Technics SH-9010 Equalizer Recap—a few questions

While @mbz makes some fine points, one thing to consider is the whole audio chain, from source to speaker. Many people specifically “remove” the tone controls from the audio path to get fewer components, traces, resistors, etc out of coloring the sound or degrading it in some way, noise to be one.

While that overall idea is not in play here since you want to add an EQ into the signal path, IMO it would be wise to keep the noise profile to a bare minimum. And that would mean film or LL caps in the audio path.

Let’s just say your amp/receiver/integrated has a signal to noise ratio of 95dB, and you’re running a CD source with S/N of 105dB. You have just increased the noise level over that of listening to the CD through headphones, if the player even has a jack. Now, you add an EQ to the mix. If it has a S/N figure higher than the amp's 95dB, you SHOULDN'T hear any EXTRA noise, i.e. only how you contour the sound with the sliders. If however, the EQ only has a 90dB S/N, you've just added even more noise to your audio signal, which will muddy the sound further. I’ll look further at the scat and list the audio path only caps, but you should be able to follow the grey arrows and see which ones are directly in the audio signal.
 
470/35V x10 (P.S., Mute, Band)
https://www2.mouser.com/ProductDetail/Panasonic/EEU-FM1V471?qs=sGAEpiMZZMtZ1n0r9vR22bhWmcKR3unnDi0pKS7kuMQ=

47/25V x10 (P.S., Logic)
https://www2.mouser.com/ProductDetail/Panasonic/EEU-FM1E470?qs=sGAEpiMZZMtZ1n0r9vR22V2Y07khHJOUCfGBjp75zd4=

10/25V x2 (P.S.)
https://www2.mouser.com/ProductDetail/Nichicon/UPW1E100MDD?qs=sGAEpiMZZMtZ1n0r9vR22RH2kZvTh%2b0aPNLUvx0KCu4=

220/25V x2 (Band)
https://www2.mouser.com/ProductDetail/Panasonic/EEU-FM1E221BJ?qs=sGAEpiMZZMtZ1n0r9vR22ZGaUoI0JcRfnEDHJr7HIZg=

120/25V x2 (P.S. set higher capacitance since modern trend low on value)
https://www2.mouser.com/ProductDetail/Nichicon/UPW1E121MED1TD?qs=sGAEpiMZZMtZ1n0r9vR22RH2kZvTh%2b0aPr6SbdIUafg=

47/16V x10 (Band) note: these are 10% tol.
https://www2.mouser.com/ProductDetail/Nichicon/UKL1C470KEDANATD?qs=sGAEpiMZZMtZ1n0r9vR22Xou3HnuIKwreNFX4Dpl7i0=

3.3/50V x10 (all locations) extras to be used in future projects
https://www2.mouser.com/ProductDetail/Nichicon/UKL1H3R3MDDANA?qs=sGAEpiMZZMtZ1n0r9vR22cK9fic32uhgdGusy23lGm0=

Alternatively, for C101,102 (input caps, the orange ones) you could splurge with these:
3.3/63V x2 FILM CAPS
https://www2.mouser.com/ProductDetail/WIMA/MKS2C043301K00KSSD?qs=sGAEpiMZZMv1cc3ydrPrF7l45uRd9dVhUVVnxqBvHE0=

Those would be my choices for the Non-Bipolar caps. In the PS you want caps that have low ESR and can handle high ripple current to smooth out the voltage. The 105C temp rating is also good for reducing heat stress on the caps. The Panasonic FM series has some of the best ripple specs out there. The other places where just general filtering is needed also benefit by this. I see no need to put “audio” caps in those locations. You may feel otherwise. Looks like the max voltage in the PS. is ~23V, so your plan to go to 100V caps is WAY overkill. Sticking with the original values is fine here and will enable bulk purchasing without going too far above rated for the lower voltage caps.

C101,102 are I put caps in the audio path which is why they used the orange Low Leakage caps. That's where UKL or film cap will make the most difference. Technic engineers decided to use bipolar in the other audio path caps (which I hadn't caught at first). That's another way to “skin the cat” so to speak. Some testing has shown lower levels of distortion when BP caps are used, and some here like Leestereo will use BP even when polarized are called for in the audio path (can't do the opposite though). So, your choice of the ES Muse series is perfect for those.

Also, as stated in the 120uF choice for PS caps, in those filtering positions you could probably increase all your capacitance values a reasonable amount (680 for 470 for instance) and could probably replace the 220uF filters with 470 from a bulk purchase standpoint.

If you wanted to go all out, you could also replace the resistors in the audio path with metal film which tend to have lower noise than carbon film and also tighter tolerances. However, looking at the specs, the S/N ratio of this EQ is at best 90dB (87dB under a different test), so you'll most likely be adding noise to your signal when you engage it. May not be noticeable during louder portions, but may add a layer of hiss to quieter passages.
 
I see a pile of 16V caps in the BOM, buy side. ConradH as said that the minimum he buys is 25V because they are just better caps. He sees so many small caps out of spec compared to the larger ones his minimum is that 25V cap. This info from the Capacitor Abyss thread.
 
I see a pile of 16V caps in the BOM, buy side. ConradH as said that the minimum he buys is 25V because they are just better caps. He sees so many small caps out of spec compared to the larger ones his minimum is that 25V cap. This info from the Capacitor Abyss thread.
Thanks, but not to worry. If you look at the parts list I posted earlier, you will notice that the replacement caps (in the right columns) all are at least twice the voltage of the original ones.

---

On another note... regarding noise and sound quality with/without EQ. I have made a simple experiment today by using my best digital source, running into my amp and then into my computer to record a song, Then, on a separate track, I recorded the same song again but with the EQ added into the signal path before the amp, EQ set to "off". And I also did a third track with the EQ set to "on", but all sliders centered.

Once all tracks recorded—and this is the important part—I normalized each track to be sure I could compare them at the exact same level. It was obvious here, that the recorded tracks with EQ needed a bit of a push. In other words, with the EQ in the signal path, the level is a bit lower, which may have caused me to believe earlier that the EQ sounded a bit "lifeless".

To compare, I simply needed to start playback and use the "solo" button for which ever track I wanted to hear. I checked with headphones, and speakers as well. The difference was very slight; I could hear just a hint more of "openness" in the non-EQ track, but it was so slight, I am not sure if I could pick out the right track in a blind test.

Just an interesting experiment...

While one may conclude that I don't need to do any recapping at all now, I still want to do it as a sort of maintenance, since the caps are all 40 years old now. Should I use the EQ for longer, and on a regular basis, a recap will certainly be a good thing.

On another note... this may not be my last EQ. After having experimented with open baffle speakers for a year or two, I have something I am very happy with now, especially the bass, which is now an 18" McCauley 6174 in each baffle. It sure needs to be EQ'd, but once you do, bass is just a whole different world compared to box speakers... so I want to stick with this type of speaker.
The Technics SH-9010 does a great job bringing in the right amount of bass, what's a bit awkward though is, that I really just need that—bass. I could live without he other four sliders! So we will see... maybe I find an interesting DIY project one day to just do the necessary bass EQ, which would mean a much simpler signal path altogether.

Time will tell... and surely many more experiments :)
 
470/35V x10 (P.S., Mute, Band)
https://www2.mouser.com/ProductDetail/Panasonic/EEU-FM1V471?qs=sGAEpiMZZMtZ1n0r9vR22bhWmcKR3unnDi0pKS7kuMQ=

47/25V x10 (P.S., Logic)
https://www2.mouser.com/ProductDetail/Panasonic/EEU-FM1E470?qs=sGAEpiMZZMtZ1n0r9vR22V2Y07khHJOUCfGBjp75zd4=

10/25V x2 (P.S.)
https://www2.mouser.com/ProductDetail/Nichicon/UPW1E100MDD?qs=sGAEpiMZZMtZ1n0r9vR22RH2kZvTh%2b0aPNLUvx0KCu4=

220/25V x2 (Band)
https://www2.mouser.com/ProductDetail/Panasonic/EEU-FM1E221BJ?qs=sGAEpiMZZMtZ1n0r9vR22ZGaUoI0JcRfnEDHJr7HIZg=

120/25V x2 (P.S. set higher capacitance since modern trend low on value)
https://www2.mouser.com/ProductDetail/Nichicon/UPW1E121MED1TD?qs=sGAEpiMZZMtZ1n0r9vR22RH2kZvTh%2b0aPr6SbdIUafg=

47/16V x10 (Band) note: these are 10% tol.
https://www2.mouser.com/ProductDetail/Nichicon/UKL1C470KEDANATD?qs=sGAEpiMZZMtZ1n0r9vR22Xou3HnuIKwreNFX4Dpl7i0=

3.3/50V x10 (all locations) extras to be used in future projects
https://www2.mouser.com/ProductDetail/Nichicon/UKL1H3R3MDDANA?qs=sGAEpiMZZMtZ1n0r9vR22cK9fic32uhgdGusy23lGm0=

Alternatively, for C101,102 (input caps, the orange ones) you could splurge with these:
3.3/63V x2 FILM CAPS
https://www2.mouser.com/ProductDetail/WIMA/MKS2C043301K00KSSD?qs=sGAEpiMZZMv1cc3ydrPrF7l45uRd9dVhUVVnxqBvHE0=

Those would be my choices for the Non-Bipolar caps. In the PS you want caps that have low ESR and can handle high ripple current to smooth out the voltage. The 105C temp rating is also good for reducing heat stress on the caps. The Panasonic FM series has some of the best ripple specs out there. The other places where just general filtering is needed also benefit by this. I see no need to put “audio” caps in those locations. You may feel otherwise. Looks like the max voltage in the PS. is ~23V, so your plan to go to 100V caps is WAY overkill. Sticking with the original values is fine here and will enable bulk purchasing without going too far above rated for the lower voltage caps.

C101,102 are I put caps in the audio path which is why they used the orange Low Leakage caps. That's where UKL or film cap will make the most difference. Technic engineers decided to use bipolar in the other audio path caps (which I hadn't caught at first). That's another way to “skin the cat” so to speak. Some testing has shown lower levels of distortion when BP caps are used, and some here like Leestereo will use BP even when polarized are called for in the audio path (can't do the opposite though). So, your choice of the ES Muse series is perfect for those.

Also, as stated in the 120uF choice for PS caps, in those filtering positions you could probably increase all your capacitance values a reasonable amount (680 for 470 for instance) and could probably replace the 220uF filters with 470 from a bulk purchase standpoint.

If you wanted to go all out, you could also replace the resistors in the audio path with metal film which tend to have lower noise than carbon film and also tighter tolerances. However, looking at the specs, the S/N ratio of this EQ is at best 90dB (87dB under a different test), so you'll most likely be adding noise to your signal when you engage it. May not be noticeable during louder portions, but may add a layer of hiss to quieter passages.
I really appreciate your efforts... but I have already ordered what I posted earlier. I did order the UKL caps to replace the orange caps, by the way.
 
Like you said earlier, you’re replacing 40 year old caps...that's a good thing. As you continue with other projects and continue the learning process, you’ll find that there are places where you really do want to use high temp / low ESR caps vs audio caps and so forth. It's fun experimenting and I'm sure the new caps will take the lifelessness away. I bet you even find some of those lavender caps are open like I did. I replace them whenever I find them in gear now based on issues I've seen them cause.
 
Thanks, but not to worry. If you look at the parts list I posted earlier, you will notice that the replacement caps (in the right columns) all are at least twice the voltage of the original ones.

Won't make much difference for decades but if Conrad is not going to install anything under 25V, I'm following that advise. His testing of removed caps shows a significantly higher out of spec situation will the caps below 25V ratings. He indicated that this was where the caps started to show similar out of spec to the higher voltage caps. So if the less that 25V caps aren't that good and in general the 25V caps cost the same as the lesser caps, I'm getting the big boys (they are also about the same size). I generally look at 25-50V caps for these and combine to get buy 10 pricing and find that higher voltage may be cheaper at times.
 
While one may conclude that I don't need to do any recapping at all now, I still want to do it as a sort of maintenance, since the caps are all 40 years old now. Should I use the EQ for longer, and on a regular basis, a recap will certainly be a good thing.

depends on the make and hours on the caps. boxes are usually sealed so corrosion should be minimal. shot gunning without performance test is wasteful. imo.
 
shot gunning without performance test is wasteful. imo.

Perhaps, but removing a cap for testing and then reinstalling it, when it's well past its design life is a waste of time if you ask me. Besides, I've found those purple caps to be the direct cause of several pieces of gear not working properly or even at all. H/K EQ8, PM655, Yamaha CA-810. So now, when I see them in gear, they get replaced even if I'm NOT intending to do a recap.
 
