Post by thread:[EE:] Commutating diode selection and snubber circ

> And what should I consider in designing a circuit to make this coil quiet. It operates at 24V .46A. /> Has to pass max current and withstand max voltage. 1N400X probably OK there. Low speed but unlikely to be a problem as long as this is off and on for longish periods at a time (eg >= 0.1 second) and not PWMd. . Want to get fancy use eg BVY26 or equivalent - but probably no need. >I guess it would just have to have a reverse bias brake down voltage grater than 24V and be able to handle about half an amp of current. /> Yes > Will the coil have enough internal resistance to dissipate the power or should I have a resistor in series with the diode? I am very green at this and unsure of myself so don't thrash me too hard and thanks in advance. /> Coil may well be OK. 24V at 0.5 amps or so suggests 48R coil. If so power dissipation will be I^2R = 12W at max current. No coincidence that this also eqwuals 24V x 0.5A. Adding extra R will increase flyback voltage of coil. Maximum disispation PROBABLY about when Rexternal = R coil (past bed time, brain protests, it says that transfer theorum probably applies, ask it in the (this) morning. HOWEVER all the above may not work. Description is a little uncertain. If the brake is a generator then adding the diode may make it self brake when you don't want it to. When you say "electromagnetic brake" do you mean eddy current, or electromagnet drives a brake show pair or ????? Russell McMahon -- http://www.piclist.com hint: To leave the PICList .....piclist-unsubscribe-requestKILLspam@spam@mitvma.mit.edu

At 10:25 AM 8/23/2004, Roberts II, Charles K. wrote: >The Brake is engaged when no voltage is applied, the break is disengaged >when there is a voltage. When I took it apart it looked as if it used a >spring to hold the brake shoe onto the shaft and a solenoid to release >it. >That's all I got, there is not much information on the manufactures >site. The diode has to have enough PRV to withstand the on-state. Then the energy stored in the inductor needs to dump through it's own resistance, so that defines the maximum one-shot energy the diode needs to sustain. You don't often see the math for this presented, I suppose it's because generally speaking, a 1N400X diode will handle most loads. I've tried to get this sort of info out of stepper chip makers. They say they have on-chip diodes, but when you press for details, they are often unable to tell you how much the diodes can handle. -- http://www.piclist.com hint: To leave the PICList piclist-unsubscribe-requestKILLspammitvma.mit.edu

One thing to watch out for when putting a commutating diode on a solenoid or relay coil is the fact that it will not loose its magnetic field as quickly as it did before the diode was added. This is because the commutating diode for all practical purposes, feeds that energy back in to the coil such that it keeps itself energized for a fraction of a second while the magnet dumps its stored energy through the now forward-biased diode. I was made aware of this as a small boy when I connected a relay with normally-closed contacts such that it buzzed when energized from a battery. The buzz or chatter was normally fierce with nothing across the coil. If you touched the coil, the shock was painful and there was enough voltage released in the field collapse to light a neon bulb. If I put a diode in reverse across the coil, the chattering turned to a faint buzzing sound and the high-voltage spikes along with AM radio static were gone. It was obvious that the relay couldn't release quite as fast as it used to. If split-second timing is important, remember that the solenoid will energize as quickly as before, but it may hang up for a split-second before releasing. -- http://www.piclist.com hint: To leave the PICList .....piclist-unsubscribe-requestKILLspam.....mitvma.mit.edu

Wouter van Ooijen wrote: >> If split-second timing is >> important, remember that the solenoid will energize as quickly as >> before, but it may hang up for a split-second before releasing. > > If timing is important you can use a zener in series with the diode, > or even a resistor. The voltage at the 'low' side of the coil will be > higher, so you must decide what the maximum voltage there can be. Another trick to increase the response time of a relay is to put a parallel capacitor and resistor in series with the relay coil. Make sure the time constant of R*C is less than half or so of the minimum relay on or off time, then set the R so that the steady state current thru the coil is just the "holding" current, but not the "pull in" current. The capacitor initially allows more current to flow when the relay is switched on, but the current will fall to a lower level once it has been engaged. This lower current and therefore lower magnetic field is quicker to switch off. ***************************************************************** Embed Inc, embedded system specialists in Littleton Massachusetts (978) 742-9014, http://www.embedinc.com -- http://www.piclist.com hint: To leave the PICList piclist-unsubscribe-requestspam_OUTmitvma.mit.edu

