb then its vital that you don't actually switch the amp on with the speaker connected. Switch on first and then connect the speaker. Keep the volume low (very low) at all times with the bulb.
The reason for the above is that the bulb severely limits current and with a real load connected (the speaker) the supplies could drop low enough to cause the amp to "latch" with the output swinging over to one of the rails.
or
we can test for real on full mains by you refitting the fuse across the bulb. When you do test on full mains be sure to first turn the bias preset back down and then with the amp running on full mains reset the bias this time aiming for around 12 millivolts. Then test with the speaker connected as normal.
Yes it is important (about the bulb) and why around 8mv seemed enough. That's because the current now drawn by the output stage will be pulling the supplies down.
So... I think the next step is to try that channel for real with a speaker.
If you test with a speaker and the bulb then its vital that you don't actually switch the amp on with the speaker connected. Switch on first and then connect the speaker. Keep the volume low (very low) at all times with the bulb.
The reason for the above is that the bulb severely limits current and with a real load connected (the speaker) the supplies could drop low enough to cause the amp to "latch" with the output swinging over to one of the rails.
or
we can test for real on full mains by you refitting the fuse across the bulb. When you do test on full mains be sure to first turn the bias preset back down and then with the amp running on full mains reset the bias this time aiming for around 12 millivolts. Then test with the speaker connected as normal.
If I understand you correctly: If you set the bias with a bulb, you need to redo it without the bulb and with the original fuse in place? Why didn't we set it around 12 volt with bulb?
What about values with bulb?
For example: If in normal operating mode there must be 80mV what will there be with a light bulb? 70mV? or still 80mV
Or do you only use the light bulb test for the first testing purposes, so till you connect the speakers?
What about values with bulb?
For example: If in normal operating mode there must be 80mV what will there be with a light bulb? 70mV? or still 80mV
Or do you only use the light bulb test for the first testing purposes, so till you connect the speakers?
12 millivolts, not volts.
We use the bulb to make sure as best we can that there are no problems.
We use the bulb to make sure as best we can that there are no problems.
The difference the bulb makes to the values is this...
As more current is drawn (higher bias) then the power supply voltages in the amp fall. That can affect the bias current. For example if we set the bias to 8 mv with a bulb then it might be nearer 14 or 15mv without. A lot depends on the actual amp and the circuitry. Yours is particularly supply dependent because it doesn't use something called "a constant current source" which would set a fixed current to flow in the bias generator circuitry irrespective of voltage. Yours use resistors and so more voltage equals more current.
As more current is drawn (higher bias) then the power supply voltages in the amp fall. That can affect the bias current. For example if we set the bias to 8 mv with a bulb then it might be nearer 14 or 15mv without. A lot depends on the actual amp and the circuitry. Yours is particularly supply dependent because it doesn't use something called "a constant current source" which would set a fixed current to flow in the bias generator circuitry irrespective of voltage. Yours use resistors and so more voltage equals more current.
Well Mooly, the left signal sounds really nice haha! YES!
But I was wondering, it was a transistor problem on the left channel... But how is it possible that the transistors measure good (with 2 ohm stand or checking for a short between c and e) still sound horrible?
Edit:
What also was strange:
The setting of the bias potmeter in wich the bulb was at his most dimm. Was the setting in wich the bias had the highest value??
Maybe I made a mistake or something, but I thought that when the bulb is at is most dim, the bias pot has the highest resistance?
So I started him up with the most resistance as you said but I had to decrease the bias instead of increasing it to 12 mV
But I was wondering, it was a transistor problem on the left channel... But how is it possible that the transistors measure good (with 2 ohm stand or checking for a short between c and e) still sound horrible?
Edit:
What also was strange:
The setting of the bias potmeter in wich the bulb was at his most dimm. Was the setting in wich the bias had the highest value??
Maybe I made a mistake or something, but I thought that when the bulb is at is most dim, the bias pot has the highest resistance?
So I started him up with the most resistance as you said but I had to decrease the bias instead of increasing it to 12 mV
Last edited:
Brilliant job. Well done 🙂
The bias pot (and you need to understand basic semiconductor theory now) gives the lowest bias current when its set to maximum resistance. Why ? Because on max resistance the Vbe multiplier Qm13 (as its called) will be turned fully on via Rm51, RM53 and the thermistor. In that state Qm13 is behaving like a "short" and no voltage develops across it. Decreasing the pot resistance turns off the transistor and so it acts a bit like a resistor and allows a voltage to develop across it. That voltage when it exceeds the turn on voltage of the driver and outputs (that's four base/emitter junctions at around 0.7 volts so 2.8 volts in all) brings them into conduction and the current flows. And we set it by measuring the volt drop across the 0.47 ohms 🙂 Simples 😛
Transistors can measure good on basic checks but still have problems.
So what next ? Other channel 😀
The bias pot (and you need to understand basic semiconductor theory now) gives the lowest bias current when its set to maximum resistance. Why ? Because on max resistance the Vbe multiplier Qm13 (as its called) will be turned fully on via Rm51, RM53 and the thermistor. In that state Qm13 is behaving like a "short" and no voltage develops across it. Decreasing the pot resistance turns off the transistor and so it acts a bit like a resistor and allows a voltage to develop across it. That voltage when it exceeds the turn on voltage of the driver and outputs (that's four base/emitter junctions at around 0.7 volts so 2.8 volts in all) brings them into conduction and the current flows. And we set it by measuring the volt drop across the 0.47 ohms 🙂 Simples 😛
Transistors can measure good on basic checks but still have problems.
So what next ? Other channel 😀
So, builded up the right channel also, and it also sounds good!
Now rebuilding the case and buy some new coolpaste I think?
I really want to thank you guys, and especially Mooly for taking time to help some noobie on his way. I think it's a good way of learning the beginning skills and a little understanding of components!
In the future there will be some more projects (little ones, like a strange transistor in my rotel ra-1412).
p.s
The only general question I still have:
When I find a amplifier at the trash or in a basement and I test it with a bulb, and the bulb stays bright then there must be a short circuit right?
But how to find it then?
Amplifier off and check for resistance?
Amplifier on with bulb and check for?
Now rebuilding the case and buy some new coolpaste I think?
I really want to thank you guys, and especially Mooly for taking time to help some noobie on his way. I think it's a good way of learning the beginning skills and a little understanding of components!
In the future there will be some more projects (little ones, like a strange transistor in my rotel ra-1412).
p.s
The only general question I still have:
When I find a amplifier at the trash or in a basement and I test it with a bulb, and the bulb stays bright then there must be a short circuit right?
But how to find it then?
Amplifier off and check for resistance?
Amplifier on with bulb and check for?
Last edited:
Sounds like you have done a good job 🙂
Fresh thermal paste on both sides of the mica washers would be a good idea but be careful as the mica is super fragile. You could always fit new insulating kits and if you did that then always check they are insulated first before switching on.
More info on the bias which is important on any amp. The actual value will probably drift quite a bit with temperature. The circuit didn't give any set up procedure but its normal to finally set it when the amp has been on for 10 minutes or so. That other component on the heatsink is a thermistor and its vitally important its in good thermal contact with the heatsink. If that has thermal paste on it then renew that too.
(Ideally the bias should remain constant from switch on and through all operating temperatures. In practice it should never go much higher than specified at any point in the temperature cycle)
Bulbs staying bright... depends on the equipment and how much current it would normally draw. For something like an amp a bright bulb would indicate a short, probably failed outputs or a short in the power supply somewhere.
Fresh thermal paste on both sides of the mica washers would be a good idea but be careful as the mica is super fragile. You could always fit new insulating kits and if you did that then always check they are insulated first before switching on.
More info on the bias which is important on any amp. The actual value will probably drift quite a bit with temperature. The circuit didn't give any set up procedure but its normal to finally set it when the amp has been on for 10 minutes or so. That other component on the heatsink is a thermistor and its vitally important its in good thermal contact with the heatsink. If that has thermal paste on it then renew that too.
(Ideally the bias should remain constant from switch on and through all operating temperatures. In practice it should never go much higher than specified at any point in the temperature cycle)
Bulbs staying bright... depends on the equipment and how much current it would normally draw. For something like an amp a bright bulb would indicate a short, probably failed outputs or a short in the power supply somewhere.
-So if you have a short, you need to find it with the amplifier off?
What about recapping the ka-1500, if it plays 100 procent do you recommend it to do, or just let it go this way?
What about recapping the ka-1500, if it plays 100 procent do you recommend it to do, or just let it go this way?
Me personally 🙂 I would enjoy recapping the amp (if I was going to keep it) Mid to late 70's is it ? That's nearly 40 years old.
Shorts... I would always do a quick resistance check on outputs etc and then if that showed nothing power up with a bulb and carefully "interpret" the voltage readings I got in the unit.
Shorts... I would always do a quick resistance check on outputs etc and then if that showed nothing power up with a bulb and carefully "interpret" the voltage readings I got in the unit.
Excellent job If you gonna keep this amp throw all caps and put fresh ones! Don't use any cheapo chinese caps but good quality caps such Nichicon Panasonic etc..Recapping is well spended money.!
