|
Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders. |
|
Thread Tools |
23rd Oct 2017, 8:26 pm | #41 |
Dekatron
Join Date: Jan 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 8,172
|
Re: Variac help please
Hi Gents, usually plenty of NTC's in scrap SMPS units, several in parallel should do.
If you go down the timer route then a 10W 10R resistor (or lower R) should do nicely. Ed |
23rd Oct 2017, 9:29 pm | #42 |
Dekatron
Join Date: Jun 2009
Location: Staffordshire Moorlands, UK.
Posts: 5,263
|
Re: Variac help please
Definitely a heavy duty NTC or three. I have one on the bench isolating transformer as it's off a B20 breaker and plenty of room in the box, but the 8A variac is off a B32 and simply dims the lights a little. It's unusual to see C-type breakers in domestic situations and they shouldn't be installed willy-nilly unless you can prove the earthing arrangements.
The only problem with using a timer-bypassed inrush resistor is what happens to the resistor if the timer fails under load. It might get very hot indeed and in a domestic environment might start a fire.
__________________
Kevin |
23rd Oct 2017, 10:16 pm | #43 | |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Variac help please
Quote:
Silicon Chip built a very nice soft starter for appliances using this method, you have to subscribe to get the whole article; http://www.siliconchip.com.au/Issue/...Current+Menace |
|
23rd Oct 2017, 10:51 pm | #44 |
Dekatron
Join Date: Jun 2009
Location: Staffordshire Moorlands, UK.
Posts: 5,263
|
Re: Variac help please
An NTC doesn't run as hot as a fixed resistor that's accidentally left in circuit. A 10R resistor is always a 10R resistor. A 10R NTC may go to a fraction of an ohm, so its capability to start a fire is much less. Over my career I've seen many soft start resistors overheat and burn adjacent boards or wiring, whether in switchmode power supplies, touring-grade amplifiers, inverter drives or what-have-you. I've never seen an NTC fail in that way. Never. If they're grossly overloaded they fail safe-open. They are a safer bet all-round.
Fast power-cycling is best with an NTC and timed bypass, so the NTC soon cools down ready for the next power-up. But if the relay fails, no danger.
__________________
Kevin |
24th Oct 2017, 5:08 am | #45 | |
Dekatron
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,190
|
Re: Variac help please
Quote:
There is an anti-surge fuse in series with the resistor (the relay shorts out the resistor and fuse). The fuse is rated so thst it will stand the current through the resistor for a short time, so that normally it's OK, but if the relay fails the current flows through the fuse long enough to blow it. |
|
24th Oct 2017, 12:28 pm | #46 | |
Dekatron
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,528
|
Re: Variac help please
Quote:
Only if the diode's thermionic! And if it is indirectly heated, it'll do a delay on characteristic all on its own
__________________
....__________ ....|____||__|__\_____ .=.| _---\__|__|_---_|. .........O..Chris....O |
|
24th Oct 2017, 9:05 pm | #47 |
Dekatron
Join Date: Jan 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 8,172
|
Re: Variac help please
If you have the space use a lamp (as in limiter). Even 15W pygmy should do to get the core flux going. If the timer sticks the bulb stays on, but cannot get any hotter than normal.
Ed |
24th Oct 2017, 10:46 pm | #48 |
Dekatron
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
|
Re: Variac help please
if you do as Ed says, and use a lamp, you can see it working by if it gives a brief flash on switch-on. Probably sometimes it will, sometimes not. Which will be educational!
|
25th Oct 2017, 12:09 am | #49 | |
Dekatron
Join Date: Jun 2009
Location: Staffordshire Moorlands, UK.
Posts: 5,263
|
Re: Variac help please
Quote:
Another type of inrush circuit uses a PTC thermistor (polyswitch). The intrinsic cold resistance will stop any inrush, then a relay bypasses it. if the relay fails, the polyswitch heats and trips off. But again, the values chosen are critical. NTC Ametherm or Surgeguard
__________________
Kevin |
|
25th Oct 2017, 2:17 pm | #50 |
Dekatron
Join Date: Nov 2003
Location: Heckmondwike, West Yorkshire, UK.
Posts: 9,637
|
Re: Variac help please
Many years ago I built a similar device for my Father's slide projector. A thermistor slowed the lamp filament, and was then shorted out by a relay. I think it was a PW design. All long gone, along with my precise memory of it.
|
27th Oct 2017, 12:34 am | #51 |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Variac help please
I should have taken photos of two large NTC's that exploded/fractured in series with the compressor motor on one of my air con units. One split into fragments, the other melted into a pile of round globules in its central section. They can fail with dissipation even though they intrinsically try to protect themselves with reduced IsquaredR losses as their resistance drops with heat.
