You are correct if you choose the 3 separate PCB modules kit - progress already 
However, it's a product that is essentially the same as suggested but is spread over 3 boards, needing extra wiring and more care with twisting supply and signal pair leads together - not always good regardless of experience because its easy to screw up and make hum and noise worse rather than better or so-so. Its not impossible to do spread over more boards, but you'll need more help with wiring dress or finding illustrations, if that's what it takes.
I would stay with a good single board module, where few mounting pillars are necessary and the the layout is good such that the rectifier is some distance from signal leads and traces. This should be cheaper but with no sound quality loss, fitted just as it comes. It's your choice of course, but KISS is best for a first rebuild project.
What some kit designs designs offer in good layout or component quality, others may lack because the money is spent on features with eye appeal or useless "benefits", simply for more sales. Start with a single, not too cluttered board unless there are compelling installation reasons not to.
However, it's a product that is essentially the same as suggested but is spread over 3 boards, needing extra wiring and more care with twisting supply and signal pair leads together - not always good regardless of experience because its easy to screw up and make hum and noise worse rather than better or so-so. Its not impossible to do spread over more boards, but you'll need more help with wiring dress or finding illustrations, if that's what it takes.
I would stay with a good single board module, where few mounting pillars are necessary and the the layout is good such that the rectifier is some distance from signal leads and traces. This should be cheaper but with no sound quality loss, fitted just as it comes. It's your choice of course, but KISS is best for a first rebuild project.
What some kit designs designs offer in good layout or component quality, others may lack because the money is spent on features with eye appeal or useless "benefits", simply for more sales. Start with a single, not too cluttered board unless there are compelling installation reasons not to.
So, some progress.
My over confidence got the better of me and I ended up going for the 3 seperate board design.
On reflection, you were right, Ian, I should've listened and it was more faff than it needed to be. The only upside I can see, is that reusing the existing heatsink was made simpler by being able to position the individual amp boards using holes already drilled by CA rather than having to butcher it with my own drill. The exuberance of inexperience!
This is the state of the wiring as it currently stands (1337 MSPaint skills for easier decyphering)
You'll note the lack of speaker grounding... I wasn't sure whether I ought to connect the speaker ground to the G[out] on the amp board (presumably that then eventually grounds back at the chassis ground point near the input phono connectors); directly to the input chassis ground; or to the power ground as they were (if I've read the PCB tracing correctly?) on the original CA PCB design ?
Other than that, before I turn it all on and have something go pop, does the wiring look correct?
(not shown are the T4A inline fuses between the toroid and the AC inputs of the PSU)
The toroid secondaries are twisted into pairs, as are each DC+/DC- , signal+/signal.gnd and the Amp<->PSU grounds. I've done my best to keep power and ground pairs at the bottom and over towards the toroid as far as possible, and the signal pairs at the top and nearer the heatsink
My over confidence got the better of me and I ended up going for the 3 seperate board design.
On reflection, you were right, Ian, I should've listened and it was more faff than it needed to be. The only upside I can see, is that reusing the existing heatsink was made simpler by being able to position the individual amp boards using holes already drilled by CA rather than having to butcher it with my own drill. The exuberance of inexperience!
This is the state of the wiring as it currently stands (1337 MSPaint skills for easier decyphering)
You'll note the lack of speaker grounding... I wasn't sure whether I ought to connect the speaker ground to the G[out] on the amp board (presumably that then eventually grounds back at the chassis ground point near the input phono connectors); directly to the input chassis ground; or to the power ground as they were (if I've read the PCB tracing correctly?) on the original CA PCB design ?
Other than that, before I turn it all on and have something go pop, does the wiring look correct?
(not shown are the T4A inline fuses between the toroid and the AC inputs of the PSU)
The toroid secondaries are twisted into pairs, as are each DC+/DC- , signal+/signal.gnd and the Amp<->PSU grounds. I've done my best to keep power and ground pairs at the bottom and over towards the toroid as far as possible, and the signal pairs at the top and nearer the heatsink
So, I went for the simple option and connected speaker grounds to the GND Out on the amp boards, turned it on and there is absolutely no hum even turning the amp up to full so huzzah.
Downside is, there's still a gremlin in the system as the left channel still doesn't work (swapping the left and right signal wires from the tone board output swaps the problem so I'm happy it's not a problem with either amp or signal wires and swapping the left and right phono ins had no effect)
Can anyone suggest a likely culprit for this annoyance? The original amp chip was nicely toasted and had scorched the nearby capacitors, so I was wondering if whatever signal overload sent that on its merry way also took out the opamp on the tone board?
Looks like I need to invest in a multimeter, what would be the best diagnostic process to identify the problem?
Downside is, there's still a gremlin in the system as the left channel still doesn't work (swapping the left and right signal wires from the tone board output swaps the problem so I'm happy it's not a problem with either amp or signal wires and swapping the left and right phono ins had no effect)
Can anyone suggest a likely culprit for this annoyance? The original amp chip was nicely toasted and had scorched the nearby capacitors, so I was wondering if whatever signal overload sent that on its merry way also took out the opamp on the tone board?
