Printable View
Anyone with a scope ever take a look at the transient voltage output of the typical switching PSU's such as those from Mean Well when the unit is switched on/off with and with out load. Despite driving with a flexmod P3, I wonder if the more sensitive diodes might be exposed to damaging voltages from these supplies.
Well, they Mean Well, don't they lol :p. Seriously tho, a transorb or a few zeners in parallel on the output of the ps might be good. My scope wont capture a spike. It's too old. Like me .
For many years I have listened to my mentor in Laser Shows and installed two BIG ones of these in every show system PSU:
http://t0.gstatic.com/images?q=tbn:A...Y6fkMdphvdu8jN
It also helps to add a small time delay relay and a set of precharge resistors to allow the caps to come up, before connecting the filtered PSU to the Amps.
Steve
Depending on the type of load you’re hooking up (transient sensitive) the circuit should have at least a 100uF Low ESR type capacitor on the supply input. Have you actually measured spikes on the PSU output or is this just an assumption? The filter stage of the PSU should prevent such happenings, unless it’s a lower quality PSU.
My OS died a couple of months ago and am waiting till the new year to replace it. The PSU's that I use are the Typical E-bay/Chinese switchers and most of them have been Mean Well brand. Out of ten PSUs two have proven defective (both Mean Well). One simply stopped producing any output voltage and the other dropped it's voltage to the minimum of its set range as soon as the slightest load was applied. So... the quality is a little iffy and I can't measure if any potentially harmful transients are present. If a general purpose filter can be applied to these 5-24V supplies this might provide some insurance unless the P3s already provide this protection and this is just redundant. Do I understand? You would filter the PSU input vs the output?
No, I would put about 100uF low ESR cap on the driver input. The problem with adding filter caps to these cheep china PSU outputs is the question of how much and can the PSU handle the extra capacitance. If the PSU you have experiences voltage drops with moderate to light loads I don’t think that adding additional capacitance is going to be too good for the “already gasping for air china PSU”.Quote:
Originally Posted by planters
OK, that sounds good. Please understand that those two PSUs are no more and I'm not even thinking about rescuing them. Just mentioning them to give an idea of my experience with the typical build quality and hence my concern when I say cheep.
Also, adding capacitors to the output of a power supply, you have to be sure that the PS can handle the inrush current. NTCs are one option.
^How does that work?
When adding capacitors (not ~.1uF decoupling caps, the larger ones;100-10,000uF etc) to the output of a power source /supply, the caps will draw huge amounts of current on start-up, appearing almost as a short-circuit to the source. An NTC (negative temerature coefficient) device placed in the line to the caps will provide a high resistance until the caps are partially charged, then reduce down to very small resistance allowing for normal operation.Quote:
Originally Posted by Xytrell
Linky:
http://www.epcos.com/web/generator/W...zHuijX0WF77w==
Based on your recommendation, I looked into these and they seem elegant, but potentially design sensitive as opposed to simple resistors. If undersized the cap would be well below line voltage and they will have heated to the point that they no longer protect and if over-sized they will not heat suffeciently and the resistance remains high. This is I suspect less of a problem than I am making out, but is there a guideline re cap size/line voltage /NTC size?
-post deleted-
Not that I'm aware of , but probably out there somewhere. Power supply design (non-switching) is something that has always been trial and error for me until I get it right. The basic transformer/ bridge/ cap/ regulator supplies are the ones I like because spikes are less likely to occur. I have never been a fan of the switchers because of the complexities of the designs (more that can go wrong w/ them).Quote:
is there a guideline re cap size/line voltage /NTC size?