Started with the power supply this afternoon. I could say it's done... but I did actually forget to order one cap! And I thought I was being very thorough :)

Also put in that one cap by the muting relay. The "pop" noise when turning on/off has not disappeared though. Should I assume the relay itself is bad? I'd love to get rid of that issue, but who knows if this type of relay is still available.

Anyway, first caps are in now—out the lavender, in the black and gold... It appears to be a little quieter now, actually.

Measured the voltage and it comes out to +17.3 and -17.3. Is it worth the trouble to try bringing it exactly to ±17.0? I heard these older trim pots are difficult to set.
 

Attachments

  • R0015636.JPG
    R0015636.JPG
    154.5 KB · Views: 26
  • R0015638.JPG
    R0015638.JPG
    130 KB · Views: 25
Looks good. Yeah, they are cheap and nasty trimmers however this is not the project for trimmer replacement (20+ x VR's!)
+/-17.3V should be fine, there is always some mains AC bounce. Main point is that it's reasonably stable
 
Looks good. Yeah, they are cheap and nasty trimmers however this is not the project for trimmer replacement (20+ x VR's!)
+/-17.3V should be fine, there is always some mains AC bounce. Main point is that it's reasonably stable
Sounds good.

By the way, tested the removed caps, they were all within spec, mostly a bit too high. Don't know about the larger ones, my meter can only test up to 400µF. Those looked a bit bulged on the bottom.

By the way, should I worry about the "gooey" stuff on the circuit board? I did my best to keep everything clean, although cleaning the pads also smeared up the glue. Why was it put there anyway?
 
The glue is used to stabilize the larger caps for the wave soldering process. Some formulations of it turn corrosive over time and can cause issues with components, eat through jumpers and/or component leads etc. Since, as hobbyists and not large manufacturers, we're doing work by hand and usually not shipping all over the world, it's best to remove as much of the glue as you can before installing new caps. 90% isopropyl alcohol or acetone usually soften the glue enough to use a scraper to remove it easily. I use a flat Xacto knife blade that looks somewhat like a screwdriver end to remove it, but anything that can get it out works.
 
Well, here it is now, 55 out of 56 electrolytic capacitors replaced :)
I went through the trouble to measure all µF values of the old ones... only one was completely off, about 50% below its rating, all others were surprisingly high, at least 20% over, often rather 30%. Is that what's to be expected when they get old?

I will still have to get that one that I forgot in my original order... to make the recapping complete.

I did my best to clean up the glue around the area where I soldered, but scraping it off the entire board would have been difficult.

There are four caps in there where I forgot to bend the leads... the caps have 3.5mm lead spacing, the holes are spaced out 5mm. I am hoping this won't cause any trouble in the long run.?

You are probably wondering how it sounds now... well, I would have to rely on my "audio memory". At first I thought this wasn't gonna be possible... but I have to be honest, I really do believe it sounds better now. I am using it just for the low frequencies, and here I am quite sure I can hear an improvement, it just sounds more "fat", seems to have more authority than before. It's working really well now for compensating for the bass loss in an open baffle woofer.

So all is good and well, but here is the thing ...
The manual states that one should calibrate the unit after replacing capacitors or resistors, which makes sense of course. Problem is, I don't have any measuring equipment apart from my Amprobe 37XR-A multi-meter :confused:

According to the manual, I need two voltmeters, a signal counter, and an oscilloscope, and a scope. But then they only provide a table with voltage values that should be matched using the trim pots... so I wonder, would I really need all those tools for this?

If you can give me any clues on how to calibrate it... maybe with simpler tools...that would be great.
 

Attachments

  • R0029979.JPG
    R0029979.JPG
    140.5 KB · Views: 26
Though not quite the same, I would treat it like recapping a tuner....w/o the correct test equipment, don't fiddle with the adjustments. Some say not to even bother with recapping tuners, however, if the caps are out of spec, recapping should, in theory, get it back closer to what it was like when originally aligned. So, with your EQ, unless you can actually perform the procedure that's outlined, just leave well enough alone. Check voltages where specified, but I wouldn't try to really adjust them. YMMV.
 
There is more to checking caps than just capacitance. I wonder how the ESR and leakage would be on the ones checking 30% high.
 