Looks like I need to invest in a multimeter, what would be the best diagnostic process to identify the problem?
Your wiring diagram looks right.
Connecting the speaker return to the PCB speaker GND is also correct.
Carefully check voltages on the amp PCB +ve to PSU zero volts and -ve to PSU zero volts.
Buy a DMM that has scales going down to 199.9mVdc and 199.9mVac and up to 600Vac and 1000Vdc
Check the voltages on the amp PCB outputs. Both AC and DC. You may get a tiny non zero reading for both AC and DC.
Connecting the speaker return to the PCB speaker GND is also correct.
Carefully check voltages on the amp PCB +ve to PSU zero volts and -ve to PSU zero volts.
Buy a DMM that has scales going down to 199.9mVdc and 199.9mVac and up to 600Vac and 1000Vdc
Check the voltages on the amp PCB outputs. Both AC and DC. You may get a tiny non zero reading for both AC and DC.
Both amps work fine, if they take a signal feed from the R+ and Rgnd from the preamp/toneboard, so I figure that the issue lays somewhere on page 5 or 6 of the schematic?
Presumably, I can narrow it down to either page 5 (tone board) or page 6 (input board) by measuring a value (what should I be looking for?) across L+ and Lgnd at the "signal in" on page 5; if there's a reading there, then the problem isn't in the input circuit on page 6 and must lay within the tone board... if there's no reading there then the problem probably lies in the switching circuitry?
Presumably, I can narrow it down to either page 5 (tone board) or page 6 (input board) by measuring a value (what should I be looking for?) across L+ and Lgnd at the "signal in" on page 5; if there's a reading there, then the problem isn't in the input circuit on page 6 and must lay within the tone board... if there's no reading there then the problem probably lies in the switching circuitry?
Hi again and good work getting a new power module to work in place of the old one As you rightly concluded, the problem lies in the signal path of the left channel which on this model is about as simple as possible, based on just 2 halves of NE5532 dual opamps, U1, U2. If you can't even hear low level hiss when phono is selected, the same as the R channel, then the problem will be after the selector - even though these switches, input sockets and solder joints can be troublesome too. Don't forget that the tape switch is in line too and could be duff or corroded, dirty etc. Make sure to jiggle those signal switches gently for signs of life.
As both opamps are shared for the R and L channels, it's unlikely they are toast, so start with the volume and balance pots by setting balance to midpoint. Assuming you have some bits of wire or even better, some crocodile clip leads, make a simple signal tracer by clipping to the VOL pot. centre terminal (the slider contact) and then to the input of the amplifier, preferably with a capacitor of 10-100 nF in series, to avoid thumps from any DC potentials. You'll see this should just duplicate existing wiring by coupling to VOL, but at th BAL pot, it should test further back and so you continue on to the output (pin 7) of U2 and then pin 7 of U1 etc. to find any point of breakdown back to the selector switch.
However, the sliding contact of the BAL control is at max. volume. That will destroy speakers at max power so, assuming the connection between amplifier input and volume pot. proves to be good, then transfer the clip lead to from the amp input to BAL pot slider and so on, leapfrogging back through the preamplifier sections until you strike a problem that is unique to the L channel.
This traces the signal back through the preamp. That's not all, but you can know what to expect by comparison with the right channel. Now, I didn't say this (for safety reasons) but I place a finger on these points briefly because on test, they will be no more than than +/- 15V, assuming I test this voltage to be sure! The resulting hum from my finger tells me that the circuit is working, if not perfect. Still, I recommend that you use a clip lead with the series capacitor included. Anyway, lets see what difficulties you may have with doing this and understanding it, even though it's straightforward in principle at least, for others who've done signal tracing before.
As you rightly concluded, the problem lies in the signal path of the left channel which on this model is about as simple as possible, based on just 2 halves of NE5532 dual opamps, U1, U2. If you can't even hear low level hiss when phono is selected, the same as the R channel, then the problem will be after the selector - even though these switches, input sockets and solder joints can be troublesome too. Don't forget that the tape switch is in line too and could be duff or corroded, dirty etc. Make sure to jiggle those signal switches gently for signs of life.As both opamps are shared for the R and L channels, it's unlikely they are toast, so start with the volume and balance pots by setting balance to midpoint. Assuming you have some bits of wire or even better, some crocodile clip leads, make a simple signal tracer by clipping to the VOL pot. centre terminal (the slider contact) and then to the input of the amplifier, preferably with a capacitor of 10-100 nF in series, to avoid thumps from any DC potentials. You'll see this should just duplicate existing wiring by coupling to VOL, but at th BAL pot, it should test further back and so you continue on to the output (pin 7) of U2 and then pin 7 of U1 etc. to find any point of breakdown back to the selector switch.
However, the sliding contact of the BAL control is at max. volume. That will destroy speakers at max power so, assuming the connection between amplifier input and volume pot. proves to be good, then transfer the clip lead to from the amp input to BAL pot slider and so on, leapfrogging back through the preamplifier sections until you strike a problem that is unique to the L channel.
This traces the signal back through the preamp. That's not all, but you can know what to expect by comparison with the right channel. Now, I didn't say this (for safety reasons
) but I place a finger on these points briefly because on test, they will be no more than than +/- 15V, assuming I test this voltage to be sure! The resulting hum from my finger tells me that the circuit is working, if not perfect. Still, I recommend that you use a clip lead with the series capacitor included. Anyway, lets see what difficulties you may have with doing this and understanding it, even though it's straightforward in principle at least, for others who've done signal tracing before.
As Andrew raised the issue of using a DMM, you should be testing for any unwanted DC voltage and particularly here in the preamp signal path, where it should not be, WRT ground. This was a fully DC coupled amplifier which may seem like a cool idea but is probably the reason for a high level of failures in use.
There are cheapo meters everywhere from ₤5 or so but if you plan on future DIY adventures, get something in the line of a UT 61 which starts to be accurate as well as precise - since these are two different qualities and just having lots of precise digits really means nothing. This is 4½ digit, true RMS reading, very good, well reviewed by pros and is sufficient for the actual needs of even advanced DIYs. https://www.amazon.co.uk/dp/B00L7R1...stripe&pf_rd_t=201&pf_rd_i=B00N9V6GE6&pf_rd_m
Otherwise there are many cheaper, generic branded, general purpose DMMs that will do this job too, with Andrew's caveat that they have at least a low 2V range with 3½ digit display - i.e. they have at least 1 mA or mV resolution if required and a mandatory class II mains safety rating. Automotive testers and cheapos may not have a mains safety rating at all and there have been some spectacular multimeter+idiot accidents at 230VAC
There are cheapo meters everywhere from ₤5 or so but if you plan on future DIY adventures, get something in the line of a UT 61 which starts to be accurate as well as precise - since these are two different qualities and just having lots of precise digits really means nothing. This is 4½ digit, true RMS reading, very good, well reviewed by pros and is sufficient for the actual needs of even advanced DIYs. https://www.amazon.co.uk/dp/B00L7R1...stripe&pf_rd_t=201&pf_rd_i=B00N9V6GE6&pf_rd_m
Otherwise there are many cheaper, generic branded, general purpose DMMs that will do this job too, with Andrew's caveat that they have at least a low 2V range with 3½ digit display - i.e. they have at least 1 mA or mV resolution if required and a mandatory class II mains safety rating. Automotive testers and cheapos may not have a mains safety rating at all and there have been some spectacular multimeter+idiot accidents at 230VAC
I suggested that the DMM should have a 199.9mVac and 199.9mVdc scales.
That is a resolution of 0.1mV on both the AC and DC scales.
I have a range of very cheap DMM and ALL three have this AC resolution. All work well without any rms reading capability.
I also have an rms reading 50000count bench DMM. I can use that to cross check the other meters.
Indeed you did, apologies for assuming you referred to a minimum 2V range that some really cheap meters still have, or such inaccuracy that makes high resolution pointless. It depends on how high your cheapie definition takes you but in flea markets and local Ebay trading here, you do get reasonable quality for DIY at approximately equivalent to ₤15 and there are plenty of autoranging, 4000 count display options and extra features on meters at that price point too.
Even at ₤5.50 you can have the sensitivity but I wouldn't trust the accuracy of all ranges, safety with mains voltages (despite the cat II label) or longevity. Digital LCD Multimeter Voltmeter AC DC Circuit Checker Voltage Tester Buzzer UK | eBay
There are, as far as I can tell, 3 versions of the amp marked as A1
1) The original, designed by Mike Creek, the volume knob is at the right hand end. Good quality, well engineered. Proper heatsinks.
2) The TDA1514 one, with the volume knob in the middle. Not bad, but the TDA1514's can explode if abused. Folded sheet metal heatsink - one of the banes of cost cutting in Cambridge Audio gear.
3) The LM4765 based one. A complete turd. The heatsink is too small, the capacitors are cheap crap, and some of them even have a stupid little EI transformer instead of the toroid. The LM4765 frequently overheats and explodes, burning a hole through the PCB in the process.
I got a broken one of the 3rd kind from Freegle, fortunately the kind with a toroid in it. I designed and built a power amplifier board to go inside it, with a better heatsink. Havent got around to doing a replacement board for the preamp/front panel controls yet.
1) The original, designed by Mike Creek, the volume knob is at the right hand end. Good quality, well engineered. Proper heatsinks.
2) The TDA1514 one, with the volume knob in the middle. Not bad, but the TDA1514's can explode if abused. Folded sheet metal heatsink - one of the banes of cost cutting in Cambridge Audio gear.
3) The LM4765 based one. A complete turd. The heatsink is too small, the capacitors are cheap crap, and some of them even have a stupid little EI transformer instead of the toroid. The LM4765 frequently overheats and explodes, burning a hole through the PCB in the process.
I got a broken one of the 3rd kind from Freegle, fortunately the kind with a toroid in it. I designed and built a power amplifier board to go inside it, with a better heatsink. Havent got around to doing a replacement board for the preamp/front panel controls yet.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.