Relay true bypass power Consumption

[jackwithoneye]

Hi,
i integrated a relay bypass circuit in a pedal, i'm surprised by the power drained by this circuit (more than 53mA, only from this circuit, not the effect).
I knew that using NE555P will drain a lot of power, but i expected more in the 25-30mA range.
I'm using NA5WK  takamisawa, fujitsu relay.

Does it ring a bell to someone?
Thank you!

[GibsonGM]

I looked at the data sheet for the relay...is this a 5v or 9v relay?  Isn't it not 'what you expect', but what the manufacturer lists for operating current of the relay? ( voltage/coil resistance)

Quick way to know is to measure the relay's current draw when you actuate it.  If it's a 5v model running on 9v expect higher draw (and probably blue smoke)!

[antonis]

According to datasheet https://www.fujitsu.com/downloads/MICRO/fcai/relays/na.pdf, your relay coil is rated at 5V where you feed it with 9V (minus a BC549 V_CEsat), hence it should draw much more than 28mA current by its own..
Adding 10 - 15mA for NE555 plus LED current (?) we come close to your measurement..

Either place a coil series resistor for 4V drop (according to coil nominal current - page 3), or use a NA9WK, or live with that current consumption and a very warm relay..

edit: Sir Mike is by far faster..

[jackwithoneye]

According to datasheet https://www.fujitsu.com/downloads/MICRO/fcai/relays/na.pdf, your relay coil is rated at 5V where you feed it with 9V (minus a BC549 V_CEsat), hence it should draw much more than 28mA current by its own..
Adding 10 - 15mA for NE555 plus LED current (?) we come close to your measurement..

Either place a coil series resistor for 4V drop (according to coil nominal current - page 3), or use a NA9WK, or live with that current consumption and a very warm relay..

edit: Sir Mike is by far faster..

I looked at the data sheet for the relay...is this a 5v or 9v relay?  Isn't it not 'what you expect', but what the manufacturer lists for operating current of the relay? ( voltage/coil resistance)

Quick way to know is to measure the relay's current draw when you actuate it.  If it's a 5v model running on 9v expect higher draw (and probably blue smoke)!

Thanks for your answers guys
i'm using NA5WK so it's the 5V version




Putting  a 142ohm resistor in serie with the relay coil (178ohm) to get about 5V on the coil should shift down a bit the current drained and releive the coils pain?

As i can't find any 9v relays like the NA9WK (unavailable), best choice would be to use this serie resistor to prevent the 5V relay from burning?

Seeing the constuctor datasheet above, wouldn't it be perfect to use the 12V version (NA-12WK), which is available, has a 1K coil, and has a "must operate Voltage" of 9V? Or this "must operate voltage" being too close to the supply voltage may introduce some troubles?

[antonis]

Powering with 9V a 12V nominal voltage relay is like continously balancing on ineffectiveness cusp..



A rather long but usefull text about relays:
https://media.digikey.com/pdf/Other%20Related%20Documents/Panasonic%20Other%20Doc/Small%20Signal%20Relay%20Techincal%20Info.pdf

I should use LED + CLR(properly sized) between relay coil and BJT Collector if I was sure about LED long life..

[GibsonGM]

Try a standard value of 150 for the dropping resistor, that should work. If you had to you could go a bit lower using 2 R's in series; test its operation.  I'd say if you HAVE the 5v relay, use it. If you BUY one, get the 9V version. There's nothing wrong with dropping the voltage.

Running the 12V on 9V does seem too close to the minimum value, and you may not really be just at 9V..what if you have 8.2V? (I'd test the operation w/dropping R down to at least 8V to be sure it works w/low battery). 

Antonis is still awake where he is - I am just waking up this morning so I am slower 

[Rob Strand]

When you run a 5V coil from 9V using a series resistor you want to match the *current* for the relay 5V spec.

If you then compare that current to the 9V coil current you will see the 5V coil + resistor requires a lot more current.
A 9V coil gets its resistance from more turns of thinner wire on the coil.   The turns contribute to the pull so you need less current.   In the 5V + resistor case the resistance lowers the current but it does not contribute more turns to the coil.



For ultimate low power look up latching relays.   The don't use any power once they are switched to a particular state. The way you drive these relays is different.

[jackwithoneye]

thanks you for your answers again.

I get that the resistor in serie could cause more problem, and that the best thing is to have a real 9V relay.

The NA9WK i would need, is discontinued, and really not findable where i live (in europe), or at a decent price (i just can't pay more than 10$ for a relay...)

I have tested/breadboarded with a RY9WK i found, which have a different footprint, different pinout and is bigger. I have to do a new pcb to fit it in.
RY9WK datasheet : https://www.datasheet-pdf.info/attach/1/3098632639.pdf
NA9WK datasheet : https://www.fujitsu.com/downloads/MICRO/fcai/relays/na.pdf

It's working good, with a maximum current drained of 30mA, which is perfect to me.
Nasty popping is just bothering me....

nb : I found this very useful for Transistor Relay driver, full of useful datas : https://www.homemade-circuits.com/how-to-make-relay-driver-stage-in/

[antonis]

Nasty popping is just bothering me....

If you post a schematic of circuit of interest (effect On/Bypassed) we might find a remedy..

[GibsonGM]

Yes, please post the full schematic.   

FYI. There is no problem with a dropping resistor; it is operating at 5V 'at the top of the relay', but since you found a 9V that will make is easier for you!

[jackwithoneye]

here we go!
Actually, the pop is not "That bad", it happens especially when the FX output volume is high.

It's a tone bender inspired fuzz with npn transistors.
Note that there is no pulldown resistors on the FX schematic, but there's one next to the JACK OUT switch of the relay (R7).

the FX part :



the relay bypass part (with the old relay before last day 9v relay):

[antonis]

Although 100nF shouldn't pop heavily, place a 1M resistor between R8/C3 and GND..

[jackwithoneye]

Although 100nF shouldn't pop heavily, place a 1M resistor between R8/C3 and GND..

I tried but it doesnt reduce the pop, which is not a bassy pop, more of a "trebly clicky pop" (it's not the click from the relay ) it's coming out from the speaker

[antonis]

Does popping occur both for ON and OFF switching..??

[jackwithoneye]

Does popping occur both for ON and OFF switching..??

yes

[Rob Strand]

yes

The effects circuit (not shown) should have a resistor to ground on the FXOUT line, say 100k; you might already have this.  R7 is probably not required (and is perhaps undesirable as it loads the input in bypass mode).  As a precaution, you could optionally add a 1M to 2M2 from FXIN to ground.

It could be the switching circuit itself.  Perhaps even caused by the relay current changing.  Add a 100uF cap from +9V_PAD to ground.   If you have a resistors in series with the supply rail for filtering the switching circuit (+relays) should be powered before the filtering resistors.

Make sure you have diodes across the relay coils, like in the link you gave.

[jackwithoneye]

i posted the effects circuit few posts above, here it is :

the FX circuit :



the relay bypass part (updated with the 9v relay, R7 became R3):

yes

The effects circuit (not shown) should have a resistor to ground on the FXOUT line, say 100k; you might already have this.  R7 is probably not required (and is perhaps undesirable as it loads the input in bypass mode).  As a precaution, you could optionally add a 1M to 2M2 from FXIN to ground.

It could be the switching circuit itself.  Perhaps even caused by the relay current changing.  Add a 100uF cap from +9V_PAD to ground.   If you have a resistors in series with the supply rail for filtering the switching circuit (+relays) should be powered before the filtering resistors.

Make sure you have diodes across the relay coils, like in the link you gave.

-ok i'll try to move the pulldown resistor R3 (ex-R7) to the FXIN instead of on JOUT.
- the 100uF after the protection diode on the power supply part of the FX circuit is not enough? Should i add the small resistor you're talking about and put this 100uF cap before?
-The diode D1 is in place over the coil.

[Rob Strand]

i posted the effects circuit few posts above, here it is :

Sorry, I somehow missed that.   It changes what I said a bit.

the relay bypass part (updated with the 9v relay, R7 became R3):

OK probably don' t need that resistor.

Also, for the FXOUT output.   No need for a resistor to ground as the existing output pot on the effect already does the same thing.

Optional 1M to 2M2 resistor on the input worth a long shot (it won't hurt).

-The diode D1 is in place over the coil.

OK looks good.

- the 100uF after the protection diode on the power supply part of the FX circuit is not enough? Should i add the small resistor you're talking about and put this 100uF cap before?

Well these types of circuits can be very sensitive to noise on the supply.

You might try:
+9V ---diode ---- 100uF to ground [this rail to switching ckt pwr] ---> 100R ---> 100uF to ground [this rail to effect power]

If you hear some improvement then the power is likely to be the cause.

[jackwithoneye]

i posted the effects circuit few posts above, here it is :

Sorry, I somehow missed that.   It changes what I said a bit.

the relay bypass part (updated with the 9v relay, R7 became R3):

OK probably don' t need that resistor.

Also, for the FXOUT output.   No need for a resistor to ground as the existing output pot on the effect already does the same thing.

Optional 1M to 2M2 resistor on the input worth a long shot (it won't hurt).

OK, this 1M/2M2 has to be directly to the JIN, right? not on the input of the FX circuit post-relay (that i tried, no improvement)?
Does it make a difference if i just solder a 1M/2M2 between Tip and sleeve on the Input jack connector, or it has to be as close as possible to relay on pcb?
And do you think this 1M/2M2 on the input would change the impedance or interfere with the response with the guitar volume pot, considering a fuzz pedal is -sensitive/reactive- for cleaning-up with the guitar volume pot, like a fuzz face?

-The diode D1 is in place over the coil.

OK looks good.

- the 100uF after the protection diode on the power supply part of the FX circuit is not enough? Should i add the small resistor you're talking about and put this 100uF cap before?

Well these types of circuits can be very sensitive to noise on the supply.

You might try:
+9V ---diode ---- 100uF to ground [this rail to switching ckt pwr] ---> 100R ---> 100uF to ground [this rail to effect power]

If you hear some improvement then the power is likely to be the cause.

Thank you a lot!!! i'll try this tomorrow (it's very late in my timezone ) and will tell you if it improves

Edit : i changed the transistor base resistance (R2) from 3K to 220K to match the optimal coil current (17mA at 9V) considering the Q1 HFE (550) like explained in the earlier posted link ( https://www.homemade-circuits.com/how-to-make-relay-driver-stage-in/ )

[Rob Strand]

OK, this 1M/2M2 has to be directly to the JIN, right? not on the input of the FX circuit post-relay (that i tried, no improvement)?
Does it make a difference if i just solder a 1M/2M2 between Tip and sleeve on the Input jack connector, or it has to be as close as possible to relay on pcb?
And do you think this 1M/2M2 on the input would change the impedance or interfere with the response with the guitar volume pot, considering a fuzz pedal is -sensitive/reactive- for cleaning-up with the guitar volume pot, like a fuzz face?

Well, there's no real reason to have the grounded resistors on the either jacks themselves.   The main reason is it loads the input unnecessarily when in bypass - kind of undermining the idea of true-bypass.

Nonetheless there's nothing stopping you doing the experiment to track down very obscure problems.  In which case it doesn't matter which end of the wire you place it.

If you are going to add a resistor to the input then FXIN is the correct place to put it.   Under normal circumstances shorting the input in bypass mode, like your circuit does, makes the FXIN resistor to ground redundant.  If you find no improvement then that's fine.  It just means it's working normally without any weird stuff happening.

FYI you can see the added input resistor on this schematic,

http://www.generalguitargadgets.com/pdf/ggg_bmp_tri_sc.pdf

Thank you a lot!!! i'll try this tomorrow (it's very late in my timezone ) and will tell you if it improves

On paper the noise through the supply theory is high on the list.  It's very much worth trying.  If it gives an improvement but does not completely remove the click that could just mean we need to do more work in that area.  It can also mean there's two problems and the extra caps only got rid of one cause.

Over the years on the forum there have been a few instances where relay circuits clicked as if the switching noise on the coil was getting through to the contacts.

Edit : i changed the transistor base resistance (R2) from 3K to 220K to match the optimal coil current (17mA at 9V) considering the Q1 HFE (550) like explained in the earlier posted link ( https://www.homemade-circuits.com/how-to-make-relay-driver-stage-in/ )

That value is not really "optimal" as such.  It's more a "minimal" base current / "maximum" base resistor case.   What is much more common for transistor switching circuits is to use a base resistor 5 to 10 times smaller than that maximum value.  So in your case that would be 220k/10 = 22k to 220k/5 = 47k.     The 3k on the original is probably OK and lets you use lower gain transistors but it's probably going a bit overboard.    It's not an area that needs a precise value.   You might only split hairs if the circuit driving the transistor has limited output current.