> The Brake is engaged when no voltage is applied, the break is disengaged when there is a voltage. When I took it apart it looked as if it used a spring to hold the brake shoe onto the shaft and a solenoid to release it. That's all I got, there is not much information on the manufactures site. /> Marks assigned: 100 Time allowed: No limit Answer ALL questions Is the brake mostly either fully on or fully off or is it modulated in some way to be partially off or on. eg do you turn it off (by applying power), roll the stage somewhere and then turn the brake on (by turning off power) or is it used to apply partial brakes by either varying the DC level to the brake or by PWM switching it? If it is on/off only then I can't see why it should be noisy for electrical reasons (unfortunately). In this case what may be happening is that the brake shoes aren't clearing the drum or disk or whatever and 'chattering" for mechanical reasons. In this case, attention to the mechanical design of the mechnism or maintenance may be what is required. If it is switched by eg PWM and electrically noisy then is it noisy when the brake is OFF but the power is ON. Is it "noisy" when the brake is off (power on) but with the stage not moving OR does the noise occur only when the stage is moving? What is the nature of the noise? Can you provide a fuller description of how the brake works and what signals are supplied when it is noisy? Russell McMahon -- http://www.piclist.com hint: To leave the PICList @spam@piclist-unsubscribe-requestKILLspammitvma.mit.edu

Rich wrote: > One of the common mistakes engineers make is using a diode that is > slower than the transient. Schottky diodes are a good choice for most > applications. That's a blanket statement that is misleading at best. Schottky diodes are indeed fast and have about half the forward drop of a silicon diode, but they are limited to relatively low voltages and can have large reverse leakage currents, especially at high temperatures. Even the common 1N5818 can leak around 20mA if I remember right. Schottky diodes are a good fit for some applications, like low voltage switching power supplies, but hardly for "most". ***************************************************************** Embed Inc, embedded system specialists in Littleton Massachusetts (978) 742-9014, http://www.embedinc.com -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

The motion stage moves a radiation detector that has some very sensitive electronics. The stage is 2 Parker 404XR's one in the X direction (side to side) and one in the Z (Up and Down). The original design of the stage only locked the Z direction when power was shut off to the motor drives, which also cuts power to the motor controller (Parker 6K). The detector electronics are picking up noise from the motor so they use 6K to cut off the motor drives and manually disconnect power from the brake, as the brake was designed to hold the stage in place when there was no power to the drives. So in the process of disconnecting the coil they get the huge emission of noise from the coil collapsing that resets there detector electronics. So the brake is only on and off, not PWM. It uses a coil to hold back a spring that clamps shoes to a shaft, at least that's how I remember it. When the coil is opened you can see a huge spike, now the other noise created by the stage, which disconnecting the power to the drives and using the brakes is to fix, is still being chased :) I hope you follow the radiation detector and electronics I don't want to talk about too much. All those patent issues, trade secrets and such. Chuck {Original Message removed}

> The motion stage moves a radiation detector that has some very sensitive electronics. The stage is 2 Parker 404XR's one in the X direction (side to side) and one in the Z (Up and Down). The original design of the stage only locked the Z direction when power was shut off to the motor drives, which also cuts power to the motor controller (Parker 6K). The detector electronics are picking up noise from the motor so they use 6K to cut off the motor drives and manually disconnect power from the brake, as the brake was designed to hold the stage in place when there was no power to the drives. So in the process of disconnecting the coil they get the huge emission of noise from the coil collapsing that resets there detector electronics. So the brake is only on and off, not PWM. It uses a coil to hold back a spring that clamps shoes to a shaft, at least that's how I remember it. When the coil is opened you can see a huge spike, now the other noise created by the stage, which disconnecting the power to the drives and using the brakes is to fix, is still being chased :) /> OK - much clearer. The diode is an excellent step in the right direction. Given the nature of the application you can afford to splurge a little on the diode. An ultra high speed diode may be very slightly better in such a one off application, although a standard 1N400x for a few cents may well do just as well. "Slugging" of the release response (ie extension of release time) due to the circultaing currents in the diode are almost certainly not going to be a problem here. If they are you could add a series resistor to the diode but just try a diode as is first. An alternative, probably not needed here, is to tailor the driver turn off so that it reduces the coil current slowly and eliminates any spike. This is liable to be substantially more complex to implement (but should still be reasonably cheap and simple). In your case I'd firstly just try a 1N400x and see what happens. It's highly likely to work. Russell McMahon -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Those are pretty impressive stages and the motors to drive them aren't small either. The cause of your original noise problem will stem from the motors being driven with a PWM drive. With motors of that size, you would need to dissipate a huge amount of heat to get reasonable performance with a non-PWM drive method. That would almost certainly solve the original noise issue, but simply isn't practical. You essentially have a low impedance driver, switching large currents at one end of a long piece of cable. ie an antenna. Because the operation of the motor controller needs some ac signal component, you are limited in the amount of filtering you can apply to the noise source, so your next best option is to add more and more shielding to the cable and keep it out of reach of the sensitive parts. Back to your original question on inductors and diodes. Put simply, an inductor will do what it takes to maintain the current flow. In other words, when you remove the external driving force, the voltage across the inductor will rise to the point where that current will continue to flow. From there the current will decay according to the equation with "e" in it. So, if you put a diode across the inductor, current will start to flow as soon as the terminal voltage reaches 0.6V. According to the rule above, the current will be the same as was being driven through it before it was turned off and decay from there. If you used a zener arrangement, then the current would start to flow at the zener voltage. The initial current would still be the same value, but will decay much quicker. You also have to consider the generator effect of any mechanical movement like a solenoid plunger or motor intertia. You can work on the assumption of a spinning motor with the power removed is now a generator. Using that as a guide, assume that the voltage generated will be on the top of the supply voltage and the current could be up to the amount that you applied. So for a pure inductor, voltage rating is determined by the supply and peak current rating is whatever the steady state current is. For anything with a moving part, both peak current and peak voltage could be twice that (as a guide). There is also a turn on delay with any diode and since that is when the inductive voltage is peaking, it is something worth minimizing. You can do that with a faster diode or an RC snubber. On 24 Aug 2004 at 8:41, Roberts II, Charles K. wrote: {Quote hidden}> The motion stage moves a radiation detector that has some very > sensitive electronics. The stage is 2 Parker 404XR's one in the X > direction (side to side) and one in the Z (Up and Down). The original > design of the stage only locked the Z direction when power was shut > off to the motor drives, which also cuts power to the motor controller > (Parker 6K). The detector electronics are picking up noise from the > motor so they use 6K to cut off the motor drives and manually > disconnect power from the brake, as the brake was designed to hold the > stage in place when there was no power to the drives. So in the > process of disconnecting the coil they get the huge emission of noise > from the coil collapsing that resets there detector electronics. > > So the brake is only on and off, not PWM. It uses a coil to hold back > a spring that clamps shoes to a shaft, at least that's how I remember > it. When the coil is opened you can see a huge spike, now the other > noise created by the stage, which disconnecting the power to the > drives and using the brakes is to fix, is still being chased :) > > I hope you follow the radiation detector and electronics I don't want > to talk about too much. All those patent issues, trade secrets and > such. > > > > > Chuck ========================================== Steve Baldwin Electronic Product Design TLA Microsystems Ltd Microcontroller Specialists PO Box 15-680, New Lynn http://www.tla.co.nz Auckland, New Zealand ph +64 9 820-2221 email: KILLspamsteveKILLspamtla.co.nz fax +64 9 820-1929 ========================================= -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

steve@TLA.CO.NZ wrote: > From there the current will decay according to > the equation with "e" in it. Not necessarily at all. The "e" equation is a specialized solution for the inductor current only when the load is resistive. > There is also a turn on delay with any diode and since that is when > the inductive voltage is peaking, it is something worth minimizing. > You can > do that with a faster diode or an RC snubber. Most diodes have nearly instantaneous turn on speed for our purposes. The difference between fast and slow diodes is primarily an issue of turn off speed. ***************************************************************** Embed Inc, embedded system specialists in Littleton Massachusetts (978) 742-9014, http://www.embedinc.com -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics