nailzgun said:
Its warming up right now on my Vandersteen 2ce's.. so far I notice extended bass.. nothing more.
Yeah in my case; it will pop a 5amp when the caps are drained off overnight.. but works fine if you just flip it off and on again.. 6 amp seems to be okay.. but that is high.. I guess a slow blow 4amp might be in order.. Before the caps I could run on a 3amp!
The thermistor will limit to 5amp.. but the 400a won't draw that much under normal operating circumstances.. I don't think Mr. Pass would have recommended it that were the case.. I think he is pretty analytical..
LOL.. My quesiton is; what fuse would I use? I think Id still be stuck with a 6 amp! And in my opinion.. that is really high for the 400a..
Anyhow... back to your question.. thermistor might be needed to keep the fuse size down.. say you blow 7 amp fuses.. the thermistor would be perfect at 5amps.. maybe run a 6amp fuse..
one way to test the amp is to either monitor it with a meter or put a fuse in; bring it up on a variac and see if you blow it while cranking the living heck out of it..
Make sure its biased if its not self biasing.. the bias will drastically change the fuse requirements!
Getting back to your question "Is there a point where increasing the size of the capacitors has no effect or is even detrimental to the sound?"
I don't know the answer.. I know someone else will.. hang on a bit..
Hi,
If you require absolute switch on reliability, then I would recommend that you fit relay bypassed soft start on both the mains side of the primary and to the rectifier side of the smoothing bank.
Use power resistors to limit transformer surge current and time the relay to about 250mS.
Use a pair of thermistors in the secondary side and time the relay to about 10S.
Select the mains fuse as small as will reliably switch on with minimal nuisance tripping.
Fit rail fuses (after the smoothing bank) that pass about F(0.5*Ipk)A into your minimum load.
The NFB filter half an octave higher @ 120mS to 150mS
The PSU a further half octave above giving RC=160 to 200mS.
For 8ohm speakers that equates to 20mF to 25mF/rail/channel
4ohm speakers requires double the capacitance
3ohm speakers require 180mS/3ohm=+-60mF/Channel. (= conclusion in post39)
If you, or others, think that compromises the mid and treble then split your speaker crossover to allow bi-wiring and feed the bass half of the crossover with a good bass amp and optimise the other power amp to drive the mid/treble. Problem gone.
If you require absolute switch on reliability, then I would recommend that you fit relay bypassed soft start on both the mains side of the primary and to the rectifier side of the smoothing bank.
Use power resistors to limit transformer surge current and time the relay to about 250mS.
Use a pair of thermistors in the secondary side and time the relay to about 10S.
Select the mains fuse as small as will reliably switch on with minimal nuisance tripping.
Fit rail fuses (after the smoothing bank) that pass about F(0.5*Ipk)A into your minimum load.
For good bass I recommend that the high pass input filter should be between 80mS and 100mS.barchetta said:
The NFB filter half an octave higher @ 120mS to 150mS
The PSU a further half octave above giving RC=160 to 200mS.
For 8ohm speakers that equates to 20mF to 25mF/rail/channel
4ohm speakers requires double the capacitance
3ohm speakers require 180mS/3ohm=+-60mF/Channel. (= conclusion in post39)
If you, or others, think that compromises the mid and treble then split your speaker crossover to allow bi-wiring and feed the bass half of the crossover with a good bass amp and optimise the other power amp to drive the mid/treble. Problem gone.
Hi
Andrew - I did similar calculations to those you just displayed and as a result graduaslly increased the bias in the Stasis 2 as described in post 39 ( which is a reprint from my earlier post on this ). I constantly found an improvement in dynamics on certain notes such as cymbols. However this was only evident on my most sensative and revealing speakers which are direct coupled to the amplifier. ( that is no crossover on the mid range )
Although I did not state it in my post I found that higher capacitance did not seem to help any further. I took this to mean that the calculations were born out in practice.
As youn correctly say I also used a thermister in the circuit. I used a 10ohm 15 amp thermister. Later I will bypass this to take it out of circuit when the amplifier is playing. Anyone got a good circuit???
Based on my experience have a go and try more capacitance. I think you may be surprised.
Don
Andrew - I did similar calculations to those you just displayed and as a result graduaslly increased the bias in the Stasis 2 as described in post 39 ( which is a reprint from my earlier post on this ). I constantly found an improvement in dynamics on certain notes such as cymbols. However this was only evident on my most sensative and revealing speakers which are direct coupled to the amplifier. ( that is no crossover on the mid range )
Although I did not state it in my post I found that higher capacitance did not seem to help any further. I took this to mean that the calculations were born out in practice.
As youn correctly say I also used a thermister in the circuit. I used a 10ohm 15 amp thermister. Later I will bypass this to take it out of circuit when the amplifier is playing. Anyone got a good circuit???
Based on my experience have a go and try more capacitance. I think you may be surprised.
Don
Hi
Andrew - I did similar calculations to those you just displayed and as a result gradually increased the bias in the Stasis 2 as described in post 39 ( which is a reprint from my earlier post on this ). I constantly found an improvement in dynamics on certain notes such as cymbols. However this was only evident on my most sensative and revealing speakers which are direct coupled to the amplifier. ( that is no crossover on the mid range )
Although I did not state it in my post I found that higher capacitance did not seem to help any further. I took this to mean that the calculations were born out in practice.
As youn correctly say I also used a thermister in the circuit. I used a 10ohm 15 amp thermister. Later I will bypass this to take it out of circuit when the amplifier is playing. Anyone got a good circuit???
Based on my experience have a go and try more capacitance. I think you may be surprised.
Don
Andrew - I did similar calculations to those you just displayed and as a result gradually increased the bias in the Stasis 2 as described in post 39 ( which is a reprint from my earlier post on this ). I constantly found an improvement in dynamics on certain notes such as cymbols. However this was only evident on my most sensative and revealing speakers which are direct coupled to the amplifier. ( that is no crossover on the mid range )
Although I did not state it in my post I found that higher capacitance did not seem to help any further. I took this to mean that the calculations were born out in practice.
As youn correctly say I also used a thermister in the circuit. I used a 10ohm 15 amp thermister. Later I will bypass this to take it out of circuit when the amplifier is playing. Anyone got a good circuit???
Based on my experience have a go and try more capacitance. I think you may be surprised.
Don
nailzgun said:
As far as I understand, the filter caps are there to supply required
current with reasonably low ripple voltage.
By definition, 1F (1,000,000uF) cap means when 1A is discharged,
the cap loses 1V. So, if the discharging cycle is 100Hz, it loses
0.01V per cycle. If an amp draws Class A bias current of 5A, it
loses 0.05V per cycle. If we replace 1F with 100,000uF cap, the
peak2peak ripple voltage is about 0.5V for the amp. If the 0.5V
ripple is equal to or less than 1% of the rail voltage in Class A
amp, I have heard that it is acceptable in most case. (Learned this from Papa's paper ~ ^^.)
So, the size of caps mostly depends on the required amount of
current supply at low ripple voltage which is equal to or less than
1% of the rail voltage.
I don't think the increased capacitors will blindly improve bass
sound or mid/high sounds. If it really does, I would not hesitage in using $ for the bi~~~g caps ^^.
Of course, there are ESR and ESL to be considered . . .
Hi,
Could you check it? I wonder if it should say 1V/S?
If the amp is fed 20Hz and draws a peak current of 12.5Apk from a supply of Vrail=+-50Vdc then 40mF results in a ripple of ? I think the ripple is about 10% to 15% of supply voltage, but almeliorated by the recharge pulses occurring @ 10mS or 7mS intervals. How far will the transformer rebuild the smoothing cap voltage during those short pulses while the amp is drawing significant output current in the respective half wave time frame?
One pole of the smoothing cap pair will get a chance to recharge during the other half of the wave form and then the rail voltage gets pulled down again. This can be easily seen if one were to use an oscilloscope to monitor the rail voltage. The recharge saw tooth superimposed on the bass note saw tooth is recognisable and the voltages are measurable.
I have not actually measured it. Instead I have listened to the resulting bass output from a few wideband amplifiers (-3db 2Hz to 50kHz) and found that inadequate smoothing does result in less extended bass and attenuated bass in comparison to the remainder of the music. This effect may not be solely attributible to smoothing capacity, as your final statement alludes, but capacity seems to be the easy solution.
Most of what you say is correct, but I cannot follow that first sentence.Babowana said:
Could you check it? I wonder if it should say 1V/S?
If the amp is fed 20Hz and draws a peak current of 12.5Apk from a supply of Vrail=+-50Vdc then 40mF results in a ripple of ? I think the ripple is about 10% to 15% of supply voltage, but almeliorated by the recharge pulses occurring @ 10mS or 7mS intervals. How far will the transformer rebuild the smoothing cap voltage during those short pulses while the amp is drawing significant output current in the respective half wave time frame?
One pole of the smoothing cap pair will get a chance to recharge during the other half of the wave form and then the rail voltage gets pulled down again. This can be easily seen if one were to use an oscilloscope to monitor the rail voltage. The recharge saw tooth superimposed on the bass note saw tooth is recognisable and the voltages are measurable.
I have not actually measured it. Instead I have listened to the resulting bass output from a few wideband amplifiers (-3db 2Hz to 50kHz) and found that inadequate smoothing does result in less extended bass and attenuated bass in comparison to the remainder of the music. This effect may not be solely attributible to smoothing capacity, as your final statement alludes, but capacity seems to be the easy solution.
AndrewT said:
Ok.. so now let me say that I am not quite familiar with all the physics (funny that many years ago physics is what I took at University) anyways I am confused, but only alittle.
With my S500 amp with the 1000w transformer 119,000uF main caps and 250w @ 8ohms output I think what I have heard from you guys is that I can significantly increase capacitance with the following (in laymans terms caveats).
1.To much capacitance will cause an inrush amperage surge on startup that will need to be moderated to keep from blowing fuses. Question here is this surge also possible during significant musical dynamics at high volume?
2.That as the capacitance increases the ripple voltage will increase and this is not desirable. Two questions here, what are the effects of the ripples and are there capacitors built to reduce this?
3. ESR and ESL need to be accounted for as I increase capacitance, I must confess I do not understand yet what these two terms are. Or their impact. As I said in my first post I'm new
.So ultimately if I increase the capacitance to say 240,000uF I'll have more current to control my speakers, (bi-wired Infinity Renaissance 90's, 2ohm) but I'll possibly experience the 3 above issues?
Thanks for all the help
Scott
Hi
I will try to answer some of your questions. However there are others her more knowledgeable than I am.
When I first tried increasing capacitance it was to cure a slightly soggy base and to give harder edges to snare drums and the like. Both are situations that require a lot of instant energy from the power amplifier capacitors. If the capacitors struggle to supply the demand then rail voltage drops and you hear a less loud note. I figured and calculated that ( other thing remaining unchanged and able to handle the capacitance/energy ) increasing capacitance should rectify that. Others please step in if you disagree.
I tried this on a power amp with a 1kva transformer and 4 X 17,500uf capacitance and 3/4 ohm speakers.
I calculated that increasing capacitance should increase the energy available to draw on when demanded by the amplifier. It did. The other note/calculation that Andrew added above is to show the time taken to recharge the capacitance and the effect on the rail voltage.
To make this effective I think you do need low low resistence capacitors.
Don
I will try to answer some of your questions. However there are others her more knowledgeable than I am.
When I first tried increasing capacitance it was to cure a slightly soggy base and to give harder edges to snare drums and the like. Both are situations that require a lot of instant energy from the power amplifier capacitors. If the capacitors struggle to supply the demand then rail voltage drops and you hear a less loud note. I figured and calculated that ( other thing remaining unchanged and able to handle the capacitance/energy ) increasing capacitance should rectify that. Others please step in if you disagree.
I tried this on a power amp with a 1kva transformer and 4 X 17,500uf capacitance and 3/4 ohm speakers.
I calculated that increasing capacitance should increase the energy available to draw on when demanded by the amplifier. It did. The other note/calculation that Andrew added above is to show the time taken to recharge the capacitance and the effect on the rail voltage.
To make this effective I think you do need low low resistence capacitors.
Don
Hi Don,
Higher capacitance will only have an effect if you are drawing enough current to drop the supply rails to a point where there is current limiting.
In other words, if the supply is not sagging to the point where you can not supply the energy, increasing capacitance will only make you feel better (and increase spikes & ringing on the rails). This will only occur if you are nearing the maximum output for your amplifier.
The other bottle neck would be the actual design of the amp. If a little sag affects the amplifier's operation, the design is defective. For these amps, increasing the supply capacitance may help but you are trying to solve the problem in the wrong place.
-Chris
Higher capacitance will only have an effect if you are drawing enough current to drop the supply rails to a point where there is current limiting.
In other words, if the supply is not sagging to the point where you can not supply the energy, increasing capacitance will only make you feel better (and increase spikes & ringing on the rails). This will only occur if you are nearing the maximum output for your amplifier.
The other bottle neck would be the actual design of the amp. If a little sag affects the amplifier's operation, the design is defective. For these amps, increasing the supply capacitance may help but you are trying to solve the problem in the wrong place.
-Chris
anatech said:
Wrong place? I disagree, if it "helps" it with no ill effects its not the "wrong place".. thats just an opinion at that point.
This is like saying that a $500 speaker cable wont make a difference to someone who feels (for WHATEVER reason) that the sound is better.
Maybe I misunderstood what you said, but thats my take.
Hi Joe,
Problems are easier to solve when the source of the problem is corrected. The overall performance will be better too.
Attacking problems that are only symptoms will bring you something like the Counterpoint (and others, it's just soooo easy!) series of amplifiers.
Once the unit is working properly, go ahead and knock yourself out with tweaks. Although you may end up compromising the performance while things are changed. I've seen this more times than I can count over 35 + years in service.
-Chris
Edit:
Problems are easier to solve when the source of the problem is corrected. The overall performance will be better too.
Attacking problems that are only symptoms will bring you something like the Counterpoint (and others, it's just soooo easy!) series of amplifiers.
Once the unit is working properly, go ahead and knock yourself out with tweaks. Although you may end up compromising the performance while things are changed. I've seen this more times than I can count over 35 + years in service.
-Chris
Edit:
Most times people don't know enough about what they are doing to determine if there are no ill effects. The average tweaker is an extremely poor judge of this. - from my experience anyway.
anatech said:
Actually this point seems pretty valid. Before I start 'tweeking' the amp for better sound I should at least bring it to original condition to have a benchmark for any further tweaks. It could just be that because the amp is 20 years old and all original that a general update might accomplish what I am after?
Hmmmm....
I rather enjoy this forum, I don't have any people in my circle that give a rats a**** about equipment. Thanks for allowing me to participate.
Scott
Hi anatech
I agree with you.
The amplifier that I experimented on was a Stasis 2. I chose it as it has a substantial power output stage and a 1kva transformer both of which could deal with additional current draw without problems. I did however fit a soft start system.
What I find in more detail is that with base and particularly with the leading edge of certain percusion notes the current draw by my 3/4 ohm speakers, even in my listening room, is very high and I believe does cause sag in the rail voltages.
The additional capacitance seemed to cure the problem.
Don
I agree with you.
The amplifier that I experimented on was a Stasis 2. I chose it as it has a substantial power output stage and a 1kva transformer both of which could deal with additional current draw without problems. I did however fit a soft start system.
What I find in more detail is that with base and particularly with the leading edge of certain percusion notes the current draw by my 3/4 ohm speakers, even in my listening room, is very high and I believe does cause sag in the rail voltages.
The additional capacitance seemed to cure the problem.
Don
AMV8 said:
Well now I'm back to my quandary
that is the amp I'm upgrading is the A/AB version of your Stasis 2. Same output stage and transformer just not running in Class A. Don, other than the soft start to keep from popping fuses, did you need to change anything else? Also your speakers, are you using the Stasis full range with them?
anatech said:
Do you live and work in Georgetown, Chris, or are you one of the lucky commuters.
Scott
Hi
I am using the stasis on a full range set up. However in my case the mid range is direct connected to the amplifier - no cross over. The treble and base then have a capacitor and inductor as a simple filter and really just assist the mid range unit at the frequency extremes. I thnk it is because my mid range is direct connected that the leading edge of notes is so clearly different with extra capacitance. ( I think this stops the rail voltage sagging when a sudden short demand of current is made - my speakers are 3 ohm )
I am using a 7 amp anti surge fuse. I also have a 10 ohm 15 amp thermister in series with the mains positive in lead. ( I plan to fit a relay across the thermister later to take it out of circuit when the amp is running )
I think the stasis is a great amplifier. I have been very impressed. It just has a great balance to the sound. Have a go. If you feel uncertain about increasing the capacitance just increase capacitance slowly and see if you can here a difference.
Don
I am using the stasis on a full range set up. However in my case the mid range is direct connected to the amplifier - no cross over. The treble and base then have a capacitor and inductor as a simple filter and really just assist the mid range unit at the frequency extremes. I thnk it is because my mid range is direct connected that the leading edge of notes is so clearly different with extra capacitance. ( I think this stops the rail voltage sagging when a sudden short demand of current is made - my speakers are 3 ohm )
I am using a 7 amp anti surge fuse. I also have a 10 ohm 15 amp thermister in series with the mains positive in lead. ( I plan to fit a relay across the thermister later to take it out of circuit when the amp is running )
I think the stasis is a great amplifier. I have been very impressed. It just has a great balance to the sound. Have a go. If you feel uncertain about increasing the capacitance just increase capacitance slowly and see if you can here a difference.
Don
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.