More Volts = More / better sound?

Hi all,Thanks very much for a lot of info & ideas here. Actually I have mod the HT power supply for my aerius, increasing its ht voltage from originally around 2800 volt to now 4800 volt. The result was fantastic that the loudness, details & clarity inprove & its allmost one year now without any problem.
As Jim Power said that there is a danger of the high voltage jumping & arc over to other components damaging trances and the circuit board as a whole.
Whats I have done mine is that, remove the internal ac transformer and now using only one external insulated step up ac transformer in metal box & its ac secondary voltage was 300v series with 10 kohm 20 watt resistor each & feeding back to ac input terminal of both (left & right ) ML. I have added one more fuse 0.25 A at the neutral terminal of the ac input in the ML to make sure if its accidenlly connected to the wall ac supply it will not harm the ML. I have tested it, & yes no problem because the new added fuse blown straight away .The existing 0.25 A fuse was not blown becaus its at the live terminal & there is resistor before the Ht voltage circuit. Actually in this case the new added fuse blown whe I connect direct to wall ac because its will shorted the ac neutral to ground ( ML ac ground wire is a must for this mod otherwise dengerous both to us & ML ).
For the aerius I increase the Ht from dc 2800v to dc 4800v by adding the capacitor on its original pc board which is allredy there (voltage doubler ) & all 4 1/2 watt 15 mega ohm cabon resistor remove & was replace with one 5 watt wire wound resitor . The stator wire was clamp with ferrite bid & this I found solve the slightly hissing noise heared when I put closse my head to ML pannel.
After all I'm heapy with my mod & was succses and in the 1992 ML broncure ( " OPEN YOUR EARS AND SEE " ) becomes a reallity also stated " the stators designed an extremely rigid metal system and a special insulation application with exceptional damping characteristics that allow safe operation up to 10,000 volt."
OK, no danger and I really in provided doing with somebody ( don'd do alone, Hight voltage is very dangerous ) & follow the advice from Jim and Joe, should be ok.

Regards,
greenipsum -Aerius serial no: 496 & 497QUOTE]

This thread is to capture discussion, ideas, suggestions and metrics around what impact does increasing the ESL Diaphragm voltage have.

As you all know, our electrostatic speakers need AC power due to the high polarization voltage required to run an ESL panel. The Diaphragm is charged with anywhere from 2,500v to 5,000v depending on models and input AC.

The effective voltage potential on the panel itself will change over time as the panel gets used, dusty, etc. So some people have had success in varying the incoming voltage with Variacs to raise or lower effective panel voltages and therefore have an effect on the sound.
Others have modded the guts of the ESL drive circuit to get substantial increases in voltage.

This thread is to discuss those approaches and their results.

greenipsum said:

Hi all,Thanks very much for a lot of info & ideas here. Actually I have mod the HT power supply for my aerius, increasing its ht voltage from originally around 2800 volt to now 4800 volt. The result was fantastic that the loudness, details & clarity inprove & its allmost one year now without any problem.

Click to expand...

Greenipsum, excellent info. In this thread on the Monoliths, we were also talking about the effects of raising the panel voltage by using Variacs to just increase the AC voltage, would of course result in slight (~10%) increase in panel voltage.

Sounds like your direct mod let’s you increase diaphragm voltage by 60%.

This subject of Diaphragm voltage increase (through Varics or direct mods like yours) is definitely worthy of its own thread.

I’ll ask the mods to move your note and this reply into a new thread I’ll start shortly with the title of “More Volts = More / better sound?”

Thanks,

Thank's for all, I will post the picture shortly when I'm ready.
For your info, I allready try using Variacs increasing the input ac voltage from 240V ( Malaysia ac supply is 240v ) to around 260v & yes there are inprovement, but finally both my internal ML ac transformer faulty ( primmary winding burn ), and its very hard here to get back the transformer.
Because of this actually why I'm using external ac transformer & do the mod.
Actually the ac transformer used was designed for fixed specify ac supply voltage of the country where we are & its can handle 10% extra primmary voltage for short time only.
Thanks,[/QUOTE]

Triacs to manage external voltage

[Bringing over this post I had made earlier in the Monolith thread]

On old panels, stepping up the polarization voltage will probably improve things, as it provides the stators a greater potential to work on. A variac is the correct tool for that.

