Hi, Hope some of you can help me with this.
I have a Marantz PM6010 KI Same PCB as the PM66 PM68 etc. I think.
There were some cracked traces around the rectifier. I've repaired the traces.
When I turn it on it can take a while for the protection relay to switch on, some times it does not, sometimes it's with in 20s. Maybe another trace/joint that I missed.
The main filter caps (C801 and C802) have a bit of a bulge in the top, so I will replace these.
I'm seeing 24-0-24 AC (ideal 28.7VAC) on the secondary before the rectifier and +36.5VDC -48VDC after the rectifier and on the power amp rails (should be +/-38.4VDC). This section is unregulated. I took these measurements with no load on the amp but, it still behaves the same with the speakers connected. When the relay clicks in the readings are the same.
Is this to be expected from these amps, is there a common fault that cusses this or is it what I suspect and caused by the a (probably) damaged rectifier and dead capacitors.
With the power delivery line losses and so forth I'm not surprised by the 24-0-24 and therefore the +36.5VDC. It's the -48VDC that is confusing me.
PSU and power amp right channel attached.
Thanks
Tom
I have a Marantz PM6010 KI Same PCB as the PM66 PM68 etc. I think.
There were some cracked traces around the rectifier. I've repaired the traces.
When I turn it on it can take a while for the protection relay to switch on, some times it does not, sometimes it's with in 20s. Maybe another trace/joint that I missed.
The main filter caps (C801 and C802) have a bit of a bulge in the top, so I will replace these.
I'm seeing 24-0-24 AC (ideal 28.7VAC) on the secondary before the rectifier and +36.5VDC -48VDC after the rectifier and on the power amp rails (should be +/-38.4VDC). This section is unregulated. I took these measurements with no load on the amp but, it still behaves the same with the speakers connected. When the relay clicks in the readings are the same.
Is this to be expected from these amps, is there a common fault that cusses this or is it what I suspect and caused by the a (probably) damaged rectifier and dead capacitors.
With the power delivery line losses and so forth I'm not surprised by the 24-0-24 and therefore the +36.5VDC. It's the -48VDC that is confusing me.
PSU and power amp right channel attached.
Thanks
Tom
Attachments
If you are measuring -48VDC at test point 160, with respect to ground, then it's possible C802 has dried up, lost capacitance, or is open circuit.
It's always a good idea to check the bridge for opens/shorts too.
It's always a good idea to check the bridge for opens/shorts too.
Definitely, measured multiple times and in different palaces.
I seem to have fixed the offset but now both are high: took the main PCB out reworked some more of the joints. Took the main filter caps out and did basic check (as much as you can) with a standard multi meter: check for short, charge and discharge; both about the same, took the bridge rectifier out and checked that, tested ok.
Put the components back on the PCB turned The amp on relay not clicking.
Checked the rectifier bridge. 233Vac mains, 63.3vac across the rectifier (31.5vac and 31.4vac from each TX out to 0v should be 28.7V on both)
Checked with continuity that chassis and PCB GND are connected (which they are).
+42.3Vdc and -42.4vdc out of the bridge rectifier to Gnd.
Q761 & Q762 +42.2V, Q763 & Q764 -42.2V. The heat sinks are getting warm not hot as is the bridge rectifier.
There is 12mVdc to 0V at R769 L751 and R770 L752
I seem to have fixed the offset but now both are high: took the main PCB out reworked some more of the joints. Took the main filter caps out and did basic check (as much as you can) with a standard multi meter: check for short, charge and discharge; both about the same, took the bridge rectifier out and checked that, tested ok.
Put the components back on the PCB turned The amp on relay not clicking.
Checked the rectifier bridge. 233Vac mains, 63.3vac across the rectifier (31.5vac and 31.4vac from each TX out to 0v should be 28.7V on both)
Checked with continuity that chassis and PCB GND are connected (which they are).
+42.3Vdc and -42.4vdc out of the bridge rectifier to Gnd.
Q761 & Q762 +42.2V, Q763 & Q764 -42.2V. The heat sinks are getting warm not hot as is the bridge rectifier.
There is 12mVdc to 0V at R769 L751 and R770 L752
Attachments
I would replace the rectifier caps before you go any further and get to a stable supply. There may be faults further into the circuit which are triggering protection and holding the relays off. When it does power up can you get to the output relays or amp outputs and measure the DC offset? This should be close to zero. If not you have other issues.
I know, since then I took the main PCB out and reworked some of the joints, now the rails are even but higher than expected. +/-42VDC
If you look at Q761 it should be +38.4 on the collector and on Q763 it should be -38.4 on the collector the same as TP157 and TP160. Which is where I originally measure +36.5VDC and -48VDC
Last edited:
Will be doing, just wanted to see, since this are quite common amplifier and PCB, if any one had any ideas.
It was mentioned the rectifier caps are bulging meaning they could have discharged their electrolyte and no longer doing their job. This could be causing the odd readings. I'm assuming tomst wants to get this amp working so it's the best place to start.
Totally analogue, getting the amp working is the plan. Like I said it's a common PCB through multiple Marantz products just the odd change hear and there, and I know that a lot of products have common faults. It would just be nice to get and idea of what it could be based on that so I can test and order parts, Rather than making multiple orders. In the UK it is often just easier and more cost-effective to buy a known suspect part and try it rather than make multiple orders.
The filter caps don't look like they have vented or leaked but they do look close and the protection relay was randomly clicking in. I do suspect another dry joint and maybe a more faulty component's in addition to the PSU filter caps.
The filter caps don't look like they have vented or leaked but they do look close and the protection relay was randomly clicking in. I do suspect another dry joint and maybe a more faulty component's in addition to the PSU filter caps.
So I've found the why the protection relay was not triggering, it was a dry joint between R801 and R802 in the PSU.
12mVDC at the speakers is acceptable on this amp so that was not triggering the protection circuit to go open, the voltage at TP146 was 0.4V not enough to trigger the relay.
TA7317P (QN04) is a Protection Circuit for OCL Power Amplifier and Speaker, this controls the speaker protection relay.
TP150 was about -18VDC, and TP149 was 0V. There was 0V on all three legs of Q801 and + 42Vdc at TP157 and 0V at R802.
Of course it did help when I nudged R801 while testing it and everything clicked in to life then clicked out of life.
Looking at this PCB I'm sure most of the problems with these are going to be from dry joints, they look like they are wave soldered with the minimal amount of solder they can get away with.
Failure designed in?
Re the +/-42 VDC the amp works both rails are equal; do I need to be bothered by a difference of about 3.6V on each rail from spec (+/- 38.4V)?
My guess is no.
I'm going to replace the power supply caps anyway.
12mVDC at the speakers is acceptable on this amp so that was not triggering the protection circuit to go open, the voltage at TP146 was 0.4V not enough to trigger the relay.
TA7317P (QN04) is a Protection Circuit for OCL Power Amplifier and Speaker, this controls the speaker protection relay.
TP150 was about -18VDC, and TP149 was 0V. There was 0V on all three legs of Q801 and + 42Vdc at TP157 and 0V at R802.
Of course it did help when I nudged R801 while testing it and everything clicked in to life then clicked out of life.
Looking at this PCB I'm sure most of the problems with these are going to be from dry joints, they look like they are wave soldered with the minimal amount of solder they can get away with.
Failure designed in?
Re the +/-42 VDC the amp works both rails are equal; do I need to be bothered by a difference of about 3.6V on each rail from spec (+/- 38.4V)?
My guess is no.
I'm going to replace the power supply caps anyway.
Not really. Thru hole solder joints can crack and "dry up" over time due to thermal cycling of components that produce heat. The result is an intermittent connection as you found. It's quite common. Congrats, BTW.
Perhaps you remember the days of CRT televisions that needed an occasional "whack" on the side to keep going.... same thing.
Electrical solder does not have much structural strength and can crack under the stress or shock from varoius sources.
Edit: as for the rail voltages, they are fine. It's an unregulated power supply so their actual value depends on the line voltage in your area.
That +/- 38.4 V is unregulated and probably based on a mains voltage of 230 V which the UK only has on paper but in reality maintained 240 V. Seeing +/-42 V, is hence not surprising. I wouldn't be bothered by a few V of difference as long as it stays below the caps voltage rating. Besides, our and your mains voltage is allowed to fluctuate up to 253 V to stay within code (which with all those solar panels around nowadays is almost a given on sunny spring and summer days, at least on our congested grid).
