The slits effectively counter eddy currents. The most important factor in cap rolling is where the leads connect. The lowest inductance design requires the in and out be at the same location. Otherwise, the currents enhance, causing proximity eddies.
The aspect ratio also plays into it, as eddies/proximity will actually starve the dielectric at higher frequencies.
Hit the city if you can. I was amazed at the engineering skills they display in their tools from several hundred years prior to Columbus.
As long as you follow it up with some 50 degree...
Oh, almost forgot..do not breath the air...
jn
When I was there for work, I quickly learned that they do not rate alcohol content by "proof". They use the term "degree", which is equivalent to percent alcohol. Least, that's what they told me...
I had thought it meant thermal temp, somewhat like hot sake. Clearly I was incorrect.
When I was on the tarmac for the return flight, we were delayed two hours because the air quality was quite bad. It was like pea soup, but it was not fog.
jn
"degrees proof" is a common, and correct,usage. See Alcohol "Proof" and "Alcohol by Volume": Definitions and Explanations
Sure, they might, but my point was: if these eddy currents were relevant, getting rid of them would translate into a lower ESR, so why doesn't the only producer using this technology (which is no longer IP-protected) provide this kind of information.
Since the manufacturer doesn't provide adequate information, I doubt the relevance of this technology. Just my reinforced caveat emptor reflex when it comes to electronic components specifically advertized for audio.
I agree absolutely with your reflex concerning ad copy, I share it as well.
The producer of these caps are unable to provide technical proof of the solution's merit. They do not know how to test the effect in a reasonable and repeatable fashion.
The consequence of eddy currents in jellyroll caps where the connections are at opposite sides of the foil is a significant dropoff of the measured capacitance as frequency rises, as well as a concurrent increase of ESR.
The problem facing the accurate measurement of the effect is the confounding interplay between the increasing ESR and the resultant reduction of the volume of dielectric which remains involved in the function of "capacitance". Separation of the two effects using a capacitance meter is very difficult. The meter does not distinguish between dissipative loss of the series resistance of the foil, and reduction of the effective capacitance because the dielectric is no longer carrying current. This is the exact same thing I encounter with measurement of inductance where proximity effect raises the resistance vs losses caused by conductor dissipation.
This eddy/proximity effect with a capacitor is most easily thought of by using a very wide stripline model where the eddy's cause the strip to NOT carry current uniformly across it's section...where there is no current being carried, there will be no capacitance within the dielectric.
ps. That earlier link on cancelling capacitor inductance was interesting... I went there thinking the person was using external turns to cancel the jellyroll windings that were between the two connections, or that somebody was returning the cap current over the cap body via a copper braid.
jn
Last edited:
George
Thanks guys, interesting information.
The billboard ads on the side of the roads basically just had 30O , 45O , and 50O, but the rest of the ad was not in English.. My guide explained it when I asked.
jn
jneutron, do you think that the dielectric starving effect of the eddy currents would act to magnify the nonlinearities of skin effect in the capacitor plates? Or perhaps the capacitance modulation itself would be a greater source of nonlinearity, since as I understand it starts with the absolute value of the rate of chance of current?
I would think that slit-film or slit-metallization would be used on high-Q film capacitors rather than electrolytics.
I would think that slit-film or slit-metallization would be used on high-Q film capacitors rather than electrolytics.
I may have missed this item of data, as you gents have been very industrious and I faced a lot of mails, but speaking of phono stages, what should I consider as the minimum acceptable overload margin, ref. 1 kHz?
Theoretically, the more, the merrier. I have seen phono stages with overload margins at 1 kHz as high as 350 mV, with PSU supplies of +/- 45V (I think?), from Pioneer, for example. But is that realistic?
Also, there's the question of cartirge output voltage. I use two cartridges by Ortofon, one (LM 20) is quoted as 0.6 mV/cm/sec, and the other (FF15Mk. something) almost twice that, 1.1 mV/cm/sec. Obviously, the second will sound louder, which is great for a better S/N ratio, but raises questions about the required overload margins, and louder is by no means better sounding all on its own. As a matter of fact, I feel the lower output LM20 is the better sounding one of the two.
All suggestions and views are welcome.
Theoretically, the more, the merrier. I have seen phono stages with overload margins at 1 kHz as high as 350 mV, with PSU supplies of +/- 45V (I think?), from Pioneer, for example. But is that realistic?
Also, there's the question of cartirge output voltage. I use two cartridges by Ortofon, one (LM 20) is quoted as 0.6 mV/cm/sec, and the other (FF15Mk. something) almost twice that, 1.1 mV/cm/sec. Obviously, the second will sound louder, which is great for a better S/N ratio, but raises questions about the required overload margins, and louder is by no means better sounding all on its own. As a matter of fact, I feel the lower output LM20 is the better sounding one of the two.
All suggestions and views are welcome.
Even the FF15 is just 5.5mV at 5cm/sec, which is fairly middle-of-the-road. No MM phono stage should overload with that.
There are too many factors affecting sound quality from a cartridge. I see the LM 20 has a fine-line stylus, so probably trackas very well. I once had an old F15E, which was nice but no world-beater.
There are too many factors affecting sound quality from a cartridge. I see the LM 20 has a fine-line stylus, so probably trackas very well. I once had an old F15E, which was nice but no world-beater.
Yes, you're right, the FF15 is no world beater, but it has a usefully high output and as such I use it for very old LPs, some of which had low output levels. For newer LPs, those made after say 1970, the LM20 does the work.
Actually, you may remember this, LM 10/20/30 all had the same magnet assembly, it was the stylus which differed. Mine is the LM20 marked body, but the stylus has been replaced with the one from LM30.
And again, you're quite right, their tracking ability is really very good. It pays off with lively, complex passages and/or very fast transients.
Thank you, that clarifies matters significantly.
I think this is where my tape-it-and-listen philosophy has paid off handsomely; most of my LPs are almost as good as new, very little damage, very little wear and tear, just a few warps on odd samples, so consequently I don't get much HF clicks and pops.
Although I still regret selling my SAE 5000 click eliminator. Ah, the folly of youth!
Actually, if I take SY's post at literal face value, it turns out that using op amps for the phono stage is quite viable. Not the best we can do, but reasonably good for modestly priced units. For the very best, it's back to discrete, I think.
Assuming +/- 15V rails, they should be able to handle around 85 mV/1 kHz worst case without too much fuss.
Assuming +/- 15V rails, they should be able to handle around 85 mV/1 kHz worst case without too much fuss.
Not really. The dielectric starving is the result, not the cause of the eddy current exclusion.
Yes, that is what I would expect. Most people do not think of a capacitor as being rate of change of current sensitive with respect to the absolute capacitance of the device. Therefore, very little consideration has been given to how to actually test it.
For speaker crossovers as an example, the amount of capacitance is pretty high in comparison to high Q film capacitors.
Think of the slit film electrolytics in the same way one would think of air core litz wire inductors. The slits and the litz perform the exact same function, reduction of dI/dt effects within the lumped element. The capacitive problem is of course, less obvious.
jn
Suprise you can still eat pig trotters , its been sometime since i had some (too much fat for old parts
) miss oxtail too, man nothing like fat for taste..Anyone took alook at the manley phono stage ..?
Not that I have any desire to eat cows brains, after all the problem a few years ago no way would I touch that. Some of those other parts I don't really know what they are. I'll stay away from the tripe unless it comes from a pig and is on a hot dog, nice crunch. I think we use the heel here to make gelatin as far as I know.
Same here.
A guy from three houses down the block was lost to CJD. 3 months from initial symptoms to death. Not pretty. Not even time to get one's life in order.
jn
- Status
- Not open for further replies.