|
27th Oct 2017, 1:58 am | #52 | |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Variac help please
Quote:
The thermistors that fractured & melted in my air con unit, looking at the one where its central structure became a mass of small molten globules or spheres, to melt that material, I'm not sure, but just guessing, it wouldn't surprise me if that got to 700 dec C or more, certainly way hotter than 500 deg C. |
|
27th Oct 2017, 2:52 am | #53 | ||
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Variac help please
Quote:
A 33V DC supply is made from a half wave silicon diode rectified 24V rms transformer output and a 10,000uF filter cap. This can be used to power a 24V DC relay, say just picking one from the RS book with a 20A contact rating and a 1440R coil. Simply about a 600R resistor is placed in series with the relay coil so that the relay's operating voltage is about 24V. A 4700uF capacitor is connected across the relay coil. The 4700uF capacitor delays the rise in voltage across the coil at power up so that it has risen to the relay's pull in voltage only after about 2 seconds. (the Thevenin resistance that the capacitor is charging from is the parallel equivalent value of the 600R and the 1440R, or about 420 Ohms) that with the 4700uF cap makes the time constant about 2 seconds, and the relay gets close to its pull in voltage maybe around 16V at that time. So when the circuit is powered, it takes about 2 seconds for the relay to close. Then a set of contacts can be used for some application, like say shorting out an NTC in a surge limiter system. When higher voltage DC relays were once available, it could be done quite easily by just half wave rectified mains, and a stepdown to 24V wasn't needed as in the example above. Then with this circuit, one set of spare relay contacts that were NC, could be placed in series with the power to the circuit, then after the relay closed (say switching on a mains lamp with the other set of contacts) and then the power feed was interrupted by the NC contact opening, the capacitor would discharge into the coil for the next half of a timing cycle until the relay dropped out and it would make a mains light bulb flasher with no thermionics and no bimetallic strips, the hysteresis being provided by the different pull in and drop out voltages of the relay. I had one of these as a teenager flashing a mains lamp in my room using a vintage post office style high voltage DC relay, a resistor a silicon diode & a capacitor. It worked a treat. A similar idea was used in early automotive lamp flashers that replaced the bimetallic strip ones. It is amazing what you can do when you combine capacitors & relays to create delays, they might not be precision delays, but sometimes that is not needed. Last edited by Argus25; 27th Oct 2017 at 2:58 am. |
||
27th Oct 2017, 6:25 am | #54 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
|
Re: Variac help please
Back in the 80's I built myself a monster 13.8v supply for the radio shack. It has an aluminium-clad soft start resistor bolted to a small aluminium plate mounted on standoffs. the plate also mounts a 100C thermal cutout switch. If the resistor gets hotter than that the switch cuts the current.
This PSU has a start button. Pressing it operates one microswitch which disables the downstream regulators. A second microswitch connects the path through the soft start resistor, and this brings up the transformer and reservoirs off-load. The reservoir voltage drives a relay coil which pulls in when there is about 10v in the reservoirs The contacts of this relay bridge the soft start resistor, the microswitch and the thermal cutout, thus latching the power on. the microswitch and thermal switch don't have to carry full running mains current. Releasing the start button enables the regulators to come on. It's not very complex but I thought it covers all the bases. It's a bit inconvenient. Put the main power toggle 'on' then poke the start button. The output comes up when you release the button. It has a big crowbar circuit. If triggered it fires a thyristor across the reservoirs. This too has a surge limiting resistor. The soft start relay coil power comes from the thyristor end of this resistor, so, if triggered, the crowbar quickly disconnects the mains so the thyristor has only to dump the reservoir contents. If something had gone wrong and this supply was in the hands of someone not familiar with it, they might hold the start button in indefinitely. The soft start resistor will limit mains current, the crowbar will protect the radios and the thermal switch will be left cycling holding the soft start resistor at 100C. For controlling the inrush current of a variac, a relay with a timer would be good. but put a thermal switch on the inrush resistor and a start button in series with the resistor. David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
28th Oct 2017, 9:19 am | #55 | |||
Dekatron
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,528
|
Re: Variac help please
Quote:
I wasn't claiming that the diode had to be thermionic but that if it wasn't, you'd still have a semiconductor in the circuit. (The diode!) The bit about indirectly heated thermionic diodes having an inherent delay on characteristic is a nod to the forum's "vintage" tag. (ISTR this very issue of problems with no face to face communication was brought up in my thread about smiley problems and operating system compatibility.)
__________________
....__________ ....|____||__|__\_____ .=.| _---\__|__|_---_|. .........O..Chris....O |
|||
28th Oct 2017, 4:20 pm | #56 |
Nonode
Join Date: Jan 2010
Location: North Somerset, UK.
Posts: 2,129
|
Re: Variac help please
I made a useful soft start unit for a large transformer.
It is a relatively simple matter to insert a series resistance for starting that is bypassed by a relay after a short time. However as has already been said, that leaves the risk that a failure of the relay or timer would leave the resistance in circuit indefinitely with consequent risk of fire. I resolved this by using a standard electric kettle as the resistance, this is connected in series with the transformer for about a second and then bypassed. Any failure that leaves the kettle connected continually can do no worse than boil the contents and then trigger the auto shut off of the kettle. A 1KW or a 3KW kettle may be selected according to the size of the load. |
28th Oct 2017, 6:49 pm | #57 |
Dekatron
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
|
Re: Variac help please
That's a good tip, broadgauge, and a cheap kettle can be had for a fiver or so!
|
28th Oct 2017, 8:03 pm | #58 | |
Rest in Peace
Join Date: Jan 2015
Location: Cottesmore, East Midlands, UK.
Posts: 858
|
Re: Variac help please
Quote:
|
|
29th Oct 2017, 5:28 am | #59 |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Variac help please
Herald1360,
But you will note that the time delay using a relay and capacitor doesn't actually need any diode at all, if the power supply you run it from is DC, all it needs is a resistor, the relay and a capacitor. The diode is only needed if you are stuck with an AC supply. So maybe we missed each other's points. But certainly no thermionics a are needed top create a delay and just those few parts. |
29th Oct 2017, 9:33 am | #60 |
Dekatron
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,528
|
Re: Variac help please
Granted. But your suggestion was for an ac operated delay with a diode. A universal solution rather than the dc operated one which would have to get dc from somewhere else in order to work with just a variac for its raison d'etre. Most ways to provide this dc other than dynamos or batteries would have semiconductors in them these days.
"Eats Shoots and Leaves" anyone?
__________________
....__________ ....|____||__|__\_____ .=.| _---\__|__|_---_|. .........O..Chris....O |