I think I made a mistake. R1 should go from BUZ+ to Common, not from BUZ+ to +1.5V. If the alarm output is a PNP, and the piezo looks like a capacitor (no resistive path), then R2 needs a pull-down, not a pull-up.

Ken

 

Ken,
Sorry, I have been away and not able to try your suggestions. I will need to order some parts and will give it a try again soon. I still need a solution and have not abandoned the project!
Thanks in the meantime.

 

Copy the below into notepad (not wordpad or Word or anything that will add a bunch of crap formatting codes) and save it on your desktop as - MCR716.sub
.SUBCKT MCR716 anode gate cathode PARAMS:
**************************************
* Model Generated by EVAL LAB *
* Feb. 22, 2007 *
* Copyright(c) ON Semiconductor *
* All Rights Reserved *
*Commercial Use or Resale Restricted *
**************************************
*SCR
*MODEL FORMAT: PSpice
+ Vdrm=400v Vrrm=400v Idrm=1u
+ Ih=1.0ma dVdt=100e6
+ Igt=25ua Vgt=0.55v
+ Vtm=1.5v Itm=8.20
+ Ton=2u Toff=15u

* Where:
* Vdrm => Forward breakover voltage
* Vrrm => Reverse breakdown voltage
* Idrm => Peak blocking current
* Ih => Holding current
* dVdt => Critical value for dV/dt triggering
* Igt => Gate trigger current
* Vgt => Gate trigger voltage
* Vtm => On-state voltage
* Itm => On-state current
* Ton => Turn-on time
* Toff => Turn-off time

* Main conduction path
Scr anode anode0 control 0 Vswitch ; controlled switch
Dak1 anode0 anode2 Dakfwd OFF ; SCR is initially off
Dka cathode anode0 Dkarev OFF
VIak anode2 cathode ; current sensor

* dVdt Turn-on
Emon dvdt0 0 TABLE {v(anode,cathode)} (0 0) (2000 2000)
CdVdt dvdt0 dvdt1 100pfd ; displacement current
Rdlay dvdt1 dvdt2 1k
VdVdt dvdt2 cathode DC 0.0
EdVdt condvdt 0 TABLE {i(vdVdt)-100p*dVdt} (0 0 ) (.1m 10)
RdVdt condvdt 0 1meg

* Gate
Rseries gate gate1 {(Vgt-0.54)/Igt}
Rshunt gate1 gate2 {0.54/Igt}
Dgkf gate1 gate2 Dgk
VIgf gate2 cathode ; current sensor

* Gate Turn-on
Egate1 gate4 0 TABLE {i(Vigf)-0.95*Igt} (0 0) (1m 10)
Rgate1 gate4 0 1meg
Egon1 congate 0 TABLE {v(gate4)*v(anode,cathode)} (0 0) (10 10)
Rgon1 congate 0 1meg

* Main Turn-on
EItot Itot 0 TABLE {i(VIak)+5E-5*i(VIgf)/Igt} (0 0) (2000 2000)
RItot Itot 0 1meg
Eprod prod 0 TABLE {v(anode,cathode)*v(Itot)} (0 0) (1 1)
Rprod prod 0 1meg
Elin conmain 0 TABLE
+ {10*(v(prod) - (Vtm*Ih))/(Vtm*Ih)} (0 0) (2 10)
Rlin conmain 0 1meg

* Turn-on/Turn-off control
Eonoff contot 0 TABLE
+ {v(congate)+v(conmain)+v(condvdt)} (0 0) (10 10)

* Turn-on/Turn-off delays
Rton contot dlay1 825
Dton dlay1 control Delay
Rtoff contot dlay2 {290*Toff/Ton}
Dtoff control dlay2 Delay
Cton control 0 {Ton/454}

* Reverse breakdown
Dbreak anode break1 Dbreak
Dbreak2 cathode break1 Dseries

* Controlled switch model
.MODEL Vswitch vswitch
+ (Ron = {(Vtm-0.7)/Itm}, Roff = {Vdrm*Vdrm/(Vtm*Ih)},
+ Von = 5.0, Voff = 1.5)