Though not quite the same, I would treat it like recapping a tuner....w/o the correct test equipment, don't fiddle with the adjustments. Some say not to even bother with recapping tuners, however, if the caps are out of spec, recapping should, in theory, get it back closer to what it was like when originally aligned. So, with your EQ, unless you can actually perform the procedure that's outlined, just leave well enough alone. Check voltages where specified, but I wouldn't try to really adjust them. YMMV.
I can check the power supply voltage, but for any of the other measurements described in the manual, I'd need several other tools.

I will just leave it as is for now. After all, if a slider should be a little off, it would probably not even make a difference—they don't have detents, so to set e.g. the left and right 60Hz sliders to the exact value is just possible by eye-balling anyway.

There is more to checking caps than just capacitance. I wonder how the ESR and leakage would be on the ones checking 30% high.
I would have liked that too, as this appears to be more interesting anyway. But I simply don't have the appropriate tools for that.
 
Actually, I was able to measure the "DC Unbalanced Voltage" as well. It measured -7.3 and -4.8 mV. I was able to adjust them to about 0.8mV, after making sure the unit ran for a little while. The pot is a tricky one to set, but not terribly difficult either.
 
was thinking of pulling the trigger on one of these(150)...doesn't seem worth the overall expense,work and sound;if the overall sound db are being decreased for the convenience of tailoring it. Thanks for the info., all.
 
Not familiar with the item so only general comments

Basic technique for cap selection is to work out the function/key attribute
of the cap and select accordingly. Some people like to exclusively use low
noise/leakage types like Wima MKS2, but I find this unnecessary in some cases.
You can make some gains by replacing audio path caps with quality replacements
like Nichicon KZ, FG, KW,,,

1) Going up 1-2 voltage ratings is ok but you do pay when you hit the big numbers
like 100V +. I try and stay away from 6.3V, 10V and even 16V as the specs (tan delta)
show them to be ordinary
2) What do you measure capacitance with and how accurate is it. The 20% tolerance
on caps is not great but the OEM may have been -10 +50%. I wouldn't bother but maybe
try on some critical caps.
3) Nichicon carry a wide range.
for audio path the pecking order is KZ then FG then KW
For smaller amps (50W?) KW is fine for main filters however increase the uf by
1/4->1/3 since the ripple rating is a little low. OK to use KW for power supplu output
and on the voltage rails. Nichicom UPW for the power supply itself.
4) My opinion only. Most of the watts we hear comes from the main filter caps supplying
the output transistors and pushing amps out to the speakers. The preamp signal is a
few volts only. UKW for mains, UPW for PSU, UKZ for psu output and rails.

Will try and track down sm to comment on the
Hello mbz,
Not familiar with the item so only general comments

Basic technique for cap selection is to work out the function/key attribute
of the cap and select accordingly. Some people like to exclusively use low
noise/leakage types like Wima MKS2, but I find this unnecessary in some cases.
You can make some gains by replacing audio path caps with quality replacements
like Nichicon KZ, FG, KW,,,

1) Going up 1-2 voltage ratings is ok but you do pay when you hit the big numbers
like 100V +. I try and stay away from 6.3V, 10V and even 16V as the specs (tan delta)
show them to be ordinary
2) What do you measure capacitance with and how accurate is it. The 20% tolerance
on caps is not great but the OEM may have been -10 +50%. I wouldn't bother but maybe
try on some critical caps.
3) Nichicon carry a wide range.
for audio path the pecking order is KZ then FG then KW
For smaller amps (50W?) KW is fine for main filters however increase the uf by
1/4->1/3 since the ripple rating is a little low. OK to use KW for power supplu output
and on the voltage rails. Nichicom UPW for the power supply itself.
4) My opinion only. Most of the watts we hear comes from the main filter caps supplying
the output transistors and pushing amps out to the speakers. The preamp signal is a
few volts only. UKW for mains, UPW for PSU, UKZ for psu output and rails.

Will try and track down sm to comment on the caps.
These are great suggestions for replacement caps. I was wondering if it is normal to replace the old caps with new higher capacitance values? I have seen this done by some members and others are just staying with the original values. The older caps seem to be much closer to their stamped values versus all the new caps seem to be on the lower end of the spec.
 
Back
Top Bottom