validation of relay power supply greatly appreciated :)

[ACHIEVEIT]

Hi guys,
I would really appreciate some feedback on a relay power supply. I have checked a few threads here + the info on the site and i would like to know whether i have understood everything correctly.

I want a simple channel switching relay which i will power from the 6v3 heater winding. The winding is center tapped so it is more accurate to say 3v15-0-3v15 or 6.3VCT. It is my understanding that in order to power the relay from a center-tapped winding, i need to keep the common of the relay (and whatever connects to it) floating, meaning disconnected from the chassis. The way i have done the wiring and if my calculations are correct (shown on the picture), with the footswitch disconnected, there will be ~7.5v across the relay and with the footswitch connected it should be ~5.6v (because of the footswitch led voltage drop).

Do you think it will operate correctly? The way i have wired it, the panel led operates regardless of whether the footswitch is connected or not and if the footswitch is connected, the panel switch is bypassed. Do you think i should add a couple of diodes in series with the panel switch in case 7.5v is too high?

I will use a cliff jack so it will be isolated from the chassis

Thanks a lot

Ps. A few details that are not shown on the picture: C=4700uf, i live in europe so mains is 50Hz and the relay is a 5v relay

[EL34]

No need to re-invent the wheel

I sell the relay power supply boards and single relay boards here.
http://hoffmanamps.com/MyStore/catalog/parts20.htm

They work great, look nice and the price is right

[eleventeen]

Delete the grounded CT from the power transformer! That cannot be there on a bridge rectifier.


Which immediately poses a quandary: You almost certainly need that grounded CT to kill hum. 95% of amps do, either that way, hard-wired, or via synthetic center tap (the two 2 qty 100 ohm resistors to ground from the "hot" 6.3 winding wires.)


So you cannot have both at the same time with your bridge rectifier scheme. Furthermore, there is the implication that the "sleeve" or ground connection of your panel-mounted footswitch jack will make such a prohibited connection unless it is insulated from the metal of the chassis.


First thought is to completely disconnect the transformer CT from any usage at all and only use a dual-100 ohm center tap. I don't see that working.


Second thought is to retain the wired CT as is and to build your relay/LED supply using a voltage doubler off only one half of the filament winding. This will save you two diodes but cost you or or two more caps. I believe this would solve your problem.


Last thought: 20 mils is on the highish end of what you want to power an LED. Probably a FAT LED can handle it.
 
[/URL]

[EL34]

You may have missed this page on the Library page?
http://el34world.com/projects/relay_switch.htm

[sluckey]

Delete the grounded CT from the power transformer! That cannot be there on a bridge rectifier.

Sure it can. You just cannot connect the negative side of the bridge to ground whenever the CT is grounded.

[Willabe]

You may have missed this page on the Library page?
http://el34world.com/projects/relay_switch.htm

That page has grown! Very nice work Doug!


              Brad     

[EL34]

Always trying to illustrate things more clearly.

[PRR]

The grounded CT is fine *IF* _all_ the switch/relay stuff (including jack and footbox!) can forever be UN-grounded. Using 1/4" plugs may be unwise. While insulated jacks are easy (Cliff), plugs that stand-up are metal. Now drop another plug (guitar, mike, metal power-box) on that plug. Heater winding is shorted through the diodes.

Yet we often get away with this.

What I see is that a 30mA relay with series LED (in footbox) may over-heat a typical 20mA LED.

Also I wonder if 4,700uFd is needed.

[ACHIEVEIT]

Thanks a lot for all your answers guys.

I have definitely seen doug's extremely informative page but since it didn't cover my case (center-tapped heater winding) i continued my search in the forum where i stumbled upon this thread:

http://el34world.com/Forum/index.php?topic=14348.0

where sluckey among others supported the floating psu design for this case.

Unfortunately i live in europe, so shipping costs are killing me if i get doug's very nice and pleasantly compact boards :(

About the led consumption, you are right PRR. Maybe it's better to use a low-current 2ma led for the panel, so that the total consumption is 32ma, and a normal led for the footswitch which would be happy with around 15ma. I guess it would be a good idea to put a resistor in parallel with the led/led-resistor in the footswitch so that the 32ma do not pass all from the led but they split.

Also, i had never thought about someone (even me :) ) accidentally putting something else in the footswitch jack. I will look for alternative connectors.

Other than these two considerations, is there any other reason why this circuit would not work? Are the voltages okay? (Especially the 7.5 v on the relay)

Would i be better off with the alternative implementation proposed in the aforementioned thread - i.e. leds are off when the relay is on and the other way around? Would it be better anyway considering i have a clean and a lead channel (so that the lead that is hum-sensitive doesn't get polluted by the relay hum)?

Thanks again!

[EL34]

Unfortunately i live in europe, so shipping costs are killing me if i get doug's very nice and pleasantly compact boards :(

I'll do cheapo air mail for you

[PRR]

> the 7.5 v on the relay

You have not said *what* relay. "5V 30mA" isn't a specific relay. Take the model number and find the data-sheet. On there it should say what the max/min voltage really is.

However note that your pedal LED is in series with the relay. If red it will drop about 1.7V; if green it may be near 3V. I suspect your relay may be more under-volted than over-volted.

This gives a 2-wire control without all relay current flowing in the foot-LED.