Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1] 2 Next   Go Down
  Print  
Author Topic: Soft start relay delay capacitor questions  (Read 5985 times)
2E0ILY
Member

Posts: 131




Ignore
« on: September 09, 2012, 02:17:10 PM »

I am trying to create a soft start circuit for a HV power supply. I have a 12 volt 30 amp relay that will short out the dropper resistor to the main transformer and send full power to it. I am testing the relay energizing circuit away from the psu. Supplying the relay energizing coil, which is rated 12 volt and measures at 16.5 ohm resistance across the coil, so circa 0.7 amp current draw, off my bench power supply at 12 volts, the relay energizes almost instantly. Adding a 2,200uF 25 volt working electrolytic across the energizing coil pins seemed to make no noticeable difference at all. I then tried paralleling another 2,200uF electrolytic with the first. Still no noticeable difference. What sort of delay SHOULD these impose? My one concern is they are possibly not true relays, but rather contactors, but surely the caps should still impose a delay? Should it be a  noticeable delay, or are we talking milliseconds here? Thanks.
Logged

Best regards, Chris Wilson.
G8HQP
Member

Posts: 124




Ignore
« Reply #1 on: September 09, 2012, 02:41:38 PM »

16.5 ohms and 2200uF give a CR time constant of 36.3ms. Unlikely to be noticeable.
Logged
2E0ILY
Member

Posts: 131




Ignore
« Reply #2 on: September 09, 2012, 03:24:11 PM »

16.5 ohms and 2200uF give a CR time constant of 36.3ms. Unlikely to be noticeable.

Thanks Dave, I attach a link to the circuit I am trying to duplicate, ideally without buying a specific relay, is this feasible in any way with what I have, or is the choice of the 24 volt relay deliberate in order to allow a more substantial delay? Thanks for your reply. http://www.gatesgarth.com/PSU.pdf
Logged

Best regards, Chris Wilson.
W8JI
Member

Posts: 9296


WWW

Ignore
« Reply #3 on: September 09, 2012, 06:37:50 PM »

Looking at the schematic, I'm totally lost.

The schematic shows a 24 volt relay coil running through a 600 ohm resistor from a half wave rectifier from the mains. You just said the coil resistance was 16.5 ohms and the relay was 12 volts . None of this makes any sense at all.

Also, you have a potentially lethal mistake in the filter system. The schematic has no connection to the center of the two oil filled caps. This potentially could kill someone servicing the unit, or could cause a capacitor to fail. The schematic effectively has no bleeder across the capacitors, because the center point floats. What will happen is the capacitors will equalize and discharge based on internal capacitor leakage and capacitance. Since they are oil filled, there is no guarantee the center point will not stay charged long after the amplifier is unplugged. Making it worse, the HV meter might read zero while the center point is charged. 

Finally, you can't run a three phase transformer efficienctly, or perhaps at all, the way you are trying to run it. The schematic shows a three-phase bridge. It looks like you decided to run two of the primary phases in parallel, and two of the secondary phases in series.

A three phase transformer with FW bridge would have an average current about .6 or .7 times steady dc on each winding into a capaitor input filter.  Peak current is about 1 times dc load current in each winding.

You now connected two windings in series to make a FW single phase bridge. Peak current might be 10 or 20 times average load current. This is because you changed the % of voltage ripple out of the rectifiers from 4 % up to around 50%. The 50% ripple drives apparent power factor and peak currents through the roof, and this will increase voltage drop and greatly increase transfomer heat. You also lowered ripple frequency from 360 Hz to 120 Hz, but it looks like enough capacitance.

You also increased the amount of no load HV about 7%, and you can bet the regulation will be much worse. 

Even if the core and winding arrangement in the transformer is a style that allows single phase (I suppose sometimes you could do that if you pick the correct two windings), you made all sorts of things fall far out of reasonable spec.

Are you sure you are comfortable with this conversion? From what I've seen, I'm not. I'd have to think a little bit about how the core in the transformer would behave. I think you could get into trouble with the wrong polarity of connections if you use adjacent windings.

73 Tom
Logged
G3RZP
Member

Posts: 4713




Ignore
« Reply #4 on: September 10, 2012, 12:30:57 AM »

Leaving aside the potential lethality of the filter caps, have you considered staying with the three phase rectification and a Steinmetz connection of the primary? That produces pseudo 3 phase, although you'll not get the full power that would be available with proper three phase. Nevertheless, there's a good chance you'll get more than just using one primary: as Tom says, using two primaries  could be a problem, depending on the way the transformer is made.
Logged
2E0ILY
Member

Posts: 131




Ignore
« Reply #5 on: September 10, 2012, 02:46:11 AM »

Looking at the schematic, I'm totally lost.

The schematic shows a 24 volt relay coil running through a 600 ohm resistor from a half wave rectifier from the mains. You just said the coil resistance was 16.5 ohms and the relay was 12 volts . None of this makes any sense at all.


Sorry, I wasn't terribly clear, the 16.5 Ohm 12 volt relay is what I have around, and would like to use in place of the 24 volt one shown in the schematic. The schematic isn't mine, it was kindly offered by another person when I enquired if it were possible to convert the 3 phase supply of my Henry 2000D to single phase. The supplier has converted several similar 3 phase RF generators to single phase using the schematic. I am not 100% sure about doing this, and I greatly appreciate your input here Tom. As I am in the UK and HV oil filled caps are as rare as hen's teeth I am building an electrolytic array using 15 off 450 volt, 470uF electrolytics in series with 50k ohm 5 watt bleeder resistors across each cap. I was aware of the lack of a centre tap on the two oil filled units being a potential issue, it could indeed leave a cap open to staying charged.

Basically I was going to break this down into stages and I was looking at the mains input side to the HV supply first. I was trying to utilize the relay I had for the soft start, but work out how to change the schematic to do this. I don't quite follow why one side of the relay coil in the schematic is grounded. I was expecting to see it go to the neutral side of the mains input, after the primary circuit breaker.



Quote
Also, you have a potentially lethal mistake in the filter system. The schematic has no connection to the center of the two oil filled caps. This potentially could kill someone servicing the unit, or could cause a capacitor to fail. The schematic effectively has no bleeder across the capacitors, because the center point floats. What will happen is the capacitors will equalize and discharge based on internal capacitor leakage and capacitance. Since they are oil filled, there is no guarantee the center point will not stay charged long after the amplifier is unplugged. Making it worse, the HV meter might read zero while the center point is charged. 


Yes, someone else noticed that, too, although I was myself aware it didn't seem to look right to have two  caps wired like that, given the bleeder resistor set up. Thanks for bringing it to my attention Tom!

Quote
Finally, you can't run a three phase transformer efficienctly, or perhaps at all, the way you are trying to run it. The schematic shows a three-phase bridge. It looks like you decided to run two of the primary phases in parallel, and two of the secondary phases in series.

A three phase transformer with FW bridge would have an average current about .6 or .7 times steady dc on each winding into a capaitor input filter.  Peak current is about 1 times dc load current in each winding.

You now connected two windings in series to make a FW single phase bridge. Peak current might be 10 or 20 times average load current. This is because you changed the % of voltage ripple out of the rectifiers from 4 % up to around 50%. The 50% ripple drives apparent power factor and peak currents through the roof, and this will increase voltage drop and greatly increase transfomer heat. You also lowered ripple frequency from 360 Hz to 120 Hz, but it looks like enough capacitance.

You also increased the amount of no load HV about 7%, and you can bet the regulation will be much worse. 

Even if the core and winding arrangement in the transformer is a style that allows single phase (I suppose sometimes you could do that if you pick the correct two windings), you made all sorts of things fall far out of reasonable spec.

I have powered the transformer up off our UK 240 volt 50Hz mains (live, neutral and earth) using the two outer primaries in phase, with the secondaries left open. It drew minimal current, and never got warm left like that all day. It drew 135 / 140 mA across the two paralleled primaries. I was surprised to see all three rectifier "blocks" (the original Henry rectifiers are 3 separate encapsulated blocks with each having a centre tap). I am wanting to re-use two of these encapsulated rectifiers in the power supply, if possible. As I said earlier, I am not 100% happy this will work as desired, and if necessary I will have to get a custom single phase transformer wound here in the UK as shipping, VAT and duty make getting a device from the USA impractical. But having already got this 3 phase supply in the RF generator, and having being told the schematic had run many such three phase units on single phase with excellent results, I thought I should attempt it. I am still unclear as to how to create a step start using the relay I have, with a 12 volt, 16.5 ohm coil. Nor am I clear as to why one side of the relay energizing coil in the schematic is grounded and not going to the neutral side of the mains input. Is this a US mains V UK mains thing? I am very glad you have taken the trouble to reply Tom, I am spending a lot of time looking at your excellent and well regarded web pages, most of which I have printed out for reading off line. Thank you.

Quote
Are you sure you are comfortable with this conversion? From what I've seen, I'm not. I'd have to think a little bit about how the core in the transformer would behave. I think you could get into trouble with the wrong polarity of connections if you use adjacent windings.


No, I am not 100% happy, but I am 100% sure that I know how dangerous a 5kV 1.8 Amp power supply can potentially be, and believe me I will be treating this with utmost respect and asking for more  critiquing as I slowly go along. Once the mains input  circuitry, step start and rectifier / capacitor / bleeder / and glitch resistor stuff is verified as OK I am hoping the worst that can happen is the output from the transformer is low, or it overheats? If it failed to produce the goods I would then have to commission the winding of a bespoke unit, but hopefully the rest of the build could stay as was?

Quote
   73 Tom 


Thanks Tom, and to all who have replied, it's obvious I am new to this, but I am hoping by asking questions and proceeding carefully, I can safely either get this working, or prove it incapable of doing as required. All the best from England!
Logged

Best regards, Chris Wilson.
W8JI
Member

Posts: 9296


WWW

Ignore
« Reply #6 on: September 10, 2012, 04:45:18 AM »

Sorry, I wasn't terribly clear, the 16.5 Ohm 12 volt relay is what I have around, and would like to use in place of the 24 volt one shown in the schematic. The schematic isn't mine, it was kindly offered by another person when I enquired if it were possible to convert the 3 phase supply of my Henry 2000D to single phase. The supplier has converted several similar 3 phase RF generators to single phase using the schematic. I am not 100% sure about doing this, and I greatly appreciate your input here Tom. As I am in the UK and HV oil filled caps are as rare as hen's teeth I am building an electrolytic array using 15 off 450 volt, 470uF electrolytics in series with 50k ohm 5 watt bleeder resistors across each cap. I was aware of the lack of a centre tap on the two oil filled units being a potential issue, it could indeed leave a cap open to staying charged.

There is a difference between having something "work" (meaning it did not melt down yet) and having it actually work to reasonable and relatively safe performance. For example, he has the relay coil returing to chassis while being powered off the mains.  That is illegal and unsafe.

Quote
Basically I was going to break this down into stages and I was looking at the mains input side to the HV supply first. I was trying to utilize the relay I had for the soft start, but work out how to change the schematic to do this. I don't quite follow why one side of the relay coil in the schematic is grounded. I was expecting to see it go to the neutral side of the mains input, after the primary circuit breaker.

My guess would be because whoever drew that has no idea what is safe and what isn't. That would be the same reason the center of the oil filled caps float.


W8JI wrote:
Quote
Finally, you can't run a three phase transformer efficienctly, or perhaps at all, the way you are trying to run it. The schematic shows a three-phase bridge. It looks like you decided to run two of the primary phases in parallel, and two of the secondary phases in series.

A three phase transformer with FW bridge would have an average current about .6 or .7 times steady dc on each winding into a capaitor input filter.  Peak current is about 1 times dc load current in each winding.

You now connected two windings in series to make a FW single phase bridge. Peak current might be 10 or 20 times average load current. This is because you changed the % of voltage ripple out of the rectifiers from 4 % up to around 50%. The 50% ripple drives apparent power factor and peak currents through the roof, and this will increase voltage drop and greatly increase transfomer heat. You also lowered ripple frequency from 360 Hz to 120 Hz, but it looks like enough capacitance.

You also increased the amount of no load HV about 7%, and you can bet the regulation will be much worse. 

Even if the core and winding arrangement in the transformer is a style that allows single phase (I suppose sometimes you could do that if you pick the correct two windings), you made all sorts of things fall far out of reasonable spec.

Quote
I have powered the transformer up off our UK 240 volt 50Hz mains (live, neutral and earth) using the two outer primaries in phase, with the secondaries left open. It drew minimal current, and never got warm left like that all day. It drew 135 / 140 mA across the two paralleled primaries.

That's good. Many three phase transformers share flux with adjacent windings, and depend on correct phase. Otherwise some bucking of flux can go on.

The remaining concern is regulation. A three phase FW supply has almost no ripple out of the rectifiers, and this means apparent power factor is very low. VA power is very close to actual power. This greatly reduces heat in transformers, and allows much smaller wire to be used for the same dc load current.

5 times the peak current or more can cause huge heating difference for the same dc load current, plus you threw away the copper in one winding set. Making it worse, you have it on 50 Hz.

I'm not saying it won't work, but if the transformer designer used normal headroom for three phase, it should smoke and have terrible regulation on single phase at full load.


Quote
I was surprised to see all three rectifier "blocks" (the original Henry rectifiers are 3 separate encapsulated blocks with each having a centre tap). I am wanting to re-use two of these encapsulated rectifiers in the power supply, if possible. As I said earlier, I am not 100% happy this will work as desired, and if necessary I will have to get a custom single phase transformer wound here in the UK as shipping, VAT and duty make getting a device from the USA impractical. But having already got this 3 phase supply in the RF generator, and having being told the schematic had run many such three phase units on single phase with excellent results, I thought I should attempt it.


People say all sorts of things. It might work, but I surely question the words "excellent results". Unless the transformer designer had huge headroom, I'd expect poor results.  I've used a three phase transformer on single phase, but it turned a very good supply into a marginal supply. I had to buy a 50 kW three phase generator to test three phase and 50 Hz things here.

Quote
I am still unclear as to how to create a step start using the relay I have, with a 12 volt, 16.5 ohm coil.


You are going to have a problem with a 16.5 ohm coil in a direct R/C time delay. 

You have almost 1 amp of relay coil current. To use reasonable size parts, you will have to use a second relay or a semiconductor system. You could use a 3055 series transistor as a capacitance multiplier, but by the time you do that you might as well build or buy a real timer.

Do you need a step-start? Most three phase transformers operated on single phase would have so much ESR they would not be outside relay or switch closure limits. I you want one just to have one, then I'd go solid state time delay or a double relay system.

Quote
  Nor am I clear as to why one side of the relay energizing coil in the schematic is grounded and not going to the neutral side of the mains input. Is this a US mains V UK mains thing?


No difference for USA. It is illegal and unwise to return mains power to the chassis. 

Quote
Are you sure you are comfortable with this conversion? From what I've seen, I'm not. I'd have to think a little bit about how the core in the transformer would behave. I think you could get into trouble with the wrong polarity of connections if you use adjacent windings.


Quote
No, I am not 100% happy, but I am 100% sure that I know how dangerous a 5kV 1.8 Amp power supply can potentially be, and believe me I will be treating this with utmost respect and asking for more  critiquing as I slowly go along. Once the mains input  circuitry, step start and rectifier / capacitor / bleeder / and glitch resistor stuff is verified as OK I am hoping the worst that can happen is the output from the transformer is low, or it overheats? If it failed to produce the goods I would then have to commission the winding of a bespoke unit, but hopefully the rest of the build could stay as was?
Quote
   

The real concern is this......

A three phase supply has three times the ripple frequency and just a few percent ripple. Average current and peak current is nearly the same.

The ratio of peak to average current can get pretty high with capacitor input on single phase, but is always very low in a three phase supply. This is why a choke input filter makes little difference in a three phase supply, and makes a huge difference in a single phase supply. 

Voltage sag under load primarily depends on PEAK current and transformer ESR (equivalent secondary resistance).  If a transformer drops 5% from idle to full load in a three phase supply, you could expect the same ESR to drop roughly 25% or more under single phase.

Say you have 1 amp peak winding current with 40 ohms ESR. You drop 40 volts off peaks. Now you change to single phase and peak current is 6 amperes or more. Now the HV drops 240 volts.

Heating is I^2 R. Most transformer heat is made on peaks, where voltage drop and current is highest. This is why changing from vacuum tube rectifiers to solid state rectifiers can increase heat in old radio transformers, the elimination of the vacuum tube greatly increases peak current on each cycle.

You've increased peak current and thrown away the copper in one primary and secondary. I  am sure that works great for some people, but it has to radically increase the heat and decrease voltage regulation. If the transformer initially had huge headroom and was initially very low ESR, then there is no problem.

A typical 25 kW three phase CCS transformer I use might be OK at 5 kW ICAS in single phase, if the much poorer regulation is acceptable. I never have been able to test things properly on single phase, so I'm scratching my head how a system can be excellent. When I see the errors in the oil cap centers, and the line current running to chassis, it doesn't make me feel better.

If the transformer has huge headroom, then it wouldn't matter. That sure isn't a cost effective way to build a supply, because one of the main advantages of three phase, besides better filtering and regulation, is to reduce size and cost of components to help offset the size and cost increase from using three phases. I have a time believing someone overbuilt the transformer so much it works very well on single phase.

The components look OK except for the transformer (and other noted errors), so if you have a problem  all you need to do is change the transfomer.

73 Tom
Logged
2E0ILY
Member

Posts: 131




Ignore
« Reply #7 on: September 10, 2012, 05:01:06 AM »

Thank you so much Tom for all of that. I have ordered an adjustable time delay relay (semiconductor) that switches 8 amps. I'll use that for switching the 30 amp contactor. At least it's flexible and useful for other things too. I guess I will build the capacitor board and the rectifier set up (not sure why the third rectifier block is shown disconnected from the third winding, yet still in circuit with the other two rectifier blocks...), and with great trepidation power it up. I have no means of safely measuring HV AC so need to at least rectify the HV. I do have a proper commercial up to 25kV DC probe and dedicated readout meter. Not sure about testing the thing under load short of connecting it to the RF deck and using the tube as the load, as designed. Would rather be able to test it as a standalone unit. Glad I posted, my concern re the grounded relay energizing coil was not picked up by a couple of people I showed the schematic to, but being a "newbie" I was jittery of showing ignorance in pointing out my worry Wink Thanks again Tom, much appreciated!! I foresee me having to get a single phase transformer built, but we'll give this a shot, if only to learn more about what DOESN'T work....
Logged

Best regards, Chris Wilson.
W8JI
Member

Posts: 9296


WWW

Ignore
« Reply #8 on: September 10, 2012, 07:12:18 AM »

I'm sorry about the way quote appear. eHam is a little cumbersome at times to inject alternating comments.

As I recall from when I looked at the schematic, the unused rectifiers can remain connected. They are biased off and serve no function.
Logged
2E0ILY
Member

Posts: 131




Ignore
« Reply #9 on: September 10, 2012, 07:14:56 AM »

Oh, no worries at all Tom, I manually did the quotes, a bit of a PITA, I know, thanks again, and all the best. 73 2E0ILY
Logged

Best regards, Chris Wilson.
G3RZP
Member

Posts: 4713




Ignore
« Reply #10 on: September 10, 2012, 08:41:32 AM »

Tom,

What about maintaining three phase rectification with a Steinmetz primary connection?
Logged
2E0ILY
Member

Posts: 131




Ignore
« Reply #11 on: September 10, 2012, 08:49:30 AM »

Tom,

What about maintaining three phase rectification with a Steinmetz primary connection?

I am glad you directed that at Tom, and not myself <LOL> Off to Google Steinmetz...... Smiley
Logged

Best regards, Chris Wilson.
W8JI
Member

Posts: 9296


WWW

Ignore
« Reply #12 on: September 14, 2012, 07:20:06 PM »

Tom,

What about maintaining three phase rectification with a Steinmetz primary connection?

I had to look that up myself. :-)

That runs motors OK in some cases. My lathe uses a capacitor for one phase when I run it on 240 single phase, but the motor lost a great deal of torque, it spins up slower than it does when I have a generator running. Since I never turn anything large it is OK.

I don't think it would work well with a power supply, unless he can connect a very large three phase motor across the primaries.

Logged
2E0ILY
Member

Posts: 131




Ignore
« Reply #13 on: September 15, 2012, 02:11:17 AM »

I am getting close to being able to power this thing up, and see if the single phase conversion of the transformer delivers the goods. I'll post the result up here, for interests sake. At least the capacitor bank and control circuitry will swap to a proper single phase HV transformer if required. Thanks for all your help.
Logged

Best regards, Chris Wilson.
W8JI
Member

Posts: 9296


WWW

Ignore
« Reply #14 on: September 16, 2012, 09:14:29 AM »

I am getting close to being able to power this thing up, and see if the single phase conversion of the transformer delivers the goods. I'll post the result up here, for interests sake. At least the capacitor bank and control circuitry will swap to a proper single phase HV transformer if required. Thanks for all your help.

How well that will work really depends on how much headroom they had in the 3-phase transformer design.

If the design was exceptionally conservative (which oddly enough would be poor manufacturing engineering) it might work OK. If the design was typical of a 3-phase transformer reasonably within limits, it will have excessive voltage sag and will also have heating issues in high duty cycle high power use.

What you are really evaluating is the headroom they had in the transformer, not if a conversion like that is universally good.

:-)
Logged
Pages: [1] 2 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!