So you need more voltage. Its trial and error I'm afraid because the filament doesn't have a constant resistance. The hotter it gets, the higher in ohms it will read. It looks like you are going to need somewhere around 470 to 560 ohms at a guess.
Ok. Is there another way we can tackle this? Get a lower voltage bulb? The reason I say this is the 750 is on the circuit board under the tone board. And I don't want to pull tone board because of the many wire wraps attached. Maybe a 4 or 5 V bulb? Would it still need to be 55 mA?
A different bulb could be very hit and miss just as this one is.
You have a relatively high 40 volts AC feeding into a 750 ohm. If we short the bulb out then 53 ma would flow. If we say the original bulb really does have (lets say) 6 volts across it then that means we have (40-6)/750 or 45 milliamps flowing.
The life of a bulb varies with voltage as well. An 8 volt bulb on 8 volts might be rated at 500 hours. Go to 9 volts and it could drop to just 100 hours. Go to 7 and it might be 1500 hours. Make that 6 and it could be 6000 hours. It varies something like that.
(If you can access the resistor from above then you can solder another resistor across the top of the 750 ohm)
Another possibility if you have easy access to the bub holder and so on might be to use an LED. That would need a much higher value resistor (so add one in series with the wire to the holder) and it would also need a small diode to protect against reverse polarity. On the up side it would last forever and all run cool.
You have a relatively high 40 volts AC feeding into a 750 ohm. If we short the bulb out then 53 ma would flow. If we say the original bulb really does have (lets say) 6 volts across it then that means we have (40-6)/750 or 45 milliamps flowing.
The life of a bulb varies with voltage as well. An 8 volt bulb on 8 volts might be rated at 500 hours. Go to 9 volts and it could drop to just 100 hours. Go to 7 and it might be 1500 hours. Make that 6 and it could be 6000 hours. It varies something like that.
(If you can access the resistor from above then you can solder another resistor across the top of the 750 ohm)
Another possibility if you have easy access to the bub holder and so on might be to use an LED. That would need a much higher value resistor (so add one in series with the wire to the holder) and it would also need a small diode to protect against reverse polarity. On the up side it would last forever and all run cool.
I thought about going w an LED. And have read about the need for resistor and diode for ac feed. But I could bridge the 750 ohm from the bottom of the board. With 44 vac, what would it take? The 2.2k you mentioned before?
You would have to try different values to get a suitable brightness. For permanent use you would need a 1 watt resistor for the 2k2.
Looking at this from another angle... you have only a couple of volts across the bulb as it stands now, and so most of the 40 volts is across the resistor. That means around 53 milliamps is already flowing and yet the bulb only drops a couple of volts at that current. That suggests the bulb isn't of similar spec to the original. It may be rated 8 volts but it can't be '8 volts @ 55ma' like the original.
Looking at this from another angle... you have only a couple of volts across the bulb as it stands now, and so most of the 40 volts is across the resistor. That means around 53 milliamps is already flowing and yet the bulb only drops a couple of volts at that current. That suggests the bulb isn't of similar spec to the original. It may be rated 8 volts but it can't be '8 volts @ 55ma' like the original.
I thought about the replacement bulb spec too. So I bought 2 when I placed order and just hooked the second one up to 4 D batteries (6V DC) and I get the brightness I want (it would be a bit brighter at 8 V). But with AC current, the bulb isn't producing desired results with 750 ohm and 42 VAC. So would a different 8V bulb also rated at 50-55 mA produce different results?
The new bulbs have a resistance of 9 ohms.
The new bulbs have a resistance of 9 ohms.
So what current (DC now) are you seeing when you connect to four D cells. The cold resistance is always low for a filament bulb. Taking 9 ohms at face value the bulb draws 8/9 which is nearly 900ma.
The bulb in your link might be OK but you would only know by trying. Ideally you want something that's rated in the 1000's of hours lifetime as well.
This is why a mains bulb tester works as it does and why its so good. When cold the filament allows current to flow and power the amp. Providing the current is low the resistance of the filament stays low and the amp gets a decent voltage. If something is wrong, the amp draws current and the bulb filament gets hotter... its resistance increases and the current limits at a low value with most of the voltage dropped across the bulb.
The bulb in your link might be OK but you would only know by trying. Ideally you want something that's rated in the 1000's of hours lifetime as well.
This is why a mains bulb tester works as it does and why its so good. When cold the filament allows current to flow and power the amp. Providing the current is low the resistance of the filament stays low and the amp gets a decent voltage. If something is wrong, the amp draws current and the bulb filament gets hotter... its resistance increases and the current limits at a low value with most of the voltage dropped across the bulb.
So you think those bulbs are not 50 mA but closer to 900?
Well I'm over 5 hours into my playing and no cutouts. So it looks like the diff pair were failing. Given that and I have a couple more 2N5401s should I replace the right channel pair too before sending this amp back to its owner? I'm not going to replace the Vas trannies as they appear to be fine.
Well I'm over 5 hours into my playing and no cutouts. So it looks like the diff pair were failing. Given that and I have a couple more 2N5401s should I replace the right channel pair too before sending this amp back to its owner? I'm not going to replace the Vas trannies as they appear to be fine.
No, the cold resistance calculates as 900ma given that its an 8 volt bulb and you actually measure 9 ohms. What actually happens in use is that as the filament heats and glows red/white hot then the resistance increases dramatically. So just couple it up to the 6 volt battery and stick your meter on a suitable current range and place it in series with the bulb. That will give the 'hot' current at 6 volts. You can then calculate the 'hot' resistance. R=V/I
Sounds like a result with the diff pair
I always like channels to match it must be said, even though it wont actually sound any different. Up to you 🙂
With any repair like this, make sure that you give it an extended soak test. It's all pretty conclusive that it was the diff pair... but be 100% sure by giving it an extended run. That's the former 'bench tech/workshop supervisor in me coming out now.
Sounds like a result with the diff pair
I always like channels to match it must be said, even though it wont actually sound any different. Up to you 🙂With any repair like this, make sure that you give it an extended soak test. It's all pretty conclusive that it was the diff pair... but be 100% sure by giving it an extended run. That's the former 'bench tech/workshop supervisor in me coming out now.
After 10+ hours of playing, the speakers cutoff during dinner. Walked into the room w stereo and they came back on and I could also hear static on right speaker. Didn't notice it on left side. Shut things down at that point and about an hour later I got a chance to check DC offsets. Both are about 2-5 mV. left side fluctuated only +- 1 mV, but right is more like +- 3-4 mV. Looks like I do need to replace the right side diff pair as well.
Last edited:
That would be the logical thing to do. Odd to have a genuine problem on both channels though... but... these things happen.
Replaced Q2 and Q4. Have both channels dialed in steady at 0v/30mV. Seeing .5 to 1.5 mV at .5s on both channels. Playing with no static.
Been playing it for about 5 hours now. Just turned down volume and checked DC offset. Left: -11 mV, Right: -4 mV.
When offset creeps up like this, is it necessary to keep resetting bias/idle current?
When offset creeps up like this, is it necessary to keep resetting bias/idle current?
So its looking good then. Those changes in offset are miniscule in the scheme of things, in fact they are darned good. Historically, offsets of -/+100mv were considered quite acceptable some years ago. So you have no worries there.
Now its true that a change in DC offset can make the bias current appear to change... what actually happens is that the bias current becomes unequally shared in the two 0.5 ohm emitter resistors because the DC offset actually causes a current to be drawn through the speaker. That unbalances things a little and is the reason why bias current should be set with no load attached.
For your amp and offset voltages its a non issue. Set and forget.
Now its true that a change in DC offset can make the bias current appear to change... what actually happens is that the bias current becomes unequally shared in the two 0.5 ohm emitter resistors because the DC offset actually causes a current to be drawn through the speaker. That unbalances things a little and is the reason why bias current should be set with no load attached.
For your amp and offset voltages its a non issue. Set and forget.
OK, got it. On a related note, in the Service Manual, the instructions for setting bias/idle current (Pages 17 & 18) indicate that one should apply a 5.1k ohm resistor across the Power In inputs on the back of the amp. I have fashioned a couple of RCA jacks with 5.1K resistors and do use them when checking/setting voltage. Is this to mimic a load of sorts or what?
Erm...
My thoughts when I see something like that is that it is more to dissuade folk from attempting the adjustments as much as anything else. The conditions for setting offset and bias are nominally done under 'no signal' conditions. Just turning the volume to minimum is quite good enough.
I can understand terminating the power input if you have the links removed simply as a means of stopping stray hum and noise pickup. Why 5.1k though ? You would normally just use a shorting plug and be done with it. The amp is AC coupled at the input and so the resistance used can have no effect on bias currents of the diff pair.
Slightly odd and weird procedures like that are all to common in older service manuals 😉
My thoughts when I see something like that is that it is more to dissuade folk from attempting the adjustments as much as anything else. The conditions for setting offset and bias are nominally done under 'no signal' conditions. Just turning the volume to minimum is quite good enough.
I can understand terminating the power input if you have the links removed simply as a means of stopping stray hum and noise pickup. Why 5.1k though ? You would normally just use a shorting plug and be done with it. The amp is AC coupled at the input and so the resistance used can have no effect on bias currents of the diff pair.
Slightly odd and weird procedures like that are all to common in older service manuals 😉
Ah Mooly, you have such great reasoning and insight ;-) Amp is still playing this afternoon with no cutouts...
- Status
- Not open for further replies.