The ‘darker’ sound is when you lower the voltage and the speaker essentially tilts more towards the bass driver, as you are basically applying a volume control to the panel by lowering the polarizing voltage. It also probably skews the FR of the panel, but I’d have to measure to see what the net effect is.
In general if your AC voltage is under 115v, a variac would be a very nice thing to have, as one can push the input by about 10% (e.g if input is 110v, then output is 121v)

Go here to see what all kinds of Variacs there are: http://variac.com/staco_Variable_Transformer_Map.htm

For ML speakers, a small 2 to 5amp amp version is about all you really need, the current draw from the polarization circuit is minimal (the big Monoliths only draw 25w). But a bigger variac won’t hurt anything.

http://variac.com/staco_variable_transformer_100_.htm and put it on the back panel (requires pulling out the woofer, drilling a hole and hardwiring. Probably only for hard-core tweakers

Or the enclosed external version: http://variac.com/staco_variable_transformer_3pn221.htm

The large enclosed versions often found on eBay are a great deal (big 10 to 15amp versions for under $100. They might look a bit ragged, but they are generally indestructible.

But this guy's (davidRiddle) store on eBay sells very nice refurbished units, such as this 5 amp version for $150: http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=170116046106&ih=007&category=73377

Thanks a lot, Jonathan. I'm getting some great results with my new(actually old) 10 amp variac. Right now, I just have it at a low setting, 70. My wish was to tame some of the brightness coming off the panels, and man does this thing work!! The sound is quite a bit darker, and the woofer is much more prominent. I'm now working with backing the woofer level off at the crossover. I may not stay at 70, but at least now I know that the variac can make a big difference in how the speakers sound. The highs, especially the strings, are silkier, and for some reason the stage seems to have spread out bigger and deeper, as I'm able to give the system a bit more volume now. I really like it. It may be the best $77 I've spent on the QuestZ. I will also experiment with raising the voltage, as you've suggested for my older panels.

Regards,
Adrian

I think my Quests really don't like it when I increase the voltage to the panels. Things already got a bit too bright for me when I moved the variac up from 70 to just 75. Isn't this strange considering that my panels are 12 years old? I love the warmth of the sound at 70, but there is a bit too much midbass. Controlling the level at the Xover isn't quite doing enough, so I may start lowering the Xover point of the woofer from 315 Hz. Hopefully that will start to restore some of the low bass extension that I miss as well. I'm sure that I'll have a lot more control over how things sound once I have the DBX DriveRack 260. Still, what I'm getting now is more enjoyable that what I had before. This is a lot of fun!!

Adrian

Cool, great to hear you’re enjoying the results. I definitely think that lowering the voltage reduces the efficiency of the panel, therefore tilting the sound to the woofer. A gain control decrease on the panel amp in a bi-amped setup would probably achieve the same thing.

I just won an eBay bid ($40) on a 5amp Variac so I can play around with the measurement rig I have and see what impact to frequency response (not just gain) this has.

I also have an offer out on an Electrostatic voltmeter to actually measure the diaphragm voltage accurately (at the panel).

Greenipsum, I sure hope you soon post your diagrams and more details on your mod.

Thanks,

OK, great...I would love to know the measured results that you get with your variac. I'm sure that there would be some differences between our panels based on age, Monoliths vs. Quests, and other factors, but a general idea would be good. I seem to have gotten the mid bass better under control. I find that there are some recordings that I want to turn the variac up for, and others I want to turn it down. I should probably find a setting that works well for most things. I'm up in Toronto for the next few days, and should be back for more listening next weekend.

Adrian

Sory for late reply. I reopen my Aerius. Once open, its normal for pupil like us must do something. I readjust the Ht voltage and finding that's for my Aerius now, 3800V is the best sounding. Higher than this - sounding fast & lest dynamic. Migt be because now I drive it's fully active ( external electronic x-over).
I attach afew picture and draw it's Ht circuit to show the mod.
Actually the Aerius Ht cicuit design by Ml engineers was super allredy in the sense that the voltage very stable and to increase or decrease the voltage by simply change the component's value on the PCb ( Zener voltage & resistor value) provided we know how the circuits working.
To measure the voltage i used meter shown taken from old test gear & I diy 10kV probe. When in use with Fluke 87 multi meter, reading acceptable, for example on my analog meter shown -reading 4kV, Fluke 's reading 395 ( actually 3950 v ). Three 15 Mega ohm & one 14 mega ohm 1/2 watt in series as to make this probe.
Finally I would like to say that at 3800v Ht voltage and drive with 25 watts
valves amp in triode mode, the sound of this 10 over year Aerius still "Hair rising".
Thanks,[/QUOTE]

...the sound of this 10 over year Aerius still "Hair rising".

Click to expand...

Greenipsum, that's great! Thanks for sharing the pictures and the details.

I'll have more questions later.

Well, I ran some test on my Monoliths, and guess what? No differences.

Pretty bizarre, as I expected at least a lowering of SPL. But nope, same volume metric (almost). The only variance was 0.1 dB SPL between 120v and 70v.


I tested Balanced 122v vs non-balanced 120v vs non-balanced 70 and 60v and no measurable differences between any of them.

My conclusion is that varying the input AC has very little effect on the speaker.

Now the question of why?

Is the Monolith ESL power supply regulated in such a way that it can adjust the output voltage to match? Don’t know, as I did not measure the ESL energizing voltage.

And I can’t yet measure the Electrostatic field at the panel, as I’ve yet to be able to procure a meter.

Next step is to open the ‘spare’ SL3 and re-run some electrical tests.

Here is a Screen shot of the impulse response of 120v (blue) and 70v (green). No difference.

Jonathan,

For some reason I'm not seeing the blue response line on the graph. I do see a few black dots, could that be it? So if there is no measurable difference if I use a variac to raise or lower the input AC voltage, can there still be a change in frequency response? I can certainly hear changes, but I don't have the ability to measure things like you can. It does seem logical that your results suggest that the power supply is compensating for any changes in the input voltage. If this is true, why are there any differences in the sound?

Adrian

Adrian,

The reason why the blue is not visible is that it’s largely hidden behind the green trace, as they are 99% identical.

Keep in mind that reflects frequency and phase response, not volume.

The SPL diff is 0.1.

I think it might be the Monoliths. I need to test my spare SL3 and maybe my Sequels to see if there is an effect with them.
I fully expected to see a diff, at least in SPL.

Will post more tests this weekend.

Has anyone considered the adverse effects that this will have on the longevity of the panels (raising the voltage)? Will doing this effectively "wear out" the diaphram prematurely? Is the subjective increase in sound quality worth possibly cutting in half the life expectancy of the design?

slownlo said:

Has anyone considered the adverse effects that this will have on the longevity of the panels (raising the voltage)? Will doing this effectively "wear out" the diaphram prematurely? Is the subjective increase in sound quality worth possibly cutting in half the life expectancy of the design?

Click to expand...

I don't believe it adversely effects the panels. the only downside to a higher Electrostatic potential is it attracts more pollutants. So vacuum more often

Life expectancy should not be materially affected IMHO.

JonFo said:

Adrian,

The reason why the blue is not visible is that it’s largely hidden behind the green trace, as they are 99% identical.

Keep in mind that reflects frequency and phase response, not volume.

The SPL diff is 0.1.

I think it might be the Monoliths. I need to test my spare SL3 and maybe my Sequels to see if there is an effect with them.
I fully expected to see a diff, at least in SPL.

Will post more tests this weekend.

Click to expand...

[Several weekends later ]

I finally grabbed the spare SL3, and set up the measuring rig to have another go at this.

Here are the results of some measurements on this topic I took today.

First, a graph of frequency response at 70 volts AC (Blue Trace), 90 VAC (red), 120 VAC (Cyan).

As you can see, the lower voltage does have an impact on lowering the high-frequency volume. I tested many other points, but basically after 100 VAC, it’s pretty much the same. I’d say the range for adjusting this is between 60 and 90VAC.

So, basically it’s a 1 to 2dB volume control in the high-end, as I noted in prior posts.

Using an Electrostatic Voltage Meter

I also had procured an Electrostatic Voltage Meter(EVM) to help me determine the impacts of the voltage changes.

The unit I got is a Chapman model EVM5000, which seems pretty basic, and it’s repeatability of measurements a bit suspect. Placement and initialization seem to have a lot to do with what measurements I get back.

If anyone has tips on how to best get readings using a handheld EVM, I’d appreciate the feedback.

It does seem that the lower voltage in, drops the EVM detected voltage (duh) but by how much, I can't tell, as getting repeatable measurements is difficult with this device.

Here’s pic of me taking measurements:

Good for all ML Fans

look at R23 R24 thay show the Bias xfourms. 90V .ac thay sound Bad Need to be Mills or others or a 10k pot.
thay need to go BUT you well fry the Zeners,but lift Z3 an bias goes from 2k400 to 4k600 more later.
thanks

tubevibb said:

look at R23 R24 thay show the Bias xfourms.

Click to expand...

Note: With original pcb if increase voltage more than 4kV sometime hiss.ss sound heared on the panel and this may be due high voltage leakage in the pcb. For my Aerius untill now 3.8Kv was the best voltage and give the most wonderfull sound without any hiss..ss eventhough i put my ear close to ML panell.thanks