Chris,Using sufficient design margin in your MMC may cost more initially, but saves you in longer-term costs. It also increases long-term up-time and reliability. An async gap can overvolt your MMC when operated too slowly, and your MMC is subjected to mains-resonant voltage rise. Also, as David mentioned, your MMC must be able to withstand the high-voltage polarity reversals that occur during normal tank circuit operation.
If your async gap firing rate becomes too low, the HV source leakage inductance and tank capacitance form a series LC circuit that causes your MMC's voltage to abnormally rise. If the gap fails to fire over multiple AC half-cycles, series resonance will ring-up the MMC voltage to increasing higher multiples of normal voltage... until something breaks. A properly-set safety gap across the async gap prevents this from occurring - cheap and effective.
As David suggested, don't scrimp on the voltage rating of your MMC - its a false economy. For maximum capacitor life, the maximum peak-to-peak voltage reversal should not exceed the combined DC voltage rating of each of your MMC series strings. To determine the minimum number of caps in each string, multiply your HV transformer's maximum RMS output voltage by 2.5 to 3, and then divide this by your individual MMC capacitors' DC voltage rating. If you plan to over-drive the HV source from your variac(s), take this into consideration in your calculations.
BTW, not all MMC caps self-heal under repetitive voltage overstress. Some percentage will eventually fail short-circuited, sometimes after heavy internal arcing. Shorted caps place additional voltage stress on the remaining caps in the string. Using a somewhat higher multiplier (such as 3) provides additional margin, reducing the odds of early cascade failures for the other caps in the string. A small increase in MMC string length goes a LONG way to delaying or preventing premature cap and string failures in your MMC.
Bert Chris Boden wrote:
We're putting together over a thousand of the 942C series caps for our new lab MMC. The last lab MMC was 1000 of them, this one will likely be 1500. It's a lot of work, but it's worth it for us where we need to change things around quite often. I'm just wondering if ASYNC will be bad for them. We're getting in to working on some new (and one radical) RSG designs and I'd like to not be killing caps a thousand at a time. ;) On Sun, Feb 25, 2018 at 9:21 PM, David Rieben <drieben@xxxxxxx> wrote:Chris, All other things being equal, about the only way that I can see for async operation to be any more distressing to the primary capacitor(s) than sync operation is with the typically available higher break rates that are associated with a proportionate increase in RMS current (but have little effect upon PEAK current). At the end of the day, any kind of SG TC duty is rather rigorous on capacitors and low dielectric loss, pulse rated caps (high dV/dT) with appropriate peak and RMS current ratings, in addition to an ample voltage ceiling rating (generally a VDC rating of at least 2.5X the maximum RMS voltage from the transfomer) is essential for a reasonable expected life span of said capacitor(s). The "self-healing" CD-942C snubber caps have proven to be quite reliable in the MMCs built for mid-level powered NST driven systems (even with a considerably lower voltage ceiling than the suggested 2.5X). However, the sheer voltage/capacitance requirements for the primary cap of a typical pole pig driven system often dictate single unit, high voltage commercial pulse caps as a more practical and economically viable solution (and much less labor intensive!) than soldering HUNDREDS of individual snubber caps in a series-parallel configuration. David ----- Original Message ----- From: "Chris Boden" <cboden@xxxxxxxxxxxxxxxx> To: "Tesla Coil Mailing List" <tesla@xxxxxxxxxx> Sent: Sunday, February 25, 2018 6:42 PM Subject: Re: [TCML] Best Spark Gap for NSTs? I'm curious, what are the dangers of running an ASYNC gap with a large(pole pig) setup? I'd always heard it was damaging to caps, etc. Is it acceptable to run totally async and not have to worry about destroying caps? On Sun, Feb 25, 2018 at 4:15 PM, David Rieben <drieben@xxxxxxx> wrote: Yes, I concur with John's assessment of the sync vs. async debate. Iexperienced the very thing that John referred to where my output spark length deminished after converting my sync SG over to sync. Just as he mentioned, I left the original STR (0.1 uFd) primary cap unchanged. Since we are talking about a pole pig fired beast, it has proven far more practial (and economical) for me to remain async with my system, as big bright sparks were my goal, anyway. A truly LTR cap, whose voltage standoff and capacitance can match up with a 14,400 volt,10 or 15 kVA rated pole pig in sync operation would have to be a real monster (> 0.3 or 0.4 uFd @ 25 kVAC or 60 kVDC) and a real drain to the typical hobbiest's budget! Probably the only type of TC drive sytem that could possibly complete with or possible beat the asnc rotary gap for pole pig fired coil systems of equivalent power levels, whose endgame is the biggest and brightest sparks possible, would be the DC resonant system. Of course, the required giant storage cap(s) - (up to several 10's of uFds @ >20 kVDC) and ultra fast recovery, high voltage diodes also render this approach considerably more pricey that the standard async system, too. Sync RSG with LTR primary caps is the best way to coax maximum spark length from the limited available power, while at the same time going easier on the fragile HV side windings of (an) NST(s), though. David ----- Original Message ----- From: "Futuret via Tesla" <tesla@xxxxxxxxxxTo: <tesla@xxxxxxxxxx> Sent: Sunday, February 25, 2018 9:52 AM Subject: Re: [TCML] Best Spark Gap for NSTs? The answer is not simple or direct. There are advantages anddisadvantages for sync or non-sync. Did you happen to try feeding the motor with 140V from a step up type variac? Often this will keep it in sync. If you created flats on the rotor that are at least 1/4 the width of the full rotor diameter, that is enough metal to remove. Some small motors are very wimpy and don't work well such as small shaded pole fan motors. And of course you should have 4 flats on your motor. Yes the preferred break rate has a lot to do with it. Sync operation is most beneficial at low break rate of 120 bps. However it's absolutely necessary to use a large enough capacitor to take advantage of this. Back in the "day", a lot of people were installing a sync gap onto their TC and getting worse results because they were using a small capacitor with a static gap previously. They left the small capacitor in place and therefore saw a reduction in spark length when they installed the sync rotary gap. For anything over 120 bps, sync doesn't really help, you might as well use non-sync or a static gap. Gary Lau posted some information about the best capacitor sizes for sync or non-sync, etc, in the past and I guess this is at his website. Also JavaTC might give some information about the best capacitor values for sync or non-sync, for NST powered coils. 120 bps sparks will not be quite as bright as higher breakrate sparks. Also they tend not to create the kind of long tendrils that can form at higher breakrates. Certain coils tend to show these long thick tendril type sparks. John -----Original Message----- From: Daniel Kunkel <dankunkel@xxxxxxxxx> To: Tesla list <tesla@xxxxxxxxxx> Sent: Sun, Feb 25, 2018 9:45 am Subject: [TCML] Best Spark Gap for NSTs? I modified a 1725 rpm motor to run in salient pole operation years ago. It will seek and then lock into phase and then break out again. I think I recall reading that if you don't remove enough rotor material it will seek. If you remove too much it will over heat. So before to I take this back to machining, is it even worth it? What is the best gap for NSTs? Is SRSG is the answer what is the preferred break rate? ~Dan To synch or not to synch Kansas City area _______________________________________________ Tesla mailing list Tesla@xxxxxxxxxxxxxxxxxx https://www.pupman.com/mailman/listinfo/tesla _______________________________________________ Tesla mailing list Tesla@xxxxxxxxxxxxxxxxxx https://www.pupman.com/mailman/listinfo/tesla_______________________________________________ Tesla mailing list Tesla@xxxxxxxxxxxxxxxxxx https://www.pupman.com/mailman/listinfo/tesla-- Chris Boden President The Geek Group National Science Institute www.thegeekgroup.org -- This email may contain confidential and privileged material for the sole use of the intended recipient. Any review or distribution by others is strictly prohibited. If you are not the intended recipient please contact the sender and delete all copies. _______________________________________________ Tesla mailing list Tesla@xxxxxxxxxxxxxxxxxx https://www.pupman.com/mailman/listinfo/tesla_______________________________________________ Tesla mailing list Tesla@xxxxxxxxxxxxxxxxxx https://www.pupman.com/mailman/listinfo/tesla
_______________________________________________ Tesla mailing list Tesla@xxxxxxxxxxxxxxxxxx https://www.pupman.com/mailman/listinfo/tesla