|
|
Post by ourclarioncall on Oct 3, 2020 17:29:22 GMT -5
Would you do it his way ? Or is there a better method? I’m not going to try it, I’m just curious
|
|
|
|
Post by thetragichero on Oct 3, 2020 19:23:35 GMT -5
basically
|
|
|
|
Post by sumgai on Oct 4, 2020 22:53:51 GMT -5
See your other thread about rubber gloves. Therein, we learned that putting both hands into the amp to measure a voltage is asking for problems of the unpleasant kind. The guy in the video above just blithely stuck all of his mitts in there like he was picking up a Sloppy-Joe sandwich. Even at his measured 8 or 9 volts, that's just not a good habit to get into. Again, see the other thread. Two, he says "anything less than 20 volts". Say what?!? How about anything period, larger as well as smaller. Three, why wait for one hour, unplugged at that?  I mean, if time is money, then every amp tech on the planet would die of starvation if they had to do this over and over, just to find a problem. That was going too far in the name of safety, which doesn't happen often, but this is a good example of when it does happen. One should wait for perhaps half a minute, but anything longer is nonsense. Four, use a proper cap-discharge tool, one that does not require two hands. In 99% of all cases, you're going to be discharging a cap that's filtering AC out of the power supply line(s), so go ahead and use the Chassis as one side of the connection. I use a well-beat-up screwdriver to make the contact with the appropriate lead, and watch the spark. Sometimes, especially in very high voltage situations, I'll touch the lead two or three times - caps can actually accumulate another, smaller charge from the very discharging action, believe it or don't. Five, regard anything that reTrEaD says after this post as also-good-to-know information. He has previously "discussed" with me the benefits and downfalls of my methodology, which is to say, he takes a bit of an issue with my statements above. Perhaps what he has to say will appeal to you more so than my pontifications. HTH sumgai p.s. If you want a short tutorial on why this procedure is necessary, just ask.  |
|
|
|
Post by b4nj0 on Oct 5, 2020 2:58:06 GMT -5
When I was a slip of a lad in the 1970s, I decided that the mains lead on my AC30 had withered too far and decided to replace it. The recalcitrant brute had been turned off and disconnected from the mains for a good eight hours. In I went and Pow! Threw me nine feet across the room where (happily) I landed on a sofa situated inside an alcove. I reckon that was the main electrolytic(s) and I was the crowbar, so I figured out that they probably didn't need replacing too.
I could go on listing my "accidents" but I've stuck at this relevant example. Of course if Jennings had stumped up for the bleed resistor(s) it wouldn't have occurred, but something else would have eventually! And thus began a lifetime's fascination with Coulombs.
e&oe ...
|
|
|
|
Post by thetragichero on Oct 5, 2020 10:09:04 GMT -5
once you've watched an old amp take hours to discharge you appreciate the added expense of a bleed resistor on one of the filter caps
|
|
|
|
Post by ourclarioncall on Oct 5, 2020 10:34:05 GMT -5
Sumgai
Sloppy joe sandwich , haha that made me laugh out loud 🤣
Yes please , bring on the tutorial , I’m fascinated with this stuff
|
|
|
|
Post by reTrEaD on Oct 5, 2020 10:40:16 GMT -5
once you've watched an old amp take hours to discharge you appreciate the added expense of a bleed resistor on one of the filter caps QFT |
|
|
|
Post by sumgai on Oct 5, 2020 15:54:10 GMT -5
p.s. If you want a short tutorial on why this procedure is necessary, just ask. Yes please , bring on the tutorial , I’m fascinated with this stuff Well alrighty then!  As B4nj0, trag and reTrEaD all alluded to, a "bleeder" resistor is normally installed to discharge the capacitor(s) after the power has been shutoff. These are arranged in a parallel configuration, directly across the cap(s). Consider the background here: A capacitor can easily pass AC through it, but it blocks DC, which is why we can use it to filter AC (shunting it to ground in the power supply circuit), whilst allowing the DC part of the voltage to proceed merrily along its way (to the tubes, etc.). That filter cap is expecting to see voltages approaching the nominal output voltage of the transformer and rectifier components. This is because the rectifier (be it a tube or solid state diode) is really cutting off only half of the AC signal coming out of the xformer - the remaining half is still going from Zero to Full Tilt voltage, but only in one direction (usually positive, but the circuit can be built to need a negative voltage, when so desired). Now, that positive voltage is what we call "pulsating DC", and it should be obvious why, but the fact is, we need it to be as close to pure DC as we can get. Enter the Filter Capacitor. I've already said that caps pass AC, and we already know that the filter cap is hooked up directly between the power supply output line and ground, so the next obvious question should be "isn't that a short circuit?". Go to the head of the class. It is indeed a short circuit, but one of such low amperage that the capacitor can shrug it off.... provided that it is rated for the desired voltage. Thus the pulsations are greatly reduced, hopefully as close to Zero as possible. (BTW, those pulsations are usually called "ripple".) And now we come to the part about the bleeder resistor. An apt description, we use one to "bleed" off the remaining charge from a cap, because as we all should know, a cap can retain any charge upon it for long periods of time. Without it, a cap could ostensibly hold a charge for years! And thus we see videos about how to discharge a cap manually.... which started this whole thread. Yes, the resistor is also a direct short between the output line and ground, but.... it's a high enough value to present a load so small when compared to the rest of the circuit that it won't cause problems. When the load of the active components is removed, and the power remains on, then these resistors will get hot, in a hurry. That's why they're often oversized, to help absorb some of that power for a short time.* And not to be too modest about it, but this is where knowing what one is doing when designing a power supply (let alone a whole amp) is of prime importance. Just ask b4nj0 about how Jennings and Denney (neither of whom were Engineers in the classical sense) decided that such "niceties weren't really necessary". Manufacturers of this stripe are damned lucky they aren't sued into the poor house. OK, enough about that. Any questions? (Silly of me to ask, of course you'll have more questions! )
sumgai * Actually, those resistors don't actually "absorb" the excess energy, they transform it into heat. But that's a nicety that doesn't need to be discussed here and now. I just wanted to stave off any possible pushback.  |
|
|
|
Post by blademaster2 on Oct 6, 2020 9:29:07 GMT -5
Okay - here is a question/comment for you, Sumgai:
The bleeder resistor, if connected across the power rail at all times (I think that is what you said), should dissipate heat from the small amount of current passing through it at all times and not just when the system is turned off. It would need to be sized such that the added power demand and dissipation was tolerable. If the On/Off switch was, say, DPDT, then the off position might be configured to connect this bleed resistor to ground, but only when the system is switched off - saving heat and power loss needlessly at other times.
I, too, have tried the 'old screwdriver' method with a high-power audio amplifier to discharge the caps. I actually waited for many minutes after shutting it off and measured the rail voltage until it was around 18 volts (operationally these sat at +/- 90Vdc). Then I grew impatient and used the screwdriver to take the rest of the charge away - and the arc melted off the corner of the screwdriver even at that low voltage.
I would definitely look at a resistor, suitably sized for power rating and bleed current, if I do this in the future. No component, caps, screwdrivers or otherwise, are happy to see current of that magnitude passing through them.
|
|
|
|
Post by thetragichero on Oct 6, 2020 12:34:17 GMT -5
power formula P = I x V
ohm's law (rearranged) I = V / R
so P = V² / R
let's use 600v because that's larger than B+ on most amps and 470k because whatever (i usually use 1M bleeder resistors, the standard fender totem pole balancing/bleeder resistors is 2x 220k)
P = (600V x 600V) / 470,000Ω = 0.766W
good idea to derate by 2 so 470k 2W
i wouldn't like all that dc going through the switch myself, when the amp is operating 470k or 1M is (as mentioned) a way higher impedance than the rest of the components in the circuit so making it switchable provides no real benefit and more complexity/risk for downside
|
|
|
|
Post by Yogi B on Oct 6, 2020 12:41:24 GMT -5
If the On/Off switch was, say, DPDT, then the off position might be configured to connect this bleed resistor to ground, but only when the system is switched off - saving heat and power loss needlessly at other times. That does also rely on remembering to flip the on/off switch, to the off position -- which might not necessarily happen if switching off at the wall instead (prior to unplugging). |
|
|
|
Post by reTrEaD on Oct 6, 2020 13:27:44 GMT -5
If the On/Off switch was, say, DPDT, then the off position might be configured to connect this bleed resistor to ground, but only when the system is switched off - saving heat and power loss needlessly at other times. This looks rather brilliant until we recognize the problem brought up by Yogi B . Another issue worth noting: Bleeder resistors can also degrade the filtering, but unless one is overly aggressive in designing for an extremely brief discharge time, this isn't really a factor. Discharge time of say 5~10 minutes to bring the cap to a 'safe' level (non-lethal in the case of tube amps, unlikely to arc weld in the case of solid state amps) seems reasonable to me. Also, we need to recognize there are often stages to the filtering that will affect the discharge time if we're relying on a single bleeder resistor.
