good morning all! sorry i couldn't get back yesterday. bathroom reno happening and SWMBO is anxious, ergo...
ok then.local weather is at 36.8C, blood pressure expected to rise dramatically soon this morning, accompanied with rapid heart beat. forecast is a mixed bag, and looks like it could go a couple different ways
seriously though, the new driver transistors have now been fitted, the original output transistors have been reinstalled. oh my poor poor board. starting to look like poop, compared to pre-aidan.
i've resisted the urge to power up, thinking i'd say hello here first to see if any of you gurus had any pre-ignition advice, or such.
Qe19 the "850" has been replaced with the "5401". the Qe17 "1735" has been replaced with "5551". in both cases, it was necessary to swap the pinouts of the collector and the base.
aidan.
ok then.local weather is at 36.8C, blood pressure expected to rise dramatically soon this morning, accompanied with rapid heart beat. forecast is a mixed bag, and looks like it could go a couple different ways
seriously though, the new driver transistors have now been fitted, the original output transistors have been reinstalled. oh my poor poor board. starting to look like poop, compared to pre-aidan.
i've resisted the urge to power up, thinking i'd say hello here first to see if any of you gurus had any pre-ignition advice, or such.
Qe19 the "850" has been replaced with the "5401". the Qe17 "1735" has been replaced with "5551". in both cases, it was necessary to swap the pinouts of the collector and the base.
aidan.
Are you happy with your work so far
If so then you can attempt to bring the voltage up slowly. The rail voltage should rise, and the bulb should remain out/dim. As you bring the voltage up make sure those new transistors are not getting hot.
If it passes that test then bring the voltage back down, switch off and then reconnect the meter to monitor the voltage between ground and the main output (junction of those 0.47 ohms) and then repeat bring the mains voltage up slowly. This new measurement should settle to a low (preferably near zero) voltage.
If that's OK then bring the voltage back down and switch off.
If so then you can attempt to bring the voltage up slowly. The rail voltage should rise, and the bulb should remain out/dim. As you bring the voltage up make sure those new transistors are not getting hot.
If it passes that test then bring the voltage back down, switch off and then reconnect the meter to monitor the voltage between ground and the main output (junction of those 0.47 ohms) and then repeat bring the mains voltage up slowly. This new measurement should settle to a low (preferably near zero) voltage.
If that's OK then bring the voltage back down and switch off.
OH MY GOD, oh my god, oh my god, oh my god.....
it passed the smoke test. bulb was faint, for just a 1/4 second, then extinguished. at 60VAC was drawing 50 mA. at 115 VAC a very nice 100 mA. this, with no load, and no input.
powered down with variac, shut down.
not sure what you mean by "junction of those 0.47 ohms". junction?
wow, this is nice if it works! mooly, thanks for your patience in this endeavour.
aidan
it passed the smoke test. bulb was faint, for just a 1/4 second, then extinguished. at 60VAC was drawing 50 mA. at 115 VAC a very nice 100 mA. this, with no load, and no input.
powered down with variac, shut down.
not sure what you mean by "junction of those 0.47 ohms". junction?
wow, this is nice if it works! mooly, thanks for your patience in this endeavour.
aidan
So far so good.
An easier way to do the measurement. I don't think this amp has speaker relays (which makes measuring easy) so you can connect your meter directly across the speaker sockets. No speakers attached yet. Bring the voltage up and there should be less than 100millivolts across the speaker sockets. Check that for both channels.
An easier way to do the measurement. I don't think this amp has speaker relays (which makes measuring easy) so you can connect your meter directly across the speaker sockets. No speakers attached yet. Bring the voltage up and there should be less than 100millivolts across the speaker sockets. Check that for both channels.
by the way, which channel are the transistors associated with? how do i tell?
thanks
aidan
If you trace the output from the channel with Qe17 you'll see it goes to the LEFT speaker and headphone socket.
you're right, no relay. and, ahhh, of course it's the left. i saw it just after i posted the question.
okay, no load, no speakers, across left speaker terminal i get .014 vdc. across right channel i read .008 vdc.
will these figures be affected by bias adjustment? holy crow, i'm really getting hopeful here. when do we connect a source and speakers?
aidan
okay, no load, no speakers, across left speaker terminal i get .014 vdc. across right channel i read .008 vdc.
will these figures be affected by bias adjustment? holy crow, i'm really getting hopeful here. when do we connect a source and speakers?
aidan
Its looking good. Power it back down... and what we need to do now is make sure the bias current adjusts OK. We do this once as a trial run and then again as a proper setting because the absolute value (might) depend a little on mains voltage.
To make this easy I recommend soldering wires to the appropriate points. If you don't then you run the risk of shorting something... hmmm been there before I think.
The reading we are interested in is the DC voltage between the emitters of the two output transistors. For now we would adjust the bias to around 20 millivolts (that's just over 20 milliamps flowing in those resistors). You might find the bulb glows a little... that's fine. If it adjusts OK then we bring the mains back down and then set it for real on full mains.
The manual suggests 40 millivolts I see. Lets keep it on the lower side for now.
To make this easy I recommend soldering wires to the appropriate points. If you don't then you run the risk of shorting something... hmmm been there before I think.
The reading we are interested in is the DC voltage between the emitters of the two output transistors. For now we would adjust the bias to around 20 millivolts (that's just over 20 milliamps flowing in those resistors). You might find the bulb glows a little... that's fine. If it adjusts OK then we bring the mains back down and then set it for real on full mains.
The manual suggests 40 millivolts I see. Lets keep it on the lower side for now.
mooly, from each leg of both of the .47 resistors (4 legs), i get .014 on left and .008 on the right side 4 legs, with reference to chassis ground.
aidan
That's fine. That's the DC offset and those values are low even by todays standards. Anything less than 100 mv is OK.
Remember the bias adjustment that we do next is measuring across those resistors.
Q7 required removal of faceplate to locate and identify. fortunately, kenwood made access to this new board a little easier by providing a way, outlined in the service manual, to remove some screws and loosen two others, and then pivot the face out and down.
collector to emitter of Q7 is 7.8 vdc
aidan
collector to emitter of Q7 is 7.8 vdc
aidan
This is what we should have and what we should expect.
Just to recap... the voltage between ground (chassis) and the 0.47 ohms is to be less than 100 millivolts. We've got that.
This is where it falls down...
If you have 7 volts across Qe7 then you have that same voltage between the two bases of the driver transistors. Possible to confirm that ?
Because of the way transistors conduct that also means that this 7 volts (less a little bit because of the base to emitter volt drops) should also be between the bases of the two output transistors. That wouldn't be possible if the output devices were good. Is it possible to check that reading, the voltage between the two bases of the outputs ?
Something very basic is amiss here.
Just to recap... the voltage between ground (chassis) and the 0.47 ohms is to be less than 100 millivolts. We've got that.
This is where it falls down...
If you have 7 volts across Qe7 then you have that same voltage between the two bases of the driver transistors. Possible to confirm that ?
Because of the way transistors conduct that also means that this 7 volts (less a little bit because of the base to emitter volt drops) should also be between the bases of the two output transistors. That wouldn't be possible if the output devices were good. Is it possible to check that reading, the voltage between the two bases of the outputs ?
Something very basic is amiss here.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Solid State
- Kenwood KA-3500 and my blooper