Cheers!
Cheers!
problem with ka-1500
Hello i have found a ka-1500 in the street and bring it to life again but because of an error it blow the left channel power transistors then after replacing them a worst problem, the insulation of the d525 fail and make a shortcut between the +34 rail and the 0 rail so it blow the rectifier diodes, after repacing them and the power transistors again and the problem started with headphones a low power hum is present in the left channer always if I rise the volume it distorts with the speakers at 0 volume you can hear a wistle or a chopped noise and it distorts much worst, so I replaced all the transistors and capacitors ( elect) and some resistence without succes so im lost and dont know what to do any help will be more than welcome
regards
Max
Hello i have found a ka-1500 in the street and bring it to life again but because of an error it blow the left channel power transistors then after replacing them a worst problem, the insulation of the d525 fail and make a shortcut between the +34 rail and the 0 rail so it blow the rectifier diodes, after repacing them and the power transistors again and the problem started with headphones a low power hum is present in the left channer always if I rise the volume it distorts with the speakers at 0 volume you can hear a wistle or a chopped noise and it distorts much worst, so I replaced all the transistors and capacitors ( elect) and some resistence without succes so im lost and dont know what to do any help will be more than welcome
regards
Max
Hi Max,
replacing all the transistors isn't really the way to go about faultfinding 🙂
The whistling sounds very much like the amplifier is unstable and oscillating. To diagnose this really involves the use of an oscilloscope to see what is going on.
Also designs like this can be quite susceptible to semiconductors of different characteristics being used, and causing problems. The original (and genuine) devices would be hard to get these days and so I'm assuming you have used either generic replacements or tried your luck with ebay (lots of fakes around).
The oscilloscope is the key tool for the job here.
replacing all the transistors isn't really the way to go about faultfinding 🙂
The whistling sounds very much like the amplifier is unstable and oscillating. To diagnose this really involves the use of an oscilloscope to see what is going on.
Also designs like this can be quite susceptible to semiconductors of different characteristics being used, and causing problems. The original (and genuine) devices would be hard to get these days and so I'm assuming you have used either generic replacements or tried your luck with ebay (lots of fakes around).
The oscilloscope is the key tool for the job here.
HI mooley and thanks for the reply the amplifier was working well until the shortcircuit and I had changed all the transistors before so the semiconductors won't seem to be the problem and the right channel works great so I haven't changed any common components just the left side ones
Max
Max
I've no magic answer I'm afraid. If the only equipment you have is a meter then all you can initially do is check that the DC offset is OK (near zero volts at the output) and that the bias current adjusts correctly over a normal range. Those two checks can tell a lot about what is going on.
Are the low value resistor around the output stage and drivers OK ?
Are the low value resistor around the output stage and drivers OK ?
I only have a meter the resistors are OK an there is around 60mv DC at the output. When I change the bias pot it starts to sound better but it still distort I don't know how to mesure the bias current
I'm sorry about my English but it's not my language so that's why I write like a telegram I will try to explain a little more about this amp. As I said I found it in the street I am not a technician but I work in an electronic components store it took me around two weeks to make it work again I had to change all the transistors for it to work again because someone tried to repair it but he did it the wrong way he short both outputs and solder and desolder to many times so the copper layer it starts to wipe out so the repair was very difficult and I think it won't stand to many more repairs. I don't have an osciloscope just a multimeter so first I have changed the output transistors because I think they were wrong because of the distortion then I changed the drivers and then I realize that when I move the bias pot it make the distortion worst or better so I changed the bias transistor and the resistors related to it then I think about a capacitor so I changed them but with no result so I wrote here for help.
Thank you very much for your time
Max
Thank you very much for your time
Max
The bias transistor for the left channel is Qm13. The bias current is measured indirectly by measuring the small voltage across either Rm73 or Rm75, both 0.47 ohms. You then calculate the current using ohms law.
With no speakers attached and with your meter on DC volts you should see the voltage across Qm13 rise and fall as you turn the bias preset. Once the voltage reaches around 2.5 volts or so then the output transistors should begin to conduct and you should see a small voltage develop across the 0.47 ohms.
So the first thing to check is whether the bias voltage across Qm13 varies and what voltage it goes to as you turn the bias higher.
With no speakers attached and with your meter on DC volts you should see the voltage across Qm13 rise and fall as you turn the bias preset. Once the voltage reaches around 2.5 volts or so then the output transistors should begin to conduct and you should see a small voltage develop across the 0.47 ohms.
So the first thing to check is whether the bias voltage across Qm13 varies and what voltage it goes to as you turn the bias higher.
- Status
- Not open for further replies.