This concerns the impact of increased capacitance in the power supply as it may, or may not create an issue with the rectifier tubes.

I need absolute clarification here.


As the needed parts are slowly being delivered, I have a few important questions that are critical for the intended upgrades I am considering.

First: The power supply.

I am replacing the entire bank of 47uf electrolytic capacitors with much higher quality MKP polypropylene  film and foil designs. I am dividing the total capacitance between a 40uf base cap + 8uf secondary base cap+ 2uf Russian paper in oil (metal canisters), to act as a large bypass for filtering the bass response far deeper than the small bypass cap can. This all gets bypassed by a high quality Russian K75-24 Hybrid PIO 0.1uf capacitor for high frequency filtering. ( everything is rated above 700 volts )

Now, due to the large increase in capacitance, (500uf over the stock collective of 470uf, not counting the small bypass) I have concern over the following due to this extra 30uf of capacitance.  This has the potential of a stiffer power supply leading to stronger bass response as a positive aspect. But, here is the negative aspect???

The ToriiMk3  has two stages of tube regulation to consider here, but the main concern is toward the effect this will have on the rectifier tube stage.


The designer of this amplifier would know this for sure, so perhaps Steve could give a confirmation concerning this situation. If anyone else has actually done something similar to what I am doing, then let me know how it worked out for you. Just remember, the ToriiMk3 topology is different than the other Torii models, so we are comparing apples to oranges here.

From what I understand about the potential risk of adding this much extra capacitance is that by doing so, this could lead to excess stress on the rectifier tubes, prematurely wearing them out, or causing them to just outright fail.  Again, just a general consensus overall concerning tube amps. I am only concerned with the ToriiMk3, NOTHING ELSE. It doesn't matter how this affects other amp designs, I only need to verify this condition with the amp I am about to upgrade.

First: Is the above statement something I need to be concerned with considering the parameters I have disclosed with increased capacitance?


The majority of the benefit comes from the small 0.1uf bypass caps.  I just need to be sure that the extra capacitance by doing so will not cause harm to the rectifier tubes.  From what I understand, the tube regulation stages are not affected by the increase in capacitance nearly as much as the rectification stage.  I need to be sure about that area as well.

Also, will this possibly affect BIAS?

Second: It is noted that excessive capacitance ( what is considered "excessive" here?), leads to increased stress upon the power transformers as well as the rectifiers. A major concern here is the potential stress caused by INRUSH current upon startup of the amp, leading to shorter conduction angles for the rectifiers during normal operation..


What I gather from information is the following in order to prevent this situation.  I would like some clarity in order to confirm the following if this is an acceptable approach.

This is the suggestion I was given:

"A 30uF increase over a 470uF filter capacitor ( combined capacitance ) is generally acceptable as long as it's NOT THE FIRST CAPACITOR in the circuit, and the rectifier tube's maximum capacitance limit isn't exceeded."



So from what I can gather from the image I have, it looks as though pin 8 from the rectifier base goes directly to the positive lead on the first filter capacitor in the bank shown with the orange circle. I will have to look inside the amp again to verify this connection. The other end is tied to the ground bus bar, as is all other filter caps.

If this is actually identified as the first filter cap ( the reservoir ), then I am to understand that this cap needs to remain a lower capacitance than the others following it downline. I am guessing that this would be okay to use a single 40uf 5% 700v cap to keep it low enough, or should I add another 8uf to it which would only increase the value by 1uf from stock design? I could also split the difference and use a 2uf paper in oil capacitor with a polypropylene 40uf cap. A total of 42uf should be within good working parameters for just the first filter capacitor after the tube rectifier, am I correct?

If all of this is true, then starting with the second section of filter caps, I can increase to whatever value I like to reduce ripple since this comes after the choke section making it completely safe to do so. This is what I make of it. Correct me if this is wrong.

Is it true that it can be beneficial to add consecutive capacitance as you progress downline in the capacitor bank? In other words, make the following capacitor larger than the previous one. Or would it be best to just keep all of the secondary filter caps the same size after the reservoir filter cap?







The following image shows the small red capacitors between the OC2 tube regulators and ground. Since these are used inline with a resistor and connected to ground, the purpose is of filtering, and stabilizing the voltage provided by the regulator tube by the way I see it.

This is what I have learned about this technique: RC low-pass filtering by combination of a resistor and a capacitor.  When configured as a low-pass filter, this circuit allows lower frequency signals (like DC and slow variations) to pass through while attenuating higher frequency signals (like noise and ripple). In the context of an OC2 regulator tube, the resistor in series with the OC2 and the capacitor across the output forms an RC low-pass filter, contributing to the overall filtering and stabilization of the regulated voltage.

