Vandersteen 5A battery replacement

[jd002200]

I wanted to describe my recent experience with replacing the batteries for my 5As in case it might help others. A little while ago it seemed like the soundstage and dynamics from my system suddenly collapsed. I was about due to replace the power tubes in my amp so I did that, but that didn't improve the sound. I then replaced the power supply tube in my ARC Ref 6 preamp, but still no improvement. I then emailed Richard to inquire if it could be the batteries in the speaker although they weren't due to expire for another 6 months. He said it probably was the batteries so I sent the crossovers in along  with the high pass filters for battery replacement, and Vandersteen provided quick turnaround. After I reinstalled everything, the system sounded like it should. I was so happy! They were Lithium batteries so they appear to lose their voltage rather quickly towards the end of their life. I'm actually glad that for this experience because it shows me how critical the battery bias is for these speakers. The difference is remarkable.

[GdnrBob]

Welcome to the forum!

Glad you got it figured out. I wonder if Lithium batteries give up the ghost quicker than standard?

From what I have read, the crossovers usually seem to go quietly into that good night rather than suddenly conking out.

B

[Richard Vandersteen]

lithium batteries do hold voltage and then drop rapidly.  Alkaline go down slowly to about 7 volts and then die.  Most do make it to our noted date for replacement.

RV

[ctsooner]

Congrats for getting it fixed and welcome. I hope you browse and decide to stay involved.  🙂 

[RWP]

I assume the speakers/crossovers still "work" when the batteries are dead, right, but that they sound less life-like, is that it?

[Richard Vandersteen]

The sound quality goes backwards dramatically.  For some it goes backward slowly so it is not noticed.

RV

RV

[Stringreen]

I had a year left on my batteries as well, but thought I'd have them replaced with 10 yr batteries, after service repaired my 5A midrange driver that was making a non-Vandersteen noise. Upon its return,  I attached all the cables, etc. and got fuzzy sound from the left, no sound from the right, a warning red light on the amp, and then shutdown. I was delighted I didn't see flames or smell smoke. Both Vandersteen service and Ayre told me it sounded like a bug in the amp....tomorrow its scheduled to be picked up by UPS.  I also asked Ayre to test my amp's input impedance at 100 k for the needed filter setting.....I don't want to blow up the amp after it comes back using the voltmeter Vandertones procedure ......all for the love of music.

[GdnrBob]

4 hours ago, Stringreen said:

I also asked Ayre to test my amp's input impedance at 100 k for the needed filter setting...

Isn't it possible for them to insert the crossover into the amp? I think @ctsooner had it installed in his integrated.

FWIW, I had the similar thing happen with my MX-R. I plugged it in when I switched from the Atmasphere amps. A slight pop and the light turned red. I sent it back and they found a capacitor had blown. Did you look inside? My amp had a definite dark spot from something burning.

bob

[TomicTime]

….yes..it’s very possible to change out the coupling capacitor for the appropriate value to create a high pass filter…..BUT…. those super high quality teflon caps take awhile to break in..AND unless a DBS voltage or similar high tube amp voltage is applied, they will lose form…. So unless the amp circuit itself keeps it formed in standby ( not likely ), the M5 will most likely sound better. RV and i just did an analysis of my tube amp and decided that the circuit would not keep the cap formed. The M5 filter is a better choice for me. YMMV. 

Jim

[ctsooner]

I had it in there. They used their best cap for it and it sounded fine.  I trust what Jim and RV are saying though.  It makes perfect sense. 

[TomicTime]

Pete - Right on, it’s circuit dependent and Ayre standby mode likely kept it formed. Jim

[Holmz]

Are there some links that talk about “the forming”… or explain it?

[GdnrBob]

40 minutes ago, Holmz said:

Are there some links that talk about “the forming”… or explain it?

That is a good question. I just accepted that caps needed a constant voltage to keep them 'formed', but never understood why.

B

[TomicTime]

perhaps…more..than you wanted to know….

 

https://en.wikipedia.org/wiki/Film_capacitor

[Holmz]

5 hours ago, TomicTime said:

perhaps…more..than you wanted to know….

 

https://en.wikipedia.org/wiki/Film_capacitor

I understand them, but the paper did not appear to talk about “forming”.
I found self healing though.

[ctsooner]

On 1/28/2022 at 4:01 PM, TomicTime said:

Pete - Right on, it’s circuit dependent and Ayre standby mode likely kept it formed. Jim

Thanks Jim.  You had me wondering.  Alex was there when I got it done and he said they would use the Wonder Caps or whatever their caps are in the Ref series and that it wouldn't deform.  If I had to do it all over again, I think I'd buy the pig tail/box from V.  With it built in, I wasn't able to have friends bring their amps and stuff over to preview etc... 😉. 

[nrenter]

I’d take the risk of sonic degradation of a single “deformed” capacitor at this particular at this stage of the system over another physical junction  (and that assuming that’s the ONLY difference - no addition in-signal soldered junctions, wires, switches, etc.). I know I’m I the minority with this opinion, as I’ve been admonished in the past for suggesting the crossover pigtail should be offered as part of a full interconnect (as an option). 

[Richard Vandersteen]

This really is so significant that I have to respectfully disagree as so much of our design innovations require a truly transparent high-pass.  Do the experiment!  In the early days of solid-state coupling caps were often used to eliminate DC offset and hence the terrible reputation of coupling caps even though tube preamps and power amps were full of them!  The big difference is the tube circuits always had a significant voltage across the caps.  Coupling caps at zero volts are never formed because the only voltage they ever see is the signal.  Not good!  Nrenter, no one will admonish you for an opinion formed by an actual experiment as this one does not require one to be an audiophile.  Over the years I have proven this fact to many engineers in our industry.  This Forum is friendly and exist to improve our knowledge of things Vandersteen and get the most out of one's investment for the enjoyment of music. 

RV

[BKDad]

I don't think that this is a question of a capacitor "reforming", in the same way as electrolytic capacitors that haven't been used for a while.   That's a somewhat different mechanism.  There, the capacitor doesn't work properly until the chemical dielectric between plates is properly formed.  

Here, we're talking about optimizing performance by establishing a significant and stable electric field across the capacitor dielectric that is modulated by the applied signal.  With no bias, the small internal dipoles of the dielectric constantly get "pushed around", especially when the field reverses at zero-crossing of the applied signal.   This is a different sort of forming.

I'm no materials scientist, but this seems like a decent explanation for us amateurs of what Richard is describing:

https://pressbooks.online.ucf.edu/osuniversityphysics2/chapter/molecular-model-of-a-dielectric/

Nice drawings, too.

 

Edited January 30, 2022 by BKDad
badd speling

[TomicTime]

Yep, good dialogue for sure. Certainly for those questioning the additional connections with the high pass box, you are just a soldering iron away from the best of both worlds….. 

Also, for those with a bit of room in the amp, you can implement battery bias with 4 caps and batteries, just be careful to not create a resonant tanking circuit. See the photo of the model 7 high pass filter internals, 4 very expensive caps in the signal path, most of the rest to control the tank where bad things happen.  Not trying to give the farm away, but also consider the cutouts so the cap axial leads can be minimized….good example of hyper detailed execution.

In the end, it’s all about tge music and our enjoyment of it. We are all on different quests. Best to all.

Jim

[TomicTime]

Here ya go. 

[Holmz]

5 hours ago, BKDad said:

I don't think that this is a question of a capacitor "reforming", in the same way as electrolytic capacitors that haven't been used for a while.   That's a somewhat different mechanism.  There, the capacitor doesn't work properly until the chemical dielectric between plates is properly formed.  

Here, we're talking about optimizing performance by establishing a significant and stable electric field across the capacitor dielectric that is modulated by the applied signal.  With no bias, the small internal dipoles of the dielectric constantly get "pushed around", especially when the field reverses at zero-crossing of the applied signal.   This is a different sort of forming.

I'm no materials scientist, but this seems like a decent explanation for us amateurs of what Richard is describing:

https://pressbooks.online.ucf.edu/osuniversityphysics2/chapter/molecular-model-of-a-dielectric/

Nice drawings, too.

 

Thanks

 

6 hours ago, Richard Vandersteen said:

This really is so significant that I have to respectfully disagree as so much of our design innovations require a truly transparent high-pass.  Do the experiment!  In the early days of solid-state coupling caps were often used to eliminate DC offset and hence the terrible reputation of coupling caps even though tube preamps and power amps were full of them!  The big difference is the tube circuits always had a significant voltage across the caps.  Coupling caps at zero volts are never formed because the only voltage they ever see is the signal.  Not good!  Nrenter, no one will admonish you for an opinion formed by an actual experiment as this one does not require one to be an audiophile.  Over the years I have proven this fact to many engineers in our industry.  This Forum is friendly and exist to improve our knowledge of things Vandersteen and get the most out of one's investment for the enjoyment of music. 

RV

Does the bias need to be higher than max signal negative voltage?

i.e. Does the polarisation reverse when it crosses zero?

And if @nrenterdid the experiment, then what y would he see?
Would it be something crossover distortion when the signal goes through zero?

[Richard Vandersteen]

The bias needs to be enough higher than the signal, so the cap does not have to reverse the negative part of the signal is simply less positive.

RV

[ctsooner]

I trust Richard over Alex for sure, but I also had no way of testing. This was purchased before I knew much about the crossovers etc. The simple solution to cut down the connections is to buy the M5 amp. They are built in and you can flip a switch when you upgrade to Kento,s. Simple as it gets. 

[BKDad]

Is there a limit on what capacitors benefit from adding DC bias?

For example, it's often a (though not common...) good idea to add damping networks with emitter/source follower circuits in order to prevent instabilities.

https://www.hifisystemcomponents.com/downloads/articles/Prevent-Emitter-Follower-Oscillation.pdf  << Note that the origin of this article was not the audio world.

(If you do the math or simulations, you find that almost all active circuits have the potential for being unstable, since the devices do not have infinite bandwidth.  Proof left to the reader.)

Anyway, in low level circuits that use devices with low internal capacitance, the cap values needed with the RC circuits are often fairly low.  Like, maybe, 47 pF or lower.  This is way lower than usually found in loudspeaker crossover networks and the high-pass filters we use in Vandersteen systems with subwoofers, either built-in or separate.  At these low values, you can use small polystyrene or even certain C0G ceramic caps to good effect. 

Would these small caps work better if they were DC biased?  In some cases, there may be a trade-off between the cap performance and power supply rejection if you add bias, unless you use a battery, of course.  Good compromise?

Another example would be the small value caps often used at a preamp or power amp input with resistors to form low pass networks in hopes of minimizing RF interference effects.

Yeah, I could do the experiments myself, but if somebody has already done the work, why not ask?