supply

[arivel]

Hi everyone .
do you recommend me?
Is it possible to combine a dual power supply with a single power supply together?
in my case the outputs of the CD40193 must drive the input of the operational amplifiers. the output of each operational amplifier goes to the input of a coil and the output of the coil goes to ground.
I prefer to do without bridge H if possible.
for the power supply I thought of using two identical transformers, each with a secondary and two outputs therefore: 6 volts 15 volts. connecting the two transformers together I get five outputs -15 -6 ground +6 +15.
the CD40193 is connected to -6 + 6. it can be done ?.
bye thank you

[Zero999]

Yes the CD40193 can run of +/-6V.

Note that the output voltage from a transformer will always be the peak voltage, not what's on the label, which is the RMS, so a 6V transformer, will give closer to 8.4V when rectifier. Smaller transformers also output a slightly higher voltage, when unloaded, so it will be higher still.

What are you trying to do?

[arivel]

Thanks for the reply .
I'm trying to do a relay volume control.
for now I managed to make two cd40193 in cascade that command two 74hc154 and it works.
the storage function of the previous volume I do with 6 bistable relays which are the subject of the question I just made.

[Zero999]

Thanks for the reply .
I'm trying to do a relay volume control.
for now I managed to make two cd40193 in cascade that command two 74hc154 and it works.
the storage function of the previous volume I do with 6 bistable relays which are the subject of the question I just made.

Please post a schematic.

[arivel]

I can't insert image.
which chip with six high current operational amplifiers do you recommend?

[Zero999]

I can't insert image.
which chip with six high current operational amplifiers do you recommend?

Why can't you attach the image? Click "Attachments and other options" and "brows", select the file and click "open". If the file is too big, scale it down using image editing software. If it's a schematic, use PNG format and reduce to colour depth to 8-bit or less, to save space. If it's a photograph, use jpg, with a quality setting of 75%.

What do you mean by high current?

Op-amps normally come in single, dual, or quad packages. I've never seen a hex op-amp package before.

[arivel]

I enclose

[Ian.M]

To drive multiple relays from logic, the easiest option is the ULN200x series of Darlington arrays.  They have seven 50V 500mA Darlington transistors, complete with base resistor networks, and clamping diodes to handle the back-EMF from the relay coils all in a 16 pin package.   Connect the relay coils between a positive supply and the ULN200x outputs and connect the ULN200x COM pin to the same positive supply.  Each relay will activate when logic '1' is applied to its corresponding ULN200x input.  For 5V logic get the ULN2003. They are not well suited to 5V relays as their on state Vce_sat voltage drop is around 1V.  If you need more channels, the very similar ULN280x series has eight Darlingtons per 18 pin chip.

That schematic doesn't look very hopeful.   The CD40193 Presettable Binary Up/Down Counters wont count as their /Load input is stuck low, and even if they did you haven't got any button debouncing  so at best they'd count erratically, and at worst would 'see' enough pulses from a single button push to increment or decrement to their limiting values enforced by the button gating circuit.

Rather than using 2x 74HC154 4 to 16 line decoders, you'd be better off using 4x 74HC238 3 to 8 line decoders with active high outputs - they can directly drive ULN2803 Darlington drivers without needing inverters in between, and two of them can be connected as a 4 to 16 line decoder with no additional components required.

If possible, stick to a single logic family.  I'd get 74HC193 presettable up/down counters and do the whole design in 74HC logic.   Supplying moderate numbers of 74HC logic chips with 5V is easy, due to their loiw quiescent current - a 78L05 regulator running off the higher voltage rail supplying the relays will do the job if there are no significant 5V loads apart from the logic, or a 7805 regulator (for its higher current capability) if you are also driving loads like indicator LEDs from the logic outputs.

[arivel]

To drive multiple relays from logic, the easiest option is the ULN200x series of Darlington arrays.  They have seven 50V 500mA Darlington transistors, complete with base resistor networks, and clamping diodes to handle the back-EMF from the relay coils all in a 16 pin package.   Connect the relay coils between a positive supply and the ULN200x outputs and connect the ULN200x COM pin to the same positive supply.  Each relay will activate when logic '1' is applied to its corresponding ULN200x input.  For 5V logic get the ULN2003. They are not well suited to 5V relays as their on state Vce_sat voltage drop is around 1V.  If you need more channels, the very similar ULN280x series has eight Darlingtons per 18 pin chip.

That schematic doesn't look very hopeful.   The CD40193 Presettable Binary Up/Down Counters wont count as their /Load input is stuck low, and even if they did you haven't got any button debouncing  so at best they'd count erratically, and at worst would 'see' enough pulses from a single button push to increment or decrement to their limiting values enforced by the button gating circuit.

Rather than using 2x 74HC154 4 to 16 line decoders, you'd be better off using 4x 74HC238 3 to 8 line decoders with active high outputs - they can directly drive ULN2803 Darlington drivers without needing inverters in between, and two of them can be connected as a 4 to 16 line decoder with no additional components required.

If possible, stick to a single logic family.  I'd get 74HC193 presettable up/down counters and do the whole design in 74HC logic.   Supplying moderate numbers of 74HC logic chips with 5V is easy, due to their loiw quiescent current - a 78L05 regulator running off the higher voltage rail supplying the relays will do the job if there are no significant 5V loads apart from the logic, or a 7805 regulator (for its higher current capability) if you are also driving loads like indicator LEDs from the logic outputs.

Hello
recommending the Darlington chip array are you referring to the relays controlled by the 74HC154? because vice versa for bistables they would not be good.
actually to verify the operation I used a simulator, in reality it can be as you say. in this case just send a negative pulse to the LOAD input of the 40193 in order to update the outputs with the inputs.
before choosing cd4093 I tried with 74HC193 but it did strange things despite having connected all four inputs to the ground.
yes the micro bounces on the relay contacts that go directly to the CD4093 inputs can cause strange things but just apply a filter.
4x decoders from 3 to 8 lines 74HC238, is there a scheme to view?

[Ian.M]

If the bistable relays have two coils,a ULN2803 is suitable (subject to their coil current requirements) and can drive four relays.   If you are driving single coil bistable relays, you need to reverse the coil polarity to choose between operate and release, so need to drive them from a H-bridge.   Dual H-bridge ICs are readily available.   If you are using a  MCU to control the relays and don't need simultaneous switching, you can economise by connecting their coils in a chain with one half-H-bridge per node (i.e. for N bistable relay coils you need N+1 half-H-bridges).

You shouldn't need a schematic for 74xx138 or 74xx238 decoders paired as a 4 to 16 line decoder.  Both have two active low enables and one active high enable, so you can connect the fourth bit to EN on one and either /EN on the other.  For a fifth bit, you've no longer got a free choice of enable polarities so need one inverter.  Above five bits, its probably best to pre-decode the high bits with a 74xx138.

[arivel]

Hello
Is it possible to use an NPN chip array with a PNP chip array together?
by this I mean if it is possible to connect the inputs of one with the inputs of the other and the outputs of one with the outputs of the other so as to have six inputs and six outputs instead of twelve inputs and twelve outputs.
the question I ask to understand if I can power the single coil relay with dual power supply -6 +6.
I don't want to use microcontrols.

[Ian.M]

First find a PNP Darlington array - they are near unobtanium. 

Even so the answer would be NO, as *either*, if the high side array contains level shifters (which usually invert), both it and the low side array would be on when the input is at logic '1' *or* if its direct to the PNP Darlington base, if it ever floated it would turn both  high and low sides on, and even with robust fast rail-to-rail drive, you'd get severe shoot-through during transitions.   

I said look at H-bridge ICs,or consider the other suggestions in the thread https://www.eevblog.com/forum/projects/how-to-drive-latching-relais/

[arivel]

I managed to do it with the whole 74XX family but this also does strange things.
the first count forward does it right all the way but then when he goes back instead of starting from the last chip from the right he jumps directly to the second from the left.

[Ian.M]

Sorry, I couldn't get my head round debugging your most recent version but I did draw up a version that I've tested the digital section of in Protel CircuitMaker 6 Pro (an ancient simulator that has interactive digital simulation).  It uses three chips of 'glue' logic to implement the zero and max end stops without using the decoder outputs, and also the input debouncing and a latch for Up/Down.  One inverter from the glue logic is used for the '138 decoders to permit four to be stacked with five bit addressing.  Two are used to clean up the reset signal, which preloads the initial volume.  There's still a spare Schmitt NAND leftover.

For a real build, in addition to swapping 74HC logic for the 74LS on the schematic, and deleting the hex displays you *WILL* need to increase the 1K input pullups to get long enough debounce times for your specific buttons.

N.B. the input debounce circuit *MUST* give a long enough low time for the limit logic and inter-counter carry to propagate.  If its too short, it will blow through the limits!

For the benefit of anyone who's still got this ancient simulator (or cares to google for a copy of its free Student Edition), I also attach the .ckt files.  The one with 'test' in the name is prepared for interactive digital simulation by removing analog RC stuff and adding simulation signal sources.  N.B The up/down SR latch (U3A,U3B) is metastable in the simulator, so press one of the buttons to set a direction, and toggle the /Reset before using the (Auto) switch, which when down, repeats the button press so you don't wear out your finger!