* Diodes
.MODEL Dgk D (Is=1E-16 Cjo=50pf Rs=5)
.MODEL Dseries D (Is=1E-14)
.MODEL Delay D (Is=1E-12 Cjo=5pf Rs=0.01)
.MODEL Dkarev D (Is=1E-10 Cjo=5pf Rs=0.01)
.MODEL Dakfwd D (Is=4E-11 Cjo=5pf)
.MODEL Dbreak D (Ibv=1E-7 Bv={1.1*Vrrm} Cjo=5pf Rs=0.5)

* Allow the gate to float if required
Rfloat gate cathode 1e10

.ENDS

 

Note that the SCR model is just the first SCR model I found and not a recommendation or anything like that. LTSpice doesn't come with an SCR model is all. Note that any included SUBs and the schematic files must be saved in the same place normally.

In this post I have attached a file.
Save the attached file to Desktop with LTSPICE.

It is a basic two transistor latch.

R2 and D1 raises your kitchen timer so that it sits at a higher level so that the low kitchen timer pulse voltage can be used just to turn on the latch.

Just to explain the latch, the pulses go through R1 and turn on transistor Q1.

Q1 turns on Q2 which then holds Q1 on making this a latch until you cut the power.

With Q2 on the SCR will trigger with C2 limiting the current.

C1 and R6 are an optional current management network.

The Capacitor will pass a high current to turn the relay on and the resistor will reduce the holding current. I specced the relay as coil L25ohms with 25 ohms DC resistance. R6 is about twice the DC resistance of the relay.

You might want to check the specs for your relay and adjust R6 so that in series with the relay it allows a little more than the relays required holding current rating.

 

Potato,
Thanks, I will have a look at LTspice as I have never used it before and I will add the parts to the order I am busy placing at the moment and give it a try soon.
Thanks again.

 

None of the parts are critical. It should work with whatever you have available.

Only caution is getting transistors with the voltage rating.

 

Why are you adding a "latching" transistor circuit ahead of a "latching" SCR? If you're going to do that, drop the SCR and drive the relay with your transistor circuit?

Ken

 

None of the parts are critical. It should work with whatever you have available.

Only caution is getting transistors with the voltage rating.

I am ordering a breadboard and also need a couple of SCR's, my only one dropped somewhere never to be seen again!

 

Why are you adding a "latching" transistor circuit ahead of a "latching" SCR? If you're going to do that, drop the SCR and drive the relay with your transistor circuit?

Ken

The original request was for a way to turn on an SCR with a very low voltage input pulse from a kitchen timer.

I don't disagree with you KMoffet. That is an option that Suzukiman might want to check on when he is Breadboarding. Does he need the SCR at all or can he just use the transistor latch.

 

Was just a curiosity.

I think that even though Suzukiman measured the output as 0.0-0.8V on a DVM, the output is probably (from my experience) 0v to 1.5V. What I've seen on an o-scope is a series of pulses with a very low duty cycle. A DVM averages this to a low "average" voltage. I don't think he will need to offset the alarm voltage.

But...we will see.

Ken

 

It looks like it would work as long as you used the right transistors.
In this simulation I picked 220 Ohms for the Coil and set the holding current to a third of relay turn on current with the 470 Ohms resistor. That should work well enough with a 2N2907 but I wouldn't try it with a 2N3906. If your relay coil was more current hungry due to lower resistance than 220Ω, for example the 25Ω that I had it set at with the SCR, then a much tougher transistor is needed.

A TIP125 Darlington PNP should manage and those are only around $0.50 and available widely.

 

Was just a curiosity.

I think that even though Suzukiman measured the output as 0.0-0.8V on a DVM, the output is probably (from my experience) 0v to 1.5V. What I've seen on an o-scope is a series of pulses with a very low duty cycle. A DVM averages this to a low "average" voltage. I don't think he will need to offset the alarm voltage.

But...we will see.

Ken

I agree with that as well and said the same thing the first time this topic was circulating.

 

I have just placed the order and should have it sometime next week.
I think you may be right that the pulse is about 1.5V, unfortunately I do not have an scope to check it, so just assumed that the multimeter showing about 0.8V was correct and seing thre was an output transistor and assumed a 0.7V drop by the transistor.

The original idea was to trigger the SCR with the pulse so that it latches and holds while current is drawn through it, which is what SCR's do if I understood that correctly, and to reset it with switch in the power line to the SCR/relay.

I will be in a better position to experiment with different configs when I have a breadboard available.

Thanks Potato pudding and Ken, for your interest in helping me to get this working.

 

It is fun and helps me get my mind back into thinking about this stuff - because I hadn't been thinking about it much for quite a while which was ashame - because it is fun. I am horribly out of practice or else I would have been able to help you much quicker and without so much relearning.

I know that there is a probably a much simpler way to do this with one transistor but it would probably be more component dependent and this will work for now unless there is something extremely strange.

If you wanted to provide details like the relay model so that I could find a data sheet that would be help me give you more refined help. I am also wondering if it was asked or just assumed, but is the relay being used to switch AC?

 

If you wanted to provide details like the relay model so that I could find a data sheet that would be help me give you more refined help. I am also wondering if it was asked or just assumed, but is the relay being used to switch AC?

The relay is a 24V dc type about 2400 ohms, another one is 1200 ohms, but I have to recheck that for you again. It is going to switch a low current DC 30V.

 

If you wanted to provide details like the relay model so that I could find a data sheet that would be help me give you more refined help. I am also wondering if it was asked or just assumed, but is the relay being used to switch AC?

The relay is a 24V dc type about 2400 Ω, another one is 1200 Ω, but I have to recheck that for you again. It is going to switch a low current DC 30V.

 

I should have asked for more details earlier because low current DC level the only reason to use the relay instead on doing it all with transistors is for the Isolation and for multipole and multithrow.

Isolation can be done with Optos more efficiently.

For multithrow and or multipole the relay works but you said a current which means it isn't the multipole. Multithrow can be done with logic circuits but some major timing issues can be covered by using the relay.

For using the relay you will have no problems switching it directly with the transistors. That 2400 or 1200Ω is going to only need a very small current.

You can save the SCRs for another project.

 

The relay is needed to completely turn a module on and off. Once I have my breadboard it will make testing a lot easier I think. Have to get a 2N2907 as well, as I have none of those on order.
I really appreciate your contributions and hope to test as soon as the parts arrive.

 

The relay is going to need lower current than I expected.

You can try a 2N3906 or any other PNP that is rated for the voltage.

Current at turn on will peak at about 25mA and with a series resistor around 820 Ohm to reduce the holding current to 10mA, any 40V PNP should manage to survive as a switch.

Interesting about the relay needed for turning a module completely on or off. That is something that could be worth using a relay for if the load has special conditions.

Transistors as a switch for example can react with the source impedance of the battery, power supply, or wiring, when it acts on the emitter; reducing current, and even causing oscillation if they have a resonant load that is current sensitive, but that is normally easily fixed by filtering the power supply with a fair to large capacitor that is across the power supply and connected in circuit very close to the emitter.

 

I've used a comparator set up as a high input resistance resetable latch many times. I don't have the exact circuit on this PC but it's similar to the one illustrated in here as "Comparator Time Delay Schematic 1" and you can reset it by pulling the line that leads to that timing cap to ground.

http://home.cogeco.ca/~rpaisley4/Comparators.html

Any desired time delay (or none) can be achieved by changing or omitting the 1M resistor and 3.3 uF capacitor. The beauty of this circuit is it will easily perform a function that's not easily done with a single 555. I have found it to be highly predictable in operation, just remember that most comparators won't sink much output current over any length of time so a transistor follower is often called for.