volume attenuator relay

Thread Starter

arivel

Joined Mar 4, 2018
42
Hello.
I would like to make a volume attenuator composed of bistable single coil latch relays in logarithmic resistor ladder configuration.
to do this I thought of using a binary counter 40193 whose outputs go to an operational amplifier. the function of the operational amplifier is to send a pulse to the TC4420 half-bridge mosfet (attention, only one pulse at each change of state of the 40193) which controls the relay coil. to drive the bistable latch relays it is customary to use a pulse in one direction or the other without leaving the coil under voltage all the time.
as you can see from the attached image the other end of the coil is connected to the zero of the dual power supply.
to simplify, only one component of each type appears in the image (opamp, tc4420, relay), but in the real circuit there are at least six.
the problem is the following:
which op amp should i use? and which configuration is the most suitable?
 

Attachments

KeithWalker

Joined Jul 10, 2017
1,857
I would use a microcontroller in place of the 40193, the op-amps and pulse generators. An ATMEGA386 would be my first choice. What are you planning on using for an input?
I made one using low power, non-latching relays. It switches from 0db to -22.5db in 1.5 db steps. I control it with a TV style IR controller.
 

Thread Starter

arivel

Joined Mar 4, 2018
42
I would use a microcontroller in place of the 40193, the op-amps and pulse generators. An ATMEGA386 would be my first choice. What are you planning on using for an input?
I made one using low power, non-latching relays. It switches from 0db to -22.5db in 1.5 db steps. I control it with a TV style IR controller.
thank you for your answer but I am not able to program, I am of a certain age and I do not really want to start studying a programming language, plus I am fond of old electronics. by input you mean what controls the inputs of the 40193? .
there are two nand gates.
 

BobTPH

Joined Jun 5, 2013
3,506
Is this a low level signal, or high power? If low level, analog switches would be a lot easier to use than latching relays.

Bob
 

Thread Starter

arivel

Joined Mar 4, 2018
42
in high level volume attenuators the reason for which relays are used is to have no semiconductors that dirty the signal
 

Ian0

Joined Aug 7, 2020
2,713
1) There are latching relays with two coils, pulse one for open and the other for closed, so you don't need a bridge driver of positive and negative supplies.
2) Latching relays are generally power relays, with silver oxide contacts, and they can make a real mess of the signal. If there is no substantial current flowing they start to behave like diodes. You must use relays with gold contacts, or reed relays which enclose the contacts inside a tube full of inert gas.
Signal relays are smaller with higher resistance coils, so there is no need for power saving and so the latching type tends not to be available.
 

AnalogKid

Joined Aug 1, 2013
9,296
In your schematic, neither opamp input is connected. Those inputs are important.

As you have discovered, the 4420 is the wrong part for the job.

If I understand you correctly, you want a relay to latch on with no continuous power when a counter output pin goes high, and latch off with no continuous power when that same counter output pin goes low.

Correct?

ak
 

Thread Starter

arivel

Joined Mar 4, 2018
42
In your schematic, neither opamp input is connected. Those inputs are important.
the inputs of the opamp are not connected because it is the object of the question.
If I understand you correctly, you want a relay to latch on with no continuous power when a counter output pin goes high, and latch off with no continuous power when that same counter output pin goes low.
this type of relay have only one coil and work as follows:
to operate the contacts, a pulse must be sent to the coil. if you want to change the state of the contacts, an impulse must be sent in the opposite direction to the previous impulse.
then the pulses flow in both directions.
is it clear now?
 

Ian0

Joined Aug 7, 2020
2,713
Why is it so important to use the latching relay? For a signal relay the coil current is hardly significant.
But if you do really need a latching relay, then go for the double-coil type.
Put a 74HC123 on each output of the counter, one half triggered on the negative edge, one half on the positive edge, and drive each coil from the Q output with a transistor.
 

AnalogKid

Joined Aug 1, 2013
9,296
IF you run the relays on one power supply voltage only

AND IF you drive each relay with a full H-bridge using both the 4420 and 4429

THEN you can eliminate the opamp and drive the 4420 and 4429 directly from the outputs of the counter with R-C networks.

If the relay coil current is below 200 mA, you can use two 555's per relay as an el-cheapo pulse circuit and H-bridge driver.

Is this school work, for a commercial product, or something else? And, where are you located?

ak
 

Ian0

Joined Aug 7, 2020
2,713
If the relay coil current is below 200 mA, you can use two 555's per relay as an el-cheapo pulse circuit and H-bridge driver.
Neat, but you need an inverter to trigger one of them.
I always see a series diode when the 555 is used to drive a relay to ground. The application note says it is for high-q inductive loads, and that the output must not be driven below ground.
I suspect that a schottky diode to ground from each output would also prevent that from being a problem.
 
Last edited:

AnalogKid

Joined Aug 1, 2013
9,296
555 output never goes through a high-impedance state when transitioning in either direction (an often-bitched-about shortcoming of the original design), and this prevents a large inductive spike from occurring. Still, a diode is cheap insurance.

ak.
 

Thread Starter

arivel

Joined Mar 4, 2018
42
both the H bridge and the dual coil relay require two commands inverted with each other and then after that it also takes an additional circuit to create the pulse, so in total 4 circuits to be added.
it seems to me that my system is less complex and with fewer components, just one opamp.
 

Ian0

Joined Aug 7, 2020
2,713
This is probably the simplest version I can think of to use a split supply. Getting the right values of C and R could take some experimentation, as the relay gets an exponentially decaying pulse, not a nice rectangular one.
Screenshot at 2021-04-15 21-36-51.pngIt requires a 6V relay, and the logic has to run off +/- 7.5V supplies (15V from Vdd to Vss)
 
Top