[arivel]

Hello and thanks .
I ran aground on the control circuit for the bistable relays which must memorize the six bits of the up / down counters without being able to go on.
it is clear to me the functioning of the single coil ones, not the same for the double coil ones.
for example I would like to know if they have a common connection.
is there a site or someone who explains it?

[Ian.M]

I wouldn't bother with bistable relays just to hold a logic state - too big, expensive and difficult to drive.

Instead, I'd use a CMOS 8 bit transparent latch with a backup battery and a PSU supervisor IC + a bit of glue logic, to latch the contents when the power goes away.   e.g.if you choose 74HC573 which will hold its state down to 2.0V and a leading brand CR2032 coin cell as the backup battery, you can expect over 30 months of power-off data retention per battery.

There are a few devils in the details: you need a pair of low leakage Schottky diodes to route both battery power and +5V to the latch so the battery takes over when the 5V goes away, cathodes to the latch Vcc pin, anodes to each power source. e.g the common cathode BAS70-05 low leakage double Scottky diode, which typically leaks less than a uA at 5V reverse voltage and 85 deg C.

The74HC573 needs a pullup on its /OE pin to its own Vcc pin to turn its outputs and something to pull that pin low when the 5V rail is up.  e.g a SN74LVC1G06 single gate open drain Schmitt inverter, powered from the 5V rail with its input fed from a potential divider with 80% of the 5V rail so it doesn't pull the /OE pin low till the 5V rail is well above the battery voltage and a circuit to take LE low and latch the current data as soon as the 5V starts to dip, but only let it go back high after the reset pulse is finished, so the data on its outputs from the previous session is latched into the counters parallel data inputs when their pre-load pins go high.  Basically whatever drives LE must be quick to go low and slow to go high, or even delayed off the reset pulse.

HTH,

Ian.

[arivel]

I wouldn't bother with bistable relays just to hold a logic state - too big, expensive and difficult to drive.

Instead, I'd use a CMOS 8 bit transparent latch with a backup battery and a PSU supervisor IC + a bit of glue logic, to latch the contents when the power goes away.   e.g.if you choose 74HC573 which will hold its state down to 2.0V and a leading brand CR2032 coin cell as the backup battery, you can expect over 30 months of power-off data retention per battery.

There are a few devils in the details: you need a pair of low leakage Schottky diodes to route both battery power and +5V to the latch so the battery takes over when the 5V goes away, cathodes to the latch Vcc pin, anodes to each power source. e.g the common cathode BAS70-05 low leakage double Scottky diode, which typically leaks less than a uA at 5V reverse voltage and 85 deg C.

The74HC573 needs a pullup on its /OE pin to its own Vcc pin to turn its outputs and something to pull that pin low when the 5V rail is up.  e.g a SN74LVC1G06 single gate open drain Schmitt inverter, powered from the 5V rail with its input fed from a potential divider with 80% of the 5V rail so it doesn't pull the /OE pin low till the 5V rail is well above the battery voltage and a circuit to take LE low and latch the current data as soon as the 5V starts to dip, but only let it go back high after the reset pulse is finished, so the data on its outputs from the previous session is latched into the counters parallel data inputs when their pre-load pins go high.  Basically whatever drives LE must be quick to go low and slow to go high, or even delayed off the reset pulse.

HTH,

Ian.

Hello and thanks .
I have evaluated your solution and I thank you but I prefer to stay on the six bistable relays.
now they make very small encapsulated and I assume silent relays costing a € or so.
there are different implementations depending on the type of relay they can be
single coil, double coil with 4 terminals or two coils with three terminals (two are shared)
this without counting the terminals of the contacts. I have thought of a solution that makes use of a dual power supply
(the voltage + 15-15 is only indicative) with transformer with central socket
(the one I like more than the others) and single coil relays but I don't know if it can work, you will tell me this.
a pattern is better than words.
it is very simplified

[arivel]

I was wrong to place the capacitor.
I would place it before the push pull along with a resistor.

[Ian.M]

IMHO you are still on the wrong track, however I suppose I should be glad you haven't decided to build a six bit magnetic core memory (see: https://sites.google.com/site/wayneholder/one-bit-ferrite-core-memory )

Your idea in reply #16 will only produce approx +/- 1.8V drive to the relay (assuming a big enough capacitor so that the drop across it is negligible).  The modification you describe in reply#17 reduces the capacitor size required but doesn't help the drive levels.  Repeated from my reply #11:

... look at H-bridge ICs, or consider the other suggestions in the thread https://www.eevblog.com/forum/projects/how-to-drive-latching-relais/

[arivel]

Hello and thanks .
which infrared chipsets can I use to remotely control the Up / down inputs of the meter?
I would also need a third button for the mute function.