So with that discovery, it is safe to assume that since this already acts as a bypass capacitor, it would be pointless to use a high quality capacitor here since it would not be audible or make any difference? That is the way it would seem to me. Just a basic tight tolerance cap here is all that really matters.  If I am wrong about this, and a higher quality cap WILL make a difference in sound quality, then please let me know to make that happen.  I think a low ESR tight tolerance cap seems to be what is important here for this purpose.






What I hear is that there is a limitation on the FIRST filter cap due to high rectification pulses which will damage the rectifier tube.

I hear that after the first filter cap, the excessive capacitance is not a problem since there are isolating resistances for the remaining B+ circuit which work to eliminate or reduce these impending pulses. I gather that the purpose of these isolation caps works to prevent feedback from stage to stage progressively through the B+ line.  


Further information I have read. "The first filter cap is considered the reservoir/primary filter cap. This directly follows the rectifier tubes. Supposedly, downstream in a latter stage, by using a resistor or choke between it and the rectifier, this reduces the impact upon the rectifier itself."


Again, confirmation concerning this would be best answered by the amp designer. I need to be absolutely sure about this upgrade before I begin. I want to get started on this immediately.


One more thing, I assume there are "bleed down" resistors installed in this amp which rapidly discharge the capacitor bank once the power is removed. Is that correct?


My plan is to unplug the power source as the amp is turned on to get the initial voltage reduced significantly with minimum residual voltage as possible. Then carefully use a bleeder resistor tool to finish removing the voltage to zero on each capacitor for safe handling. Of course I will use a multi-meter to verify before handling. I don't need any shock therapy yet, but I might later. One hand behind my back, and be careful what I touch..and perhaps, I might not get a wake up call!  500 + volts could really ruin one's day. I think I read that this amp was designed to be regulated at 410vdc. Is that correct?

Would it be a good idea to let the amp sit for a couple of days to insure complete voltage bleed down for safe handling?

I also understand that there could potentially be an issue with dielectric absorption caused by capacitors regaining a small charge even after they have been discharged. It would be a good idea to check each capacitor to verify before its removal if the amp has sat for awhile during a break just to be sure. I like to assume it has a charge, and always discharge it again with the discharge resistor tool to ground prior to handling.


Thank you for providing me feedback on this for clarification. I can't begin working on the upgrades until I get a verified response on this.

Good evening there CAJames.

Not to worry about lawyers. They are just bottom feeders who live well off of our waste.



Just joking, I think.




Anyway, thanks for having the decency to respond to my questions here. I appreciate your prompt reply. I did find it helpful.

What you stated triggered me to explore deeper into this. I just took some time taking a refresher on tube rectification, whether it be by tubes, or by way of diodes in a 4-way full wave rectification scheme. I learned what I needed to know about overloading the rectifier tube with too much capacitance used in the first primary filter capacitor (reservoir). It makes perfect sense once you understand the entire principle and how changes affect the rectifier tube itself by way of too much demand leading to stress and premature failure. It would seem that this is not so much of an issue with diode rectifiers. But you know, what would a tube amp be without those large glowing tubes!

It really isn't so much about the inrush current as much as it is concerning over stressed tubes which fail due to too much capacitance demanding a forced current flow beyond safe limits of the tube design. After all, the choke limits the inrush factor from what I gather.

I don't need to reinvent the wheel here, just trying to improve upon a few things, mainly better capacitors. On that note, it is best to remain within the original design parameters and there won't be an issue.


I do not want to be restricted to using a GZ34 since that limits my choice. What if I don't like the sound signature using the GZ34's? That doesn't leave me much choice. I would prefer to stick with 5U4G or its compatible variants.

I have the solution now after taking some time to think it through and gather some valuable data which answered my own question. I just need to verify the routing from the rectifier to the first filter cap. That stage is to be left without an increase of capacitance. A better quality cap with a small bypass yes, no more.

After that stage, the limits are off. I am increasing the size of the amp base to make it deeper with much more room. So stuffing large capacitors in there will be no problem. I plan to fabricate a retainer platform which creates a second level directly below the original layout. This provides a secure platform to attach and secure the large caps while keeping the wiring short as possible. I am stepping up to a much larger mil spec hook up wire which is silver layered OFC copper core.

As for why to unplug the cord as the amp is still on is from advice which I have seen on several audio forums from guys who have been doing this type of work for a long time. They claim that it has an extra draw on the capacitors as the power bleeds down. Hey, I am not going to argue about that or even be concerned with it. It won't hurt to do it, and if they are right, then the amp will have less voltage remaining in the capacitors. It doesn't matter, as I still need to bleed everything down manually anyway with a bleeder resistor tool. Just check with a meter to verify it.  Time to get to work rather than worry about minor details.


Just take things at face value and consider that not everything is what it is claimed to be.  I look at it this way. If you want to verify this, then try it both ways. First by just turning the amp off, then unplug it.  Check to see the capacitance level in each cap. Then try it by unplugging the amp while it is on (volume turned down of course). Check the capacitance remaining this time again. See if there is a difference in the level of capacitance remaining between the two test parameters. That is the only way to find out. It doesn't hurt to experiment a bit as long as nothing gets damaged or someone doesn't get hurt.


This is most likely not worth doing, but I have noticed many who suggest its practice. Either way, I need to verify a safe condition before I touch anything. This is kind of like always being aware of your proximity to running jet engines, especially as they ramp up for taxi. Get just one foot too close and you are history. There have been more than one who found out the hard way with nothing more to show of their existence than a splatter of bloody tissue all over the flight-line. Fatigue leads to bad judgement which in turn leads to death.  There are others who just screw up due to complacency. If I failed to follow procedure during engine runs at 70 percent thrust around F-15 vary ramps, I wouldn't be here tonight. I have a history of being involved with dangerous work. Just use common sense and be aware. Then there is a chance of seeing another day.


Thanks again for the advice. If I don't return, you know what happened!

My solid state amp has 2.102 million micro Farads of capacitance in it, and boy oh boy you can hear it.

I know that you have to have a 'soft start"  on it to keep the inrush down and causing problems. It takes around a minute for my amp to click entirely on.
It will also play for at least 45 seconds with the plug pulled from the socket!

In my case, the main thing that you can hear is the silence, everything is so quiet.

The person who made my amp is Angela-Gilbert Yeung and they also build external capacitor banks that can be added to other amps.
https://www.angela-gilbert.com/sp-capacitor-pack
Maybe this can give you some ideas.

I think that more capacitance is a good thing and hope that you can add a lot more to your Torii 3.

Put the brakes on right there!



Funny how a simple request always manages to get twisted around in another direction, without the initial situation being focused upon the way it should be.

I was expecting Steve/Decware to give a simple yes or no answer on this important question. As a long time customer, I don't think that is too much to ask, is it?

But that is okay, never mind. All is well.

Anyway, as I made clear, I solved my own problem and found the answer. But thanks to all for their suggestions.

All I wanted to know is what capacitance is recommended for this particular amp involving the first primary filter cap after the rectifier tube, since the designer just happens to be on this forum every day, and I didn't think it was too much trouble to get a simple and direct response concerning my exact question. I don't expect anyone else to know that answer unless you have actual technical experience doing what I am about to do, specifically with a ToriiMK3 amplifier.

When we have engineering issues with a 767, we don't call Lockheed Martin to resolve the issue. No, we get in touch with the design engineers at Boeing who designed and built this particular aircraft. That is the logical approach. But hey, I come from a professional background with decades of experience, what do I know.

The amp is 13 years old, the caps are starting to leak electrolyte from two of them in the filter stage. These electrolytic capacitors have been used well over their usable lifespan.

You read my response incorrectly, and took it the wrong way.

If you're worried about safety, like you seem to be,

As for being "worried"? Are you serious?  Being worried is when you are in direct gunfire, or when you travel down enemy territory knowing that at any time you may set off a land mine, leaving you dead, or going home without legs.

I seemed to have managed 8 years of active duty service without being "afraid".  We do not fear the challenges we face in life if we expect to accomplish anything.

So NO, if you knew my background, and my level of courage taking on difficult challenges, you would never bring that to question.

I was only making small talk which somehow got interpreted as some type of "fear" which is not even a part of this thread.

If I was that stupid, I would send it to a shop and pay them a huge bill to do what I can easily accomplish for free.

If I seem to be able to change out a home HVAC system with lethal voltages of 240 involved, as well as a gas line which would level the house if ever ruptured with exposure to a spark, then why would I worry about something on the 120v level? Sure, it could kill you if fully charged, and you have both hands touching the ground path. That is a good way to stop the heart. Otherwise, it is a nasty shock which is just another hard lesson in life.

I was simply making a point of that.  I really wish I had not brought it up.

I have everything under control. As I stated, I solved my situation and answered my own question. Thanks to all that replied and yes, it was very much appreciated. Just please refrain from assuming someone is afraid of something just because that safety concern was brought up.  

And BTW: I will be changing parts in small incremental stages over time, checking the amp powered on between each stage. This way, if there is an issue, then it is much easier to resolve the exact cause of it. Since I am doing this to experiment with major upgrades to be implemented concerning the capacitors + bypass experimentation, wiring, terminals, etc.., I intend to document what each and every change within the system does for the sound of the amp. This way I can review the changes more effectively before moving on to further upgrades if I choose to do so in the future.

There is only one thing for certain. There is nothing certain in life.


I hope that makes perfect sense to you now.


Now it is time to move on and get something done.