The picture doesn't show it clearly, but it is on spikes about an inch high.
Ehm, small caps usually have the lowest esr and small caps can usually have alot higher ripple current when you take their sizee into account. A small 1000 uF cap can in many cases have a ripple current of 4 A, while big 15.000 uF and similar usually have 7-10 A ripple current. Big caps have higher ripple current, yes, but its cheaper to usw alot of small caps if you want high ripple current. But they have the drawback of low total capacity.
Hmm, you might have a point about the ripple current capability, but I always assumed caps should have a ripple current that not only could withstand the RMS current but also the peak current. Peak current can easily be 2 x RMS current.
But I guess peak current doesnt matter and its the RMS current you should look at? So 10A RMS out need atleast a 10A ripple current cap, irregardless of peak ripple current?
Actually spikes on ceramic tiles on wood. It's a temporary arrangement while I put together better stands.
Hi,
is ceramic tile as a foot any better/different/worse than spikes+ceramic tile or no foot at all?
No. Lower ESL, yes, but higher ESR.
More capacitance = lower ESR and higher ESL
That's why some added small caps can be profitable after R, not before R
By sim (PSUD) of one F5 channel using a 30mf - R- 30mf PSU, the established RMS ripple current into the first cap is about 2,4A only.
A cap given for 4A ripple at 105°C (or 85) will at least ( good caps more) double its ripple capability at 40°C
Ripple rating is related to heat producing into the cap, so doubling this current for a half second is not an issue.
Note that using C-L-C increases a lot the ripple into the first cap.
Wih C-R-C, don't bother too much the ripple current.
Last edited:
Hmm, doesnt it depend on the cap? Can find several small 1000 uF caps with ESR
in the range of 0.012-0.020 ohm.
However in my case, with the 18.000 uF caps im going to use, adding a few small caps really lowers the ESR. The caps have about 0.023 Ohm ESR, the small 1000 uF caps have 0.016 Ohm ESR. The small caps nearly halves the total ESR of the first C.
But what do I know, Im no power supply expert, would be nice to get a definitive answer.
Last edited:
You can use just crappy caps before R provided the ripple capability is ok, the amp sees the caps after R, there, low ESR / ESL are good.
Though, I am not an expert too.
But i know that if i say something wrong, someone will correct me.
Edit: PSUD tells you nothing about the sound.
Though, I am not an expert too.
But i know that if i say something wrong, someone will correct me.
Edit: PSUD tells you nothing about the sound.
Last edited:
This is a very good point most people don't think about.
Eg If you have a 105 degree cap rated at 4A and an 85 degree cap rated at 5A and you operate both of them at 5A at 60 degrees the 105 degree cap will last longer even though it has a ripple rating of 4A (because it is rated for 4A at 105 C not 85 C).
Hi,
Why not just use small clip-on heat sinks? Cviller's board can easily accomodate something like this:
Digi-Key - HS220-ND (Manufacturer - 579302B00000G)
That first cap is charged by very large, short CURRENT PULSES, not some ripple scenario according to a SIM - ripple current ratings will give you an indication if the first cap will take the punishment, but little else, and nothing at all regarding the sound.
Have a good close look at the CURRENT out of your bridge and you'll see that the peak current is over 10A after cap is charged, for an total supply current of about 1.5A rms, regardless of ripple.
Many of the newer compact capacitors have very misleading capacitor ratings and at working temp and power, are not suitable for the job of that first cap - adding a number of them in parallel seems to work okay, but still not the right cap for the job - even when using better diodes (lower reverse current spike) the forward current pulse is a shocker.
The second cap is the main contributor to the sound of the amp and this is where the SIM is useful. Capacitor specs (what little is written) are very -> totally misleading in this respect and, as Bod said, tell you Nothing at all about how it sounds.
For some strange reason, this doesn't seem to be a consideration for a lot of builders.
Have a good close look at the CURRENT out of your bridge and you'll see that the peak current is over 10A after cap is charged, for an total supply current of about 1.5A rms, regardless of ripple.
Many of the newer compact capacitors have very misleading capacitor ratings and at working temp and power, are not suitable for the job of that first cap - adding a number of them in parallel seems to work okay, but still not the right cap for the job - even when using better diodes (lower reverse current spike) the forward current pulse is a shocker.
The second cap is the main contributor to the sound of the amp and this is where the SIM is useful. Capacitor specs (what little is written) are very -> totally misleading in this respect and, as Bod said, tell you Nothing at all about how it sounds.
For some strange reason, this doesn't seem to be a consideration for a lot of builders.
I know but not many people look at them. They just look at the ripple current spec without puting things into context (eg operating temp).
Never tried the tiles alone. There is a difference with the spikes over no feet for sure. The tiles were to prevent the spikes from digging into the wood. The best sound was with the spikes on a 1" thick marble top with heavy wooden legs which I built for my turntable.
Manufacturers tell about RMS value of the ripple current, not about peaks.
So, to compare apples to apples, i look at RMS value of the ripple in PSUD.
Is that wrong? I really would like to know.
Then, our choice has to go towards caps made for power supplies. Or at least computer grade.
I usually chose screw terminals caps with high ripple capacity, intended for power supplies. Might be overkill but i like how they look in an amp. And they don't even need a PCB.
AS diyers, lets go for the best quality, otherwise let's buy commercial products.
Last edited: