Puck SCR help / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Puck SCR help

Move Thread

LAN_403

dexter

Wed Jan 28 2015, 06:09PM

Registered Member #42796 Joined: Mon Jan 13 2014, 06:34PM
Location:
Posts: 195

I recently bought this DYNEX DCR803SG1818-1258 scr from ebay to upgrade my cap bank, but since i never used puck scr's before i have some questions

1. I couldn't find a datasheet for it and the only info i manage to find is from a surplus on-line store which only say 1800V 1045A
Has anyone got any extra info on this?

2. I now these SCR's require about 4.5-5kN of force to work properly so i made a 2 plates 4 screws clamp but i have no way of measuring the clamping force.
How flexible these SCR's are n term of possible exceeding the clamping force?
Is there any other indicator for the SCR when reaching the required clamping force except for the usage of a torque measuring key that i don't have?

Kolas

Thu Jan 29 2015, 07:21AM

Registered Member #102 Joined: Thu Feb 09 2006, 08:15PM
Location: Raleigh, NC
Posts: 169

Puck SCR's are probably the most difficult packages to work with.
Along with The clamping force required (this is mostly for very low heat transfer resistance) the SCR's must be centered almost exactly. Do you see the little divots in the middle of the SCR? These are in fact what you use to line them up. They use small ball bearings: ones that fit ever so perfectly help line the conductive heat sink up.
If you fail to center them, they will crack and be left useless. These packages are very susceptible to breakage due to misalignment and abuse while in the clamping system.
With that said this package is otherwise extremely robust as far as handling current and dissipating heat.

Clamps for these are cheap. They are quite simple with insulated bolts that you torque in a very specific pattern to achieve the right force. Most puck datasheets will show that the harder you clamp them, the more current they can handle. Since you're most likely going to do discharges with them, they wont require such a huge clamping force.

The very most important thing about these clamps is a spring washer. Once the right force is archived this preset spring washer will be forced closed and cause other washers in the system to rotate freely. Generally I would tend to clamp them with just enough force. Exceeding the given force can cause mechanical failure inside the device.
I recommend acquiring a clamp from a company that makes them for a living.
When using them for rectification, heat transfer becomes very important. Infact the water cooled "plates" are very expensive.
Good luck with them, they are total bitches.

Edit:

This is really all you need. Sadly its listing information about multiple devices, but your critical info is there.

...

Thu Jan 29 2015, 03:33PM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

I do not have much experience with pucks, but the few times I have used them went well. I did not put much effort into getting the exact force or getting it perfectly level, simply used 4 1/2" diameter bolts and 2 plates of 1/4" aluminum (with a 1/8" thick layer of pvc plastic on one side to insulate the bottom of the scr, since for pulse duty the thermal performance does not really matter) and cranked the bolts as hard as I could given the odd shape of the contraption (was about 50 ft-lbs). It worked fine and never gave me any trouble. I have heard of people who had SCRs die at unexpectedly low currents with very low or no clamping force, but as long as you have it under some reasonable force it should be fine. I personally was most worried about the fact that the clamping force was coming from the edges of the plate, not from the center as it should have been, thus resulting in uneven clamping force due to the plate deforming slightly.

Alternatively, if you paid more than $100 for your scr, you might want to invest in a real clamp, they are not that expensive see

dexter

Thu Jan 29 2015, 10:31PM

Registered Member #42796 Joined: Mon Jan 13 2014, 06:34PM
Location:
Posts: 195

thanks for the infos guys
as i said in my first post the SCR is used as a switch in a cap bank so thermal management is a non issue
based on the datasheet provided by Kolas i estimated how tight to screw the screws (the type i have requires 9-12kN so not that much force)
i used mu homemade clamp because the ones from ... links are more expensive then my SCR :)

i used 6mm bolts and tighten the nuts in a cross pattern in small increments as much as the spring washers allowed
i did some test-fires from 100V to 400V on a old coilgun coil and barrel assembly only using 2 450V 2200uF inverter grade german caps and all worked just fine

next thing is to put more caps and make some current and discharge time measurements

johnf

Fri Jan 30 2015, 06:59AM

Registered Member #230 Joined: Tue Feb 21 2006, 08:01PM
Location: Gracefield lower Hutt
Posts: 284

Just use springs of the correct compression force to set the torque

Signification

Sat Jan 31 2015, 05:29AM

Registered Member #54278 Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 367

I am trying to do virtually the same thing as DEXTER. I have a large puck type SCR that I plan to use to drive a very high current pulse from a 1-2kJ capacitor bank. This is for an 8AWG coilgun. I plan to use a type of washer spring that resemble truncated cones. These "belleville washers". See:

They were obtained from McMaster Carr. Like DEXTER's photo, I will use four identical sets of washers--one set in each corner. My design assumed that the total compression on the SCR would be the sum of the four spring sets. For example: If I wanted 4000 lb, I would use 1000 lb of compression in each corner. It has been argued with me that this is incorrect and that the total compression force on the puck (with four 1000 lb springs--one in each corner) is just 1000 lb, I just don't see it this way. Am I missing something?

Electroholic

Sun Feb 01 2015, 08:04AM

Registered Member #191 Joined: Fri Feb 17 2006, 02:01AM
Location: Esbjerg Denmark
Posts: 720

You will need a hard surface for the Belleville washers to work against, if you just have it on an ALU plate, it will just dig in and not work properly. Regarding spring force, you are correct that the total force add up. Same thing happens if you stack them up in the same direction. Displacement stays the same and force double. I.e. <<
Put them back to back and you double the displacement, but the force stays the same. I.e <>

Signification

Sun Feb 01 2015, 07:39PM

Registered Member #54278 Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 367

yeah...I think I overdid things as far as the hardware goes. I wanted to be sure the SCR didn't stress so I used very thick steel and bolts. I was concerned as the SCR is so large 4-1/4" diameter x 1-1/4" thick @ 3.6 pounds. I am concerned a bit about the inner insulating tube on the inner portion of the plates holes (to insulate the bolts from the plates). They stick out a bit through the bottom of the lower plates where the washers ate tightened down. I may shave it off a bit.

This is for a large coilgun. Electroholic, I seem to remember you talking a bit about SCR "action integrals" Caan you point me to a good source of tech info on this integral?

I am new to this forum, and will now attempt to upload a couple of photos of the SCR assembly:

OK, when I dug out this SCR assy. for the photo last night, it had been a long time since I looked at it--I remember spending a great amount of time on this but seemed to have forgotten a couple things. I just looked at the surfaces that the belleville washers make contact with--both sides (belleville) are against a fairly thick flat metal washer. One metal washer then presses against the nut and the the other metal washer presses against a plastic (for insulation) flat grey washer which in turn is against the flat plate--so there are all flat surfaces around the bellevilles, HOWEVER, does anyone see a problem with the metal-against-plastic washer area? I think it is OK since it is an all-flat, large surface area contact. Also the metal washer is not Al (magnets like it), but does anyone see any reason for concern about the belleville 'digging' into this arrangement? I did not use lock washers as I was figured this would interfere with the belleville spring constants.

I think the next PUCK will use thinner plates--this thing weighs 20lb with just single SCR and hardware!

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob