I got a nice deal on some Panasonic TH-A series caps 63v 27,000uF 105c types .
The question I have is, can I run them at 64.4v rails at idle? The data sheet says surge voltage is 79v. With load the rails drop to 63.5 at Quiescent load and with music etc they drop further.
Getting another Transformer is not my first choice... I am trying to understand what safety margins are built into these caps...
Data-sheet for these caps is here: Access Denied
The question I have is, can I run them at 64.4v rails at idle? The data sheet says surge voltage is 79v. With load the rails drop to 63.5 at Quiescent load and with music etc they drop further.
Getting another Transformer is not my first choice... I am trying to understand what safety margins are built into these caps...
Data-sheet for these caps is here: Access Denied
You also need to take into consideration mains voltage variations. If you are running at the lower end of the allowable range, if the voltage goes up you will be even further past the 63v rating.
Usually the advice for an unregulated supply is to add 10% (or whatever the local regulations say is the max above nominal) for mains variation as well as whatever percentage the transformer regulation is to the nominal under load voltage for sizing your caps.
I wouldn't risk it.
Tony.
Usually the advice for an unregulated supply is to add 10% (or whatever the local regulations say is the max above nominal) for mains variation as well as whatever percentage the transformer regulation is to the nominal under load voltage for sizing your caps.
I wouldn't risk it.
Tony.
> can I run them at 64.4v rails at idle?
If they blow-up, who will die?
Sure, I've run minor projects with 36V on 35V caps. One actually ran years that way. I think I was very lucky.
But caps the size of yours, IF they blow, will surely spook the cap and stink-up the house. If your face is over the caps when they burst you could be quite sorry.
If they blow-up, who will die?
Sure, I've run minor projects with 36V on 35V caps. One actually ran years that way. I think I was very lucky.
But caps the size of yours, IF they blow, will surely spook the cap and stink-up the house. If your face is over the caps when they burst you could be quite sorry.
The safety margin is what you design in. It is called component derating.
Standard derating for capacitor voltage is 0.6. This is used by all companies/agencies that value reliability for their products. So if you have a 63V rated capacitor - you should not run it higher than 0.6*63 = 37.8V for continuous operation. Any transient voltage conditions should not exceed the max voltage which is 63V.
I'd disagree with the capacitor derating advise in #4. The derating should relate to your mains voltage variation and safety margin. Have you ever measured your mains voltage? Do you know the range over which it is allowed to change, and which you can't really seek any compensation if it goes over?
E-capacitors have a designed in over-voltage capability for short duration - that is a compliance requirement. On higher voltage caps it is +10% of rating, which allows for valve amp loading. That capability allows mains born transients, fluctuations, and short term swells to be handled, but not long term deviations. As such, your 63V rated caps should not really be used for continuous 64.4V operation.
E-capacitors have a designed in over-voltage capability for short duration - that is a compliance requirement. On higher voltage caps it is +10% of rating, which allows for valve amp loading. That capability allows mains born transients, fluctuations, and short term swells to be handled, but not long term deviations. As such, your 63V rated caps should not really be used for continuous 64.4V operation.
Thanks guys.
What does the 79v surge rating ( in the data sheet) specify? I assumed that should take care of AC mains variations. I also tend to disagree with the 60% advice... I have used these caps in a project since at least 2012/13 with the Dc voltages at 62 with the same caps. They were fine, no bulging or explosions.... I really doubt 2 volts would cause a catastrophic failure... but then again, I tend to take chances sometimes... I realize not the same thing, but I have used the old MJ15003/4 with +/- 90v rails in a high bias power amp. Every self respecting manufacturer does rate their products conservatively...
However you guys have given me something to think about...
Is there an easy way to drop a few volts? Eg. A few Diodes in series
What does the 79v surge rating ( in the data sheet) specify? I assumed that should take care of AC mains variations. I also tend to disagree with the 60% advice... I have used these caps in a project since at least 2012/13 with the Dc voltages at 62 with the same caps. They were fine, no bulging or explosions.... I really doubt 2 volts would cause a catastrophic failure... but then again, I tend to take chances sometimes... I realize not the same thing, but I have used the old MJ15003/4 with +/- 90v rails in a high bias power amp. Every self respecting manufacturer does rate their products conservatively...
However you guys have given me something to think about...
Is there an easy way to drop a few volts? Eg. A few Diodes in series
It's a crap shoot and depends largely on the quality and condition of the caps. I've run 150V caps at 150V for years without problems, and I've had 450V caps explode with 300V on them.
I run 600V caps with 590V on them (mainly due to not having the space for series connected parts) and they have been fine, too.
Like PRR said, if they blow up, what is the consequence?
I run 600V caps with 590V on them (mainly due to not having the space for series connected parts) and they have been fine, too.
Like PRR said, if they blow up, what is the consequence?
Read the manufacturer (Panasonic) product catalog application guidelines section to learn very useful informations about operating conditions; the maximum rated voltage is one among several conditions that can influence reliability and product life. A slight overvoltage will increase the leakage current. This will increase the internal dissipation and therefore the capacitor temperature. You will not be able to safely use the capacitor over the entire temperature range, but it probably will work just fine on most applications. Having said this, it is never a good engineering practice to place a component right at the edge of the manufacturer specification, because component tolerances and unpredictable environement conditions may trigger a premature failure. Do not assume that a manufacturer will always place a safe margin on the specification. You must at least test it: what happens if you increase your mains voltage value by 10% and the load is turned off? Does your capacitor become hot? Do you have some means to contain the electrolyte spillage and the short circuit current if the safety valve will open?
If you have an non-potted toroidal mains transformer, you could wind a few turns for each secondary with the opposite direction which takes away voltage at the cost of sligthly higher copper losses... but you can use thick wires here with no problem so total loss will be minimal. This IME is much better than trying to use the full natural voltage and just waste power in a string of series diodes or power Zeners in the secondary outputs (plus you need them in bipolar version).
The Dumm Blonde here was testing / fixing dead PC power supplies back in the late 80's. I got a van full from a PC clone store that went out of business. Many were "defective" returns that had nothing wrong. Others had blown internal fuses since we lived in the lightning capital of North America. I got to testing the "bad" ones by jumping the fuse and plugging them in with a car headlight as a test load. Many just needed a new fuse. I was about halfway through the pile when IT happened. As soon as I stuck the plug in the wall the lights dimmed, there was a loud 60 Hz buzz followed by a HUGE explosion that broke the PC board and embedded some of that cap in the ceiling. There were still fragments of that cap stuck in the ceiling 30 years later when I sold the house. I have seen big caps spew their guts many times. It stinks and makes a mess. Seeing one explode like a large M-80 firecracker less than two feet from your head changes your perspective rather quickly.
When you have multiple caps in parallel and one shorts, the stored energy in the others will make the shorted cap explode.....Learned this one the hard way too.
Have I run caps over the voltage spec, yes. Would I run a ten year old cap over spec continuously in a high current environment. Probably not, as stated what happens if it blows up. Is it inside a thick metal cabinet, or out in the open where the shrapnel could hurt someone.
The power supply caps in a class AB power amp see a ripple current, and a large dynamic current every time there is a big transient or loud bass note. What are the conditions the amp that these cape are going in will see? Low volume simple music, or feeding the subwoofer at a dance club running at the edge of clipping. It makes a difference in the internal heating the cap will see from AC and audio current.
When you have multiple caps in parallel and one shorts, the stored energy in the others will make the shorted cap explode.....Learned this one the hard way too.
Have I run caps over the voltage spec, yes. Would I run a ten year old cap over spec continuously in a high current environment. Probably not, as stated what happens if it blows up. Is it inside a thick metal cabinet, or out in the open where the shrapnel could hurt someone.
The power supply caps in a class AB power amp see a ripple current, and a large dynamic current every time there is a big transient or loud bass note. What are the conditions the amp that these cape are going in will see? Low volume simple music, or feeding the subwoofer at a dance club running at the edge of clipping. It makes a difference in the internal heating the cap will see from AC and audio current.
I have considered this as well. I am glad you brought it up. Should I use the same gauge wire or thicker?
At least the same gauge. What I normally use for bucking windings is #14 or #12 solid core THHN building wire. The insulation is thick enough where you don’t need another over-wrap on the transformer, but thin enough where it doesn’t get in the way. And will stand mains voltage without issue. What I usually do is put a short winding in series with the primary, to add a few turns in phase. This lowers the secondary, reduces the idle current, and it only takes one winding instead of two. Less wire nuts in the assembly.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.