Axiom Home Page Forums General Discussion Technical Questions Amplifier Capacitance ? Axiom amp versus my monos

In the research that I have done, it appears that capacitance is also important in an amplifier design, and in some respects more important than talking about all these "watts" available as most of us get hung up on. In addition, things like damping factor and current delivery are important.

In purchasing my Odyssey Monoblocks some time ago, one of the advantages over other amp designs I was looking at, was that each of my monoblocks has 120,000 microFarad capacitance, for a total of 240,000 uF for both of my m80's.

The new Axiom 1400-8 amplifier has 140,000 uF total for the entire 8 channel amp. Now, I am not totally up on digital versus class A A/B designs, so maybe the more efficient 1400-8 does not require as much capacitance?

I was just thinking that if it is able to route up to 1400+ watts to any given speaker, if needed but unlikely, than it would also require a lot of capacitance for reserve to be available when called upon?

This may be a question more suited for Ian, Alan, or Tom, as they are obviously more in tune with the design. I'm just trying to understand this part of the equation.

Thanks in advance, Randy

I would like to hear a reply on this as well.
Indeed capacitance has been a major factor in how effectively some of the amps i've tested have played back speakers.
This big Coda 10.2 i tried once had enough reserve to playback for a second to two AFTER shutting the unit off (longer at lower volumes).
That's alot of reserve. I believe they had typically used about 120,000 uF in a Class A/AB amp.

As a side note, my Audiobytes play for a second or so after shutting the unit off--more if it's turned up louder.

dang, I figured many people would have input on this one.

I'll give it a free bump.

I'm curious, too, so double bump.

Well, Randy, with all this bumping going on and since I've finally read this, I'll point out that maybe some of the research that you mention has accidentally led into the land of the practitioners of voodoo audio. Watts are in fact the bottom line(often fewer of them than is sometimes imagined)and if a watt is deivered with flat response over the 20-20KHz audible range(or maybe more realistically 15KHz for most of us, as Ian among others has pointed out)and with inaudibly low noise and distortion, that's all that any amplifier can do; there ain't no more. All the other factors in amplifier design technology(damping factor, capacitance, slew rate, etc. etc.)lead to this end result.

Some of the material linked in the past here includes the capacitance discussion on Rod Elliott's Westhost technology site in the "Increasing capacitance" section and in the following "Major myth regarding capacitance" section. There are rules of thumb regarding the amount of capacitance needed(e.g., 100uF per watt)in the power supply section to smooth out the pulses of AC power enough so that the the PSRR(power supply rejection ratio)of the amplifying section can handle it without audible flaws. This and more precise calculations result in requirements in the tens of thousands, not hundreds of thousands of uF; more doesn't somehow become audibly better.

Another topic discussed here in the past was damping factor and the explanation by engineer and speaker designer Dick Pierce in Audioholics was cited. Bottom line is that anything in the double digits is fine. Just about any well-designed modern amp does this, except for some tube designs which have a factor around one or two and are obviously for this and other reasons not designed with the highest fidelity in mind. Damping factor specs in the hundreds or even thousands are seen on occasion, but are audibly meaningless; again, more doesn't mean better.

Way to squash the capacitor party, John.



On a serious note, there is a common belief that a less than LARGE capacitance reserve might be completely depleted during loud and/or percussive passages, leaving the capacitors sucked dry and unable to deliver for the next passage. This isn't the case, as stated in the first article you linked, copied below:

"With large capacitors, the momentary current peaks created by the program material will not be of sufficient duration to discharge the caps to the full load voltage levels, so there is more voltage available on a more or less consistent basis. This equates to more power for transient signals, and lower ripple voltages the rest of the time."

So John, to put this in laymans terms...

If the amp reliably delivers the advertised watts at a flat freqency response with acceptable THD+N then you are good to go for all that other stuff mentioned above?

Fred

Randy. Its important to keep in mind that you are talking about 2 very different amp designs with the Odyessey monos being a Class AB design and the Axiom being based on a Class D design. That presents entirely different sets of tradeoffs and design issues of which capacitor size and array are actually a small part of it.

The articles cited above are very good for describing issues involving Class A or Class AB amps but people should be cautious about making generalizations when discussing other classes of amps.

Class A and AB amps are the most common and can provide excellent audio performance but their achilles heal are that they are energy inefficient which is why they need fans or heat sinks which adds to their weight and size . Typically 40-50% of the energy gets lost as heat which puts alot of stress on internal circuits and components. That energy inefficiency also necessitates the need for somewhat larger caps to deal with energy loss. Some would suggest that larger caps provide better transient performance but I'm on the fence on that one since so much depends on how the circuits and components have been configured from amp to amp.

With Class D switching amps like the Axiom, they are very energy efficient, with the A-1400-8 being close to 95% efficient. The switching circuitry requires very fast voltage response and gate time in order to keep distortion to a minimum, so more smaller caps carefully laid out would be a better design choice. The dynamic power handling of the Axiom amp also means that the power supply rails have to turn on a dime and dump power quickly into the channels demanding more watts. Again smaller caps would be a better choice.

Finally there are size constraints. The A-1400-8 is a small technological marvel and there aren't that many places to fit large caps.

Thanks everyone for your input.

Jakeman, I figured we were not talking apples to apples comparing Class D to a Class A-A/B and that not all theories apply to all situations.

I'll read some more on this topic as well. I've noticed that most of the well known reputable amp companies out there design their amps with higher capacitance, especially if they are type A or A/B or A-A/B. So what you say makes sense about efficientcy. I'm sure they design the amps that way for a reason and are not just doing it to be snake oil salesman.

Heck, Tom C from Axiom himeself used to design huge monoblocks back in the day, and I highly doubt he would be creating amps with extra capacitance unless it was required.

I also forgot to add after reading the article that JohnK mentionned above, that each of mono's are made up of 8 15,000uF capacitors for a total of 120,000uF each monoblock. The article hints that there are benefits of having more capacitors to make up the total, versus say 4 30,000uF caps.

I'm guessing that Axiom must use 4 35,000 uF caps to make up the total of 140,000uF,at least that is what it appears in the pictues.

Looks like the ripple effect and other specs change based on design.

You got it, Fred; there ain't no more.

Perfect, back to fat dumb and happy for me. Funny, I'm usually the guy who wants to know all the intimate details and geeky stuff.

Now, if I could just sort out those pesky Blu-ray HD audio formats...

Good stuff, Randy. I should also mention that the other factors that determine the size and number of capacitors has to do with powersupply rails and whether or not the amplifier modules are bridged or not. The size and number of caps influence how fast voltage can be drawn from the powerrails and output. A good comparison is to look at two Bryston amps that have identical frequency response, THD, IMD, S/N and other specs.

I used to have a Bryston 14BSST a few years ago, which was the stereo version of the 7BSST monoblocks... the ones Axiom used as a basis for its development of the A-1400-8. Both these Bryston amps used the same components and had similar specs. However they differed somewhat in how the capacitors were arrayed and rail voltage.

The 7BSST monos had 16 x 10,000 mfd small capacitors close to the output section. The 14BSST used 4 x 27,000 mfd large capacitors instead of 16 small ones in the 7BSST. Also the 14BSST have a slightly lower powersuply voltage of 60 volts instead of 65 volts for the 7BSST.

People comparing these two similar amps reported a tighter more responsive sound with the 7BSST or a "softer" sound in the 14BSST, which can be attributed to the use of many small capacitors in the 7BSST and its higher voltage powerrails. Basically the 7BSST was able to respond more quickly to power demands during transients because of these small differences between the two amps.

I should check but I thought the Axiom used an array of different sized caps 4 big ones and 4 smaller ones to get better response time. The number and size of the caps also depends on the amount of voltage in the powerrails. Going by memory, I recall that the A-1400-8s voltage was higher than most amps at around 80-85V. The bottom line is that its a very "fast" responsive amp. It has to provide dynamic power to whatever channel needs it.

Stop trying to talk me into one.

Just calling it the way I hear and see it. Wasn't it Joe Friday who used to say "just the facts ma'am."

Jakeman -
How do you think the A1400-8 stacks up against the Sherbourn 7/2100A w/mono blocks? http://www.sherbourn.com/products/7_2100A.php
Would love to know your thoughts on it.
Thanks.

Ditcin, some random thoughts. From your link its Scott I believe.

Firstly let me say the Sherbourn 7/2100a is a fine example of a terrific class AB HT amp. Its the kind of powerful 7 x 200W@8ohm monoblock amp that should form the foundation for any HT, in my opinion. I've always been a fan of monoblock designs and before the A-1400-8, it was the type of amp I would recommend to my friends who were looking for the best HT amplification. I've never owned the 2100/7 but did have a Bryston 9BSST for a while which I really enjoyed, a 5 x 140w monobloc design with gain controls to boot.

To put your question into perspective, the A-1400-8 design embraces the monobloc phylosophy and takes it a step further. If you check out the video on the amp you'll hear Tom and Ian talk about how they set out to replace a stack of 7 monos with one unit that was more dynamic, didn't clip and provided much more power than available with monos.

To get the 2100/7 to output at its max you need to plug both cables into separate outlets otherwise your limited to 1440W total or 205w max over the seven channels. Setting aside the 8th channel and dynamic handling for now, the axiom design will actually output more than the 205w max because of its ability to draw more power out of the line, supposedly up to 285w per channel all driven.

It is rare though that all channels are driven simultaneously at the same power levels and that is where the Axiom really shows its technological advantage. If needed the Axiom can direct over a kilowatt of energy into any channel. It achieves that because of how its switching circuits can route power from its 85v+ powerrails and single massive (over 1600kva !) transformer. With that much power on tap this amp will clip under only the most massive loads and demands. Despite all my attempts one afternoon with no one home, I could not get the 1400-8 to clip. One thing about monos the size of Sherbourne is that they will clip in demanding scenes or in complex musical passages.

Another main difference, aside from the Axiom being half the size and weight and running much cooler, arises from the much faster response time of the a-1400-8. AB designs take longer to ramp up and down whereby a fast switching type D amp will respond much faster to power demands. What helps make it so fast, is the way the 1400-8 has 4 x 33000 mfd caps repowering smaller 4 x 2000 mfd caps as they drain.

Slew rate can be measured several ways but the Axiom weighs in at around 8V/ms while the Sherbourne is a bit over half that rate or 4-5V/ms. I know some folks think that its inconsequential but amp designers sweat for months trying to get amps to respond as quickly as possible since it impacts dynamic presentation and how loud and sharp transients will sound.

Where I believe the Sherbourne and monos in general have an advantage is that there is zero crosstalk associated with them. Any amp with a single transformer will have some level of channel crosstalk though in the Axiom it is well in the inaudible region. Some would argue that there is more crosstalk in the DSP circuits of a processor but the less the better. The other stat I pay close attention to is signal to noise and both amps have great specs.

Very informative, John. Thanks for summarizing all that and making it easy for us (me) to understand.

Thanks Sean. I'm glad you found it informative.

John,
Thank you very much for a great reply. In fact the A-1400-8 I ordered is on its way. I can't believe it shipped so fast!
I was curious and tempted to get one, but at the same time was hungry for more info. There hasn't been really that much posted/printed about it in the terms written in your reply. It obviously takes one who has experienced the mono block powering of an amp to answer my questions and concerns.
I hope it lives up to my expectations.

Now I just have to wait for Purolator and DHL to figure out where the amp is - living in the States there is limbo, a kind of twilight zone where packages disappear for 24 to 48 hrs before they turn up scanned in the US. It's a horrible waiting period since literally neither Purolator or DHL have any sign of it, or where it is.
It's supposed to be scanned when it enters the US but sometimes it isn't. I want to be available for delivey since DHL is great for leaving packages on the stoop or between the doors.
In the past I have actually stepped on smaller packages left on the stoop! (Smaller ones in envelopes)
I've even had packages sent to Ohio then back to NY then to NJ and back to NY where I live.
Grrrrrrrr
Any way, I'm very much hoping to pair my Anthem D2 with a new partner in hopes it's a match made in heaven.
Again, thanks John!

I hope some day I can be an elite member of the 1400 club.

That a helluva great pre-amp, Scott. I lusted after one for a while but ended up getting an Arcam unit because of the long delay while the Anthem engineers perfected it. A friend of mine has one and it makes me want to sell my Arcam every time I see it. The A-1400-8 should really bring out the nuances in the D2.

Don't want to alarm you but I hope Purolator doesn't damage the container. A friend in Sacramento got his last week and the amp was DOA as a result of being damaged in transit. These delivery companies should be held accountable for all the damage they cause. Axiom sent him a replacement pronto but I think the real solution is to use a more reliable company like Fedex.

John,
No one ever stated truer words - I've had on going bad experiences with DHL.

The worst I posted here about a year ago when my Epic 80 7.1 System was upgraded to the VaSSallo Series - one of my M80s got scratched by a fork lift that pierced the box. Axiom was great about it, but DHL was horrible.

When I had ordered an Axiom system for my brother as a Christmas gift it was sent via FedEx without a problem.

Even though the track record for DHL is poor I have to hope for the best and put out some positive vibes that the amp "will" get here in working order.

... or else I'll have send some of my cat's droppings to one of the head honchos at DHL again. (They gave me a free envelope for overnight shipping when they scratched my M80 - so I filled it with kitty waste in a ziplock and forwarded to some exec who was into "quality control"

...never hear from him. Imagine that.

BTW - you "lust" for the D2 is well placed. It was another experiment for me since I was basically happy with the Sherbourn. In the end the D2 was beyond anything i could have hoped for. The video processing alone is a knock out. I really does up convert 480i DVD to 1080p with HD like results. I had a friend from FL visit this past holiday season and he though it was HD.

If you decide to order one you will not be disapointed.
I was going to post a pic but don't know how to get it into this posting.

Too funny.

I hear you about the D2. I have an Integra 9.8 on order but many users are reporting problems. If it doesn't meet my expectations, I'm going to do what I should have done last year and step up to the D2.

The D2 is still in my target. I know they have a representative doing business only 2 miles from my place. And they have one in show room... just too scared of not being able to resist buying on the spot!

John, thanks for the amplifier explanation.

Well if you give into temptation rest assured on the tech side that if you need it it will be there. I have called them twice and they have been very, very helpful. I mean, Axiom-like helpful. It was refreshing and pleasure.
The first time I contacted them it had to do with a strange issue that the decoder was showing 4.0 audio (as with films mixed in the 1950's) as being simply "Stereo". Although it was decoding it right it was not registering it right on the front pannel.
They were completly unaware and asked me for some titles it was doing this with so they could write a fix as a future firmware update.
They actually cared and took time to understand and try to solve a simple question on a issue most wouldn't care about.
I dealt with other companies, like Denon and then only heard excuses.
Yes the D2 is big bucks, but at least you know it comes with the support it should have - and with it piece of mind.
(Hope this temps you!

I have a question. Did Anthem exist as a separate company and was then purchased by Paradigm or was it created as a brand name? Or am I mixed up and it is the other way around! In other words, is it Paradigm giving support or the "original Anthem company"?

(also, I guess you meant "peace of mind", otherwise you'll hear a piece of MY mind! ).

He he he. I just remembered why the D2 was on the backburner for me. No HDMI! I am not ready to give up the "single-connection" per device just yet! Is the D3 on its way?

Eric. Paradigm's first major foray into electronics was the purchase of Sonic Frontiers in the 90's. It was regarded as one of the top tube/solid state preamp and amp boutiques. Paradigm rebranded Sonic as "Anthem". Many of the people from Sonic are still there today.

You may be thinking about the D1. Actually the avm 30, avm 40, avm 50, and D2 have 4 HDMI inputs.

DG56. Glad you found it helpful.

But the D2 does have HDMI. That's how I have my Oppo connected. Here's a link to the fact (PDF) sheet.


http://statement.anthemav.com/HTML/Products/D2/Literature/downloads/D2_PDS.pdf

I have that from http://statement.anthemav.com/HTML/Products/D2/Specifications/D2_Specs.html. Are those specs wrong?

Old spec sheet. There were hdmi revisions. Check out this current comparison sheet. http://www.anthemav.com/NewSitev2.0/Anth...050_D2_Comp.pdf Try not to get a nosebleed looking at the suggested retail prices. You can get them much cheaper.

Any idea when they will upgrade reon processing to 1024p/24f?

Strange to have more recent info not accessible through their web-site!!! Or is it an unofficial sheet for an upcoming version of the D2?

Thanks. They should really cleanup their web-site. I should not be able to access old spec sheets by direct clicking on what's on screen.

On the other hand, good news... not for my bank account...