Leak TL/12 Plus / Point One Plus

[Tony L]

Just done the power-up into a shorted-out input lead as suggested above and no buzz/noise of any kind.

I guess the next thing to try is the Quad 34 in place of the Audio Synthesis stepped attenuator?

PS If it is something external I don’t understand how it can consistently (100%) affect one of the two amps and not the other. The issue unquestionably follows that amp chassis (i.e. not the valves, interconnect, mains lead, preamp/CD channel, speaker it is connected to). No exceptions here, when the issue occurs it is always with amp #2.

 

[Radford Revival]

Interesting that it doesn't manifest with a shorted input...

Another thing, has the offending amp always had the same EF86 in it?
Edit: Sorry I can't read! Nevermind!

 

[Radford Revival]

Is pin 2 (the shield pin) OK on the bad amp? It looks like the Leak only uses pin 2 where as on a Radford pins 2 and 7 are tied together (both are shield pins on an EF86).

I'm wondering if the Octal thing is a red herring and the physical force of unplugging/replugging the octal lead is moving something else in the amp

 

[Tony L]

Is pin 2 (the shield pin) OK on the bad amp? It looks like the Leak only uses pin 2 where as on a Radford pins 2 and 7 are tied together (both are shield pins on an EF86).

I will double-check that. I made a little meter lead to test continuity from valve-pins to the base or circuit below.

I'm wondering if the Octal thing is a red herring and the physical force of unplugging/replugging the octal lead is moving something else in the amp

I am positive it isn’t mechanical. The interconnect leads are a meter long and very compliant, plus I can’t make any impact or noise at all with really quite violent banging on the amp. Replugging the RCA always clears it. Banging the amp or wiggling valves never clears it. 100%, no exceptions to date.

I’ll report back with a resistance reading for pin 2 of the EF86 base tomorrow, plus I’ll dig out the Quad 34 just to see what happens with a mains earthed active preamp. Mine is the later type with RCA connectors so easy to do.

PS I could easy run a wire between pin 2 and 7 if that would help either diagnose or fix the issue.

 

[Radford Revival]

I just thought of another possibility. Something in the "bad" amp doesn't like this particular input arrangement with the octal lead and is starting up in an oscillation mode, and unplugging/replugging the lead is kicking it out of that condition so it remains stable even with the lead plugged back in (it is possible for oscillatory behaviour to "latch" like that, especially on startup). There might be nothing wrong with the amp, it could just be an unlucky marginal difference in the layout.

The reason that would cause hum is due to the oscillation frequency being so far out of bandwidth of the amp any PSU ripple and such may get re-modulated down as hum. Possibly a small value capacitor across the input might cure it.

This would certainly explain a lot of the counter-intuitive behaviour of the fault like the fact that it only happens on warm up.

 

[Radford Revival]

Did you mention earlier in the thread the Leak is the only grounded thing in the system? That'll often nudge things toward that kind of odd behaviour - experimentation with grounding perhaps the preamp may cure this. Just a theory of course!

 

[Tony L]

Did you mention earlier in the thread the Leak is the only grounded thing in the system? That'll often nudge things toward that kind of odd behaviour - experimentation with grounding perhaps the preamp may cure this. Just a theory of course!

Yes. The system as it stands is a Marantz SA8005 SACD player, which, like many CD players, doesn’t have an earth pin on its IEC connector. The “preamp” is an Audio Synthesis ProPassion which is a single-input dual mono stepped attenuator. One thing that occurred to me just now is the Audio Synthesis is a ‘shunt’ attenuator, i.e. the signal goes through a fixed resistor and the attenuation is on the ground. I’ve no idea if this is significant (it works perfectly with the Stereo 20).

Here’s an internal pic of the Audio Synthesis. It couldn’t be more simple conceptually. Just two totally independent mono stepped attenuators in a box. It is just a single input per channel for a single source system such as this one.

Taking as read that I clearly don’t know what I’m talking about at all (I am very obviously not an electronics engineer) the best theory I can come up with is that there is a ‘differential’ or ‘imbalance’ between the two amps as they warm up so there is a ‘charge’ or ‘offset’ somewhere. This is one reason I’m very curious to use the Quad 34 tomorrow as that will, I assume, present a solid ground for the interconnect. Whether this just causes a proper old-school ground loop is yet to be seen, but I’ll certainly try it.

The thing I don’t understand is why it is only the one amp though. I’d have thought any imbalance would be more random than that, especially given that I’ve ruled everything out bar the one amp chassis (i.e. it doesn’t stay with preamp channel, doesn’t care which amp is powered up first etc).

 

[John_73]

On the later Stereo 20s there's two bypass components across R5 (4.7k with 300pF) which are meant to help with stability. Does the TL12+ have those? If so is the capacitor original and perhaps the leads are corroded or something so their affects are now compromised?

 

[Radford Revival]

Tony - did you see the post above my other one - I reckon it's instability that's the issue - earthing issues possibly exacerbating it but not in the traditional ground loop sense. It explains why unplugging/replugging the interconnect makes it go away etc

 

[Tony L]

Tony - did you see the post above that - I reckon it's instability that's the issue

Yes, many thanks. I’m still processing it! I’ll try changing things upstream (preamp etc). What value/type of cap would you suggest for the input if swapping out the preamp doesn’t fix it?

On the later Stereo 20s there's two bypass components across R5 (4.7k with 300pF) which are meant to help with stability. Does the TL12+ have those? If so is the capacitor original and perhaps the leads are corroded or something so their affects are now compromised?

The TL12+ has the flat wafer cap thingy, but not the additional resistor of very late S20s. This is an interesting area as my earlier S20 has neither, but has a different value for R5. I find it interesting that the TL12 Plus had the wafer cap, then it was removed for early S20s, and finally brought back with an additional resistor for very late grey S20s!

 

[MUTTY1]

Yes. The system as it stands is a Marantz SA8005 SACD player, which, like many CD players, doesn’t have an earth pin on its IEC connector. The “preamp” is an Audio Synthesis ProPassion which is a single-input dual mono stepped attenuator. One thing that occurred to me just now is the Audio Synthesis is a ‘shunt’ attenuator, i.e. the signal goes through a fixed resistor and the attenuation is on the ground. I’ve no idea if this is significant (it works perfectly with the Stereo 20).

Here’s an internal pic of the Audio Synthesis. It couldn’t be more simple conceptually. Just two totally independent mono stepped attenuators in a box. It is just a single input per channel for a single source system such as this one.

Taking as read that I clearly don’t know what I’m talking about at all (I am very obviously not an electronics engineer) the best theory I can come up with is that there is a ‘differential’ or ‘imbalance’ between the two amps as they warm up so there is a ‘charge’ or ‘offset’ somewhere. This is one reason I’m very curious to use the Quad 34 tomorrow as that will, I assume, present a solid ground for the interconnect. Whether this just causes a proper old-school ground loop is yet to be seen, but I’ll certainly try it.

The thing I don’t understand is why it is only the one amp though. I’d have thought any imbalance would be more random than that, especially given that I’ve ruled everything out bar the one amp chassis (i.e. it doesn’t stay with preamp channel, doesn’t care which amp is powered up first etc).

Now that’s craftsmanship... anyone fancy their chances soldering like that!

 

[Radford Revival]

Yes, many thanks. I’m still processing it! I’ll try changing things upstream (preamp etc). What value/type of cap would you suggest for the input if swapping out the preamp doesn’t fix it?

Unsure yet - there may be a workaround that doesn't involved components at all (if I'm even right about this, I could well be completely off the mark)

The TL12+ has the flat wafer cap thingy, but not the additional resistor of very late S20s. This is an interesting area as my earlier S20 has neither, but has a different value for R5. I fund it interesting that the TL12 Plus had the wafer cap, then it was removed for early S20s, and finally brought back with an additional resistor for very late grey S20s!

The best combination for stability is likely the resistor and capacitor but you'd need a scope / signal gen to actually confirm this.

 

[John_73]

The TL12+ has the flat wafer cap thingy, but not the additional resistor of very late S20s. This is an interesting area as my earlier S20 has neither, but has a different value for R5. I find it interesting that the TL12 Plus had the wafer cap, then it was removed for early S20s, and finally brought back with an additional resistor for very late grey S20s!

Might be an idea to check the leads aren't corroded and/or it's still in spec? Stability might be marginal if it's not working correctly, and partially/temporarily corrected (like Will said above) when unplugging and replugging in the interconnect.

 

[Radford Revival]

Just to hopefully clarify my other post, my theory is that there is an unstable condition, caused by some unintentional feedback somewhere, that the amp can only achieve while warming up with the lead connected to the rest of the system. Literally just the act of having the amp connected up that way while it warms up. No mechanical issues involved. The condition for oscillation isn't met with the amp fully warmed up, with or without the lead connected, but while the amp is warming up WITH the lead connected up, it gets trapped in an unstable condition and the only way to break out of it is by disconnecting the lead. This isn't necessarily an intuitive concept but this can happen and explains why unplugging the lead causes it to stop, even after plugging it back in. It's probably that one of the amps is just a bit different enough which is why it always happens on the same one.

This might explain why it didn't happen with the RCA end shorted out, because it was just on the end of the lead and not some relatively big capacitive equipment hanging off the end.

The unintentional feedback might be made worse by the fact that the casework on the rest of the equipment is floating. Hopefully I've explained it better in this post!

 

[Tony L]

The best combination for stability is likely the resistor and capacitor but you'd need a scope / signal gen to actually confirm this.

If needed I have a scope and, ahem, a Moog Voyager! I can get pretty decent analog sine waves, square waves etc, I could obviously use a digital file cut to CD or via a computer too if any better.

Hopefully I've explained it better in this post!

Yes, thank you. It makes a lot of sense and sounds like a very logical explanation. I will try changing the stuff upstream.

PS I’ve just taken the multimeter to the Audio Synthesis and there is no continuity between left and right signal ground. The RCA sockets are entirely isolated from the case, it really is two entirely isolated mono attenuators in a box. That coupled by a floating CD player upstream may provide a slightly odd environment I guess. I’m very curious as to what the Quad 34 does as that has to change the picture quite significantly from an electrical perspective. I can obviously also try a different source component. I have no idea what the Marantz presents to the world, though it seems perfectly happy driving the passive pre from a sound quality perspective.

 

[Radford Revival]

I wonder if it would stop if you connected the casework of the passive to the power amp with a clip lead...

 

[Tony L]

Might be an idea to check the leads aren't corroded and/or it's still in spec? Stability might be marginal if it's not working correctly, and partially/temporarily corrected (like Will said above) when unplugging and replugging in the interconnect.

C15? It does look more than a bit shabby. If it is suspect and anyone can point me towards a good replacement I’ll swap them out in both amps. It is awkward to get to as it is right under R5, so I don’t want to fart about trying to lift an end to test it, if it’s in the frame even slightly I’ll just replace it with new/known good. My meter isn’t great right down in the pF range anyway so I’m not confident I’d get a meaningful reading anyway. I have no idea what that type of resistor is even called, but if you point me to a vender I’ll go grab a couple.

PS You can spot some C15 crustyness in this pic:

Shame the resistor is on top!

 

[Tony L]

I wonder if it would stop if you connected the casework of the passive to the power amp with a clip lead...

The casework of the passive is completely isolated from the signal returns, but I can certainly try that as it even has an earth terminal on the back panel! FWIW I suspect this terminal is actually bogus, the dual mono ProPassions like this one are so crazy rare/limited market that AS couldn’t didn’t bother printing up a dedicated back panel, it actually has the labelling from a phono stage silk-screened on the rear (they all seem to), and I suspect that is why it has an earth terminal. I suspect it was a special order. Regardless, yes, I can ground the case to anything you like, but don’t think it will impact the signal path through the attenuators. I’ll stick it on the list to try and report back.

 

[John_73]

C15? It does look more than a bit shabby. If it is suspect and anyone can point me towards a good replacement I’ll swap them out in both amps. It is awkward to get to as it is right under R5, so I don’t want to fart about trying to lift an end to test it, if it’s in the frame even slightly I’ll just replace it with new/known good. My meter isn’t great right down in the pF range anyway so I’m not confident I’d get a meaningful reading anyway. I have no idea what that type of resistor is even called, but if you point me to a vender I’ll go grab a couple.

Yep, was looking at your photos again and the leads are quite corroded, but might mean nothing as long as the connection on the soldered areas is still electrically good. I’d start with Will’s suggestion first, and look into that cap as a last resort. But its condition did jump out in the photo, and it is doing (I believe?) the same job as the C/R network on the later Stereo 20s, so you never know. It might be affecting the stability just enough - along with that passive - to cause your issue.

 

[Tony L]

Yep, was looking at your photos again and the leads are quite corroded, but might mean nothing as long as the connection on the soldered areas is still electrically good.

I’m confident ‘my’ end of the lead is solid, but I’d not put any money on the entry into the cap being good at all. I don’t understand the construction of this type of cap at all (I don’t even know what they are called), but it looks real manky to me!