Getting back to the video ... Why did he put the resistor in the middle of the wire? Just an extra soldering connection that wasn't necessary. Place the resistor at one clip or the other. (But actually, I wouldn't make such a cliplead system in the first place.) I think it would make far more sense to use a 'stackable' dual banana plug for a meter. Attach the resistor between the two plugs using the hold-down screws. Then plug a lead with a banana plug on one end and an aligator clip on the other end to the 'ground' side of the stackable. A lead with banana plug on one end and a test probe on the other plugs into the 'hot' side of the stackable. That allows you to make a ground connection with your alligator clip and contact the hot side of the cap with the test probe. You can read the voltage while you're discharging the cap. Naturally, if your resistor is such that the discharge time is long, you would use an alligator cliplead on the hot side instead of the test probe lead. And of course you could have a few of these set up with different resistance values for different applications. The resistance for a bleeder for a solid state amp would be much lower than one for a tube amp.  |
|
|
|
Post by blademaster2 on Oct 6, 2020 15:03:25 GMT -5
If the On/Off switch was, say, DPDT, then the off position might be configured to connect this bleed resistor to ground, but only when the system is switched off - saving heat and power loss needlessly at other times. That does also rely on remembering to flip the on/off switch, to the off position -- which might not necessarily happen if switching off at the wall instead (prior to unplugging). Hmmm. I am in Canada, so we do not generally have wall-switching for the outlets (I wired my studio for that, but I would never use that in place of the amplifier's power switch). Otherwise if there is no "Centre-off" for the DPDT you could get that bleeder in there and never need to give it another thought. However ... I would not expect manufacturers to add this to their products only to make servicing their equipment safer - that seems like something they would discourage (unless by qualified service personnel, yadda yadda, who already know the hazards). |
|
|
|
Post by ourclarioncall on Oct 6, 2020 15:11:47 GMT -5
Really enjoying the discussion gentlemen , keep it coming
So does this guy win a prize for the best diy discharger? Or could it be improved ?
|
|
|
|
Post by thetragichero on Oct 6, 2020 15:18:34 GMT -5
I've seen similar ideas using a wooden handle. wood being an insulator is good. don't want my hand anywhere near the resistor (likely to get hot) and the surrounding connections (as i found out on one build the heat shrink may not offer great protection when there's current passing through)
|
|
|
|
Post by ourclarioncall on Oct 6, 2020 16:05:27 GMT -5
I've seen similar ideas using a wooden handle. wood being an insulator is good. don't want my hand anywhere near the resistor (likely to get hot) and the surrounding connections (as i found out on one build the heat shrink may not offer great protection when there's current passing through) I love it when a good idea gets scrutinised then exposed as not as a great as it could be .Feeling the heat could possible be a positive thing? As it’s a sensory proof that what you did actually had an effect . A bit like watching a multimeter drop. The only thing I would critique, it the aligator clip that clips on the chassis is not insulated so wouldn’t there be a chance that your other hand or even your right hand with the probe might touch the aligator clip at the same time you stick the probe in? Or would the resistor slow down the release so much that it wouldn’t hurt that bad if you did touch it ? Think I’d prefer the aligator clip to be permanently connected to the bananna plug thing as well, I don’t like that it’s loose, well not loose but can come loose .Thanks for the feedback. I tend to research the four corners of a topic until I see the lay of the land before making a choice So far I’ve seen for cap discharge methods I’ve seen 1. Screwdriver 2. Light bulb 3. Multimeter 4. Alligator clip > wire > resistor (and solder) (covered in heat shrink) > wire > alligator clip 5. Copper rod (with heat shrink ) > resistor (with heat shrink) > wire > aligator clip 6. Retreads suggestion which I can’t remember but am interested in so will read again 7. Trags alternative wooden handle 8. My imaginary one that is unbeatably fast and safe that is a combined multimeter, oscilloscope, light bulb, heat sensing , Sparking etc. All the benefits of current methods plus more . For the ten year in us it will still give off a spark while lighting up a bulb which then dims, plus you get to feel the heat pass through the Multimeters external heat shrinked resistor and watch the numbers go down on the meter and the visual representation on the oscilloscope screen. There must be a few more I’d be interested y’all’s opinions on to know which is best to worst , Isn’t there something like this on the market already and if not why not ? I know I would buy one IF it’s safety superseded all other methods. Price is not a concern. It’s a tool for life that can save your life . There you that’s the marketing strategy tag line “A TOOL FOR “LIFE” “ “LIFETIME GUARENTEE” “THE SAFEST CAPACITOR DISCHARGE TOOL IN THE WORLD” only $99.99 |
|
|
|
Post by ourclarioncall on Oct 6, 2020 17:03:58 GMT -5
Couple of questions
1. With a cap discharge tool , where does the electricity go once it comes out of the cap? Does it turn into heat and disappears ?
2. What is a standby switch on amp for ? How does it work ? Do you really need one ?
|
|
|
|
Post by ourclarioncall on Oct 6, 2020 17:40:39 GMT -5
|
|
|
|
Post by ourclarioncall on Oct 6, 2020 17:42:54 GMT -5
|
|
|
|
Post by reTrEaD on Oct 6, 2020 18:10:43 GMT -5
Isn’t there something like this on the market already and if not why not ? As you've already found there is something like this on the market. (although there should not be.) Top Three Reasons why a commercially available 'capacitor discharge tool' should not be on the market. Reason #3: Real men design their power supplies with appropriate bleeder resistors. Reason #2: The appropriate size of the discharge resistor depends on both the value of the capacitor and the voltage applied to it. For instance, the resistor employed to discharge a 20uF cap charged 500v (think tube amp) would be woefully slow in discharging a 50,000uF cap charged to 50v (think solid state amp). Likewise, an appropriately sized resistor to discharge the 50,000uF cap would be WAY too fast for the tube amp. One size does NOT fit all. Reason #1: Careless people who don't know what they are doing should not be mucking around inside an amp. If lethal voltages are present, we have the opportunity to cull that particular herd.
😮 |
|
|
|
Post by ourclarioncall on Oct 6, 2020 18:23:47 GMT -5
reTrEaD so do any amps have these bleed resistors in them ? If not would it be easy/possible to to modify say a classic fender amp to have them ? the video I linked, was that similar to what you were describing ? The double banana plug with the resistor I saw aanother video with a guy discharging caps with just his multimeter. It was quite slow. how is that working ? Is it because there is a resistor or I assume many resistors inside the multimeter ? would using a multimeter be a good idea since you can kill two birds with one stone, 1. Discharging the caps 2. Making sure you really did discharge them by seeing it on the meter or is there the chance of killing 3 birds ? Dead multimeter i agree with number 1 on your list ! I have a healthy and safety streak in me. Get those bleed resistors into action amp builders
|
|
|
|
Post by sumgai on Oct 6, 2020 18:31:25 GMT -5
Then I grew impatient and used the screwdriver to take the rest of the charge away - and the arc melted off the corner of the screwdriver even at that low voltage. I'm sorry, did I mistakenly give the impression that the screwdriver is still capable of performing its original design function? My bad, I didn't meant to do that.  I would definitely look at a resistor, suitably sized for power rating and bleed current, I really do have to stop rushing my answers, sigh. Yes, I use an old 5-watt 1kΩ power resistor (the square kind) now (all connected with some 16ga stranded speaker wire I had laying around), but not the first time I ever needed to do the job. Back then, I just picked up the nearest handy tool that could be sacrificed, and had at it. That was along time ago, trust me. BTW, trag's short tutorial on finding the right values for a bleeder (or more than one) is on the money. And yes, best practice says that you shouldn't depend on one resistor for the entire chain of several caps (which are isolated with variously-valued resistors for different voltages), you should use a bleeder in between each stage. But I do admit, this is best practice, and some designers will only put a bleeder on the first filtering stage, and then one more at the end of the rest of the filtering stages. Sometimes a compromise like that works out to be safe enough, but I'd have to under some pretty tense pressure from the bean counters before I'd go that far. Speaking personally, of course. In other news..... No, heat shrink is not meant to handle power greater than about 1 amp, though many users to try to use to replace Electrician's Tape. When I do that, I double up on the tubing, in stages of course. For expected current flow of greater than 2 amps, I break out the real-meal-deal, or else un-cork a bottle of wire nuts. Depends on how much room I have to play - my fingers ain't so nimble any more.  HTH sumgai |
|
|
|
Post by sumgai on Oct 6, 2020 18:39:50 GMT -5
Couple of questions 1. With a cap discharge tool , where does the electricity go once it comes out of the cap? Does it turn into heat and disappears ? I actually answered that in my treatise, at the very end of the message - the energy is converted to heat, hence the reason for a resistor be large enough to handle that kind of power (volts times amps, and all that). Strictly speaking, that's off-topic. Please start another thread about standby switches, and I'm sure we'll get way more than one reply. HTH sumgai |
|
|
|
Post by ourclarioncall on Oct 6, 2020 18:49:47 GMT -5
sumgaijust heat ? i thought that but also thought because it’s clipped to the chassis that it going from the chassis to ground . So it’s being grounded , or some of it is turning into heat but some of it is going to ground. but then the amp is not plugged in so it can’t get to ground 🤨🧐 new standby switch started 👍
|
|
|
|
Post by reTrEaD on Oct 6, 2020 19:18:38 GMT -5
so do any amps have these bleed resistors in them ? If not would it be easy/possible to to modify say a classic fender amp to have them ? Fender didn't make a habit of intentionally adding bleeder resistors but some of their amps used equalizing resistors which had the side benefit of also acting as a bleeder. Here's a Super Reverb schematic with no bleeders and two 20uF caps in parallel (40uF total) as the primary filter: This is a slightly later schematic for the same amp. It uses two 70uF caps in series (35uF total).  The series arrangement (and the 220k equalizing resistors) allow caps to each see only half the voltage. They also have the same effect of having a 240k bleeder resistor in parallel with the 35uF total capacitance of that filter section. |
|
|
|
Post by ourclarioncall on Oct 7, 2020 5:25:04 GMT -5
Another good reason to buy a super reverb then 😊
If my wife objects I’ll just say but retread says it’s safer because of the ....
|
|
|
|
Post by thetragichero on Oct 7, 2020 13:20:10 GMT -5
just heat ? i thought that but also thought because it’s clipped to the chassis that it going from the chassis to ground . So it’s being grounded , or some of it is turning into heat but some of it is going to ground. but then the amp is not plugged in so it can’t get to ground is inside conservation of energy would dictate the potential difference (as that is what voltage is a measurement of) stored in the capacitor(s) would have be converted into *some* sort of energy, whether it be electrical (the spark using a metal tool. i would highly suggest not using this method for discharging caps) or heat (as the resistor heats up as current follows through it) so ground. it's all relative the ground you're speaking of (earth ground, the bottom plug on your electrical outlet) is connected at one side to a bolt right inside the chassis, and the other side to 6-8ft metal rod(s) buried into the ground (also generally the cold water pipe). the neutral (the wider of the two plugs) is also connected to earth ground at the service entrance now circuit ground is something completely different. since everything after the rectifier is dc think of anything between the input and output transformers as a circle (sounds like circle, huh? probably greek): at rest (powered down) everything *wants* to be equal, so when you've got an area of high potential energy (like a capacitor) and low/zero potential energy (like circuit ground) it *wants* to be "0V". the fact that circuit (DC) ground is usually connected to the chassis (for our purposes) and earth/safety (AC) ground is also bonded to the chassis is not coincidental but practical that's not exactly an EE explanation but hopefully sheds some light |
|
|
|
Post by b4nj0 on Oct 7, 2020 18:21:28 GMT -5
Trag that has to be regional. Over here we have PME (protective multiple earthing) Everything metallic in a building should be bonded to neutral and find its true "ground" back at the substation. Due to the potential length of conductors involved, there is a literal potential difference (however slight) between earth stakes at the premises and neutral. This is not a happy situation.
Personally I have to take RF "ground" into consideration too which has similar implications for PME. So far as I'm aware, it's not illegal for me to disconnect PME and install my own earth stakes, but I wouldn't want that responsibility. I stick to balanced antennas for that reason but I'm swerving way OT. Next up I'll be on about lightning "protection" so I'll get me coat now!
e&oe ...
|
|
I mean, if time is money, then every amp tech on the planet would die of starvation if they had to do this over and over, just to find a problem. That was going too far in the name of safety, which doesn't happen often, but this is a good example of when it does happen. One should wait for perhaps half a minute, but anything longer is nonsense.