If I were replacing those caps, I might go up one step in voltage if size would allow it. But I'd probably first put them on my LCR meter and see if capacity and tanδ were still within spec. Those big caps are usually still fine after 25 years.
Wave soldering doesn't allow for much variation. Wave too high and you have a lot of shorts, too low and a lot of unsoldered joints remain that all need to be touched up by hand ($$$). Getting it right the first time is the most economical for the manufacturer.
Last edited:
Schematic says 28.7-0-28.7 VAC.
Assuming this is at 230 V, that's almost a perfect 8:1 primary:secondary ratio.
At 240 V this results in a secondary voltage of 30 VAC.
At 250 V this becomes 31.25 VAC. Rectified - 2 diode drops of 0.6 V leaves you with 43 VDC.
So the measured +/- 42 VDC are very possible and leads me to believe the mains voltage was around 246 V at the time of measurement.
That main PCB looks almost the same as the one in an early 1990s PM40SE that I used to have.
Assuming this is at 230 V, that's almost a perfect 8:1 primary:secondary ratio.
At 240 V this results in a secondary voltage of 30 VAC.
At 250 V this becomes 31.25 VAC. Rectified - 2 diode drops of 0.6 V leaves you with 43 VDC.
So the measured +/- 42 VDC are very possible and leads me to believe the mains voltage was around 246 V at the time of measurement.
That main PCB looks almost the same as the one in an early 1990s PM40SE that I used to have.
Last edited:
Jitter, I wouldn't be surprised if it was the same PCB as the PM40 too
As of now with no load.
234VAC mains into TX.
2 X 31.5VAC out of TX, which measures 62.9VAC over bridge rectifier,
+/-41.8 after rectifier and at the collectors of Q763 & Q761 Which, considering the TX output, is not far off calculated value (+/- 43.6VDC).
This was taken with a 15 year old UNI-T Automotive multimeter, that has been repaired more than once.
I remember reading somewhere the KI Toroidal TX versions normally have a higher VA than the standard TX so the rail measurements are probably about right?
It's working now and sounds as I would expect it to.
Any one got any ideas of reasonably priced LCR meters that go to around 20,000μF or more.
Thanks for the responses
As of now with no load.
234VAC mains into TX.
2 X 31.5VAC out of TX, which measures 62.9VAC over bridge rectifier,
+/-41.8 after rectifier and at the collectors of Q763 & Q761 Which, considering the TX output, is not far off calculated value (+/- 43.6VDC).
This was taken with a 15 year old UNI-T Automotive multimeter, that has been repaired more than once.
I remember reading somewhere the KI Toroidal TX versions normally have a higher VA than the standard TX so the rail measurements are probably about right?
It's working now and sounds as I would expect it to.
Any one got any ideas of reasonably priced LCR meters that go to around 20,000μF or more.
Thanks for the responses
Last edited:
I have a V&A VA520 which goes up to 20 mF (20,000 µF) that I bought in 2015. This is one of several rebrands of which PeakTech 2170 or Voltcraft LCR-300 are others.
Looking at what Uni-T has to offer, the UT622A looks interesting for the price. It has what you want in addition to the normal measurements, i.e. D-factor (tanδ), Q-factor (1/tanδ), phase angle (θ) and ESR. It doesn't support 100 kHz but for those big caps (and many smaller ones), tanδ is usually given at 120 Hz anyway. Capacity measurement goes up to a whopping 100 mF.
I suggest you look into what D/Q/tanδ/ESR mean and what they tell you about the quality of a capacitor or inductor. If you already have some knowledge of the AC-behaviour of capacitors and inductors, then this video is worth watching to find out how an LCR-meter works and what those values mean.
Looking at what Uni-T has to offer, the UT622A looks interesting for the price. It has what you want in addition to the normal measurements, i.e. D-factor (tanδ), Q-factor (1/tanδ), phase angle (θ) and ESR. It doesn't support 100 kHz but for those big caps (and many smaller ones), tanδ is usually given at 120 Hz anyway. Capacity measurement goes up to a whopping 100 mF.
I suggest you look into what D/Q/tanδ/ESR mean and what they tell you about the quality of a capacitor or inductor. If you already have some knowledge of the AC-behaviour of capacitors and inductors, then this video is worth watching to find out how an LCR-meter works and what those values mean.
Last edited: