Question about transformer rectification in an EHX Echoflanger

[Govmnt_Lacky]

I am powering an Echoflanger with a 24VAC center tapped transformer rated to 90mA. The AC inputs each feed a 1N4001 rectifier diode and then the converted DC output from the diodes is summed and sent to a 15V regulator.

Now..... I am reading 13.7VAC coming from BOTH of the AC inputs BUT, I am only reading 16.5VDC at the summing point and at the regulator input 

I thought that I should see a HIGHER DC voltage after the rectifier diodes. It looks like I am only seeing HALF!

I doubt that the circuit is drawing more than 90mA.

This started because I am getting some serious hum when I crank the amp. I removed the transformer from the circuit and fed 24VDC regulated into the 15V regulator and the circuit is silent (no hum) so I know that the problem is either the rectifier diodes OR the transformer itself.

Thoughts?? Bad transformer? Is my assumption incorrect about the rectified DC output of the 4001s?

[Tubebass]

What do you have for filtering? I'd want at least 1000 uf to ground after the rectifiers.

[R.G.]

You've hit one of the versions of the law of diminishing returns. 

From your description, you're feeding each end of a 24Vct winding to a diode, and grounding the CT. Each half reads 13.7Vac. That means that each half winding will alternately provide a maximum of 13.7Vac * 1.414 = 19.37 to the rectifier, and you'll then lose 0.7V to the diode. Down to 18.67.

You don't mention a filter cap. A filter cap will charge to the peak ( see http://www.geofex.com/Article_Folders/Power-supplies/powersup.htm) it can, but this will be reduced by the resistive losses in the windings. On such a tiny transformer (24v*0.09A = 2.16W) the resistive losses can be severe. You could easily lose 2V to current peaks from the rectifiers.

If you don't use a filter cap, your meter reads the DC average of the incoming voltage, not the peak held up by a cap. That's lower still.

I'd say you're getting about the right amount from that setup.

[Govmnt_Lacky]

Thanks for the replies.

I am using a 470uF/63V electro to ground directly after the rectifying diodes and in series with the 15V regulator input. I also have a 470uF/25V to ground in series with the regulator output. Both in step with the original schematic.

I just do not understand why the rectified output is SO low. I would figure to see at least 20-22Vdc into the regulator.

I understand what you are saying RG but, something still does not seem right. Only getting 16.7v into the regulator seems wrong.

[armdnrdy]

If you were using this center tapped transformer to create a bipolar supply (secondary windings in series) you would read 24+ volts at the output of the transformer.

In this application the transformer is used in parallel so your reading is correct.
12volts X 1.414 = 16.97

[Govmnt_Lacky]

If you were using this center tapped transformer to create a bipolar supply (secondary windings in series) you would read 24+ volts at the output of the transformer.

In this application the transformer is used in parallel so your reading is correct.
12volts X 1.414 = 16.97

I see now....

Well, right now I am wondering if the huming is coming from the transformer itself OR if it is due to the low input voltage to the regulator and it is struggling to output 15V (it is only putting out 14.75V with the current transformer)

I do have the same transformer with a series output (true 24V). I wonder if that would help 

[armdnrdy]

Looking at this again...

I gave a rough calculation of rated voltage, R.G. gave calculations of "actual" voltage.

I see what R.G. means by his statement "I'd say you're getting about the right amount from that setup."

This is a small transformer! but it's working as rated.

The regulator seems to be doing it's job as rated too.
LM7815
MIN.   TYP.   MAX
14.4    15     15.6


HUM......Hmm
Is any of the signal wiring in close proximity to the transformer?

[Govmnt_Lacky]

HUM......Hmm
Is any of the signal wiring in close proximity to the transformer?

I already thought about this. I actually put some shielded braiding around the power feeds from the transformer to the PCB input. No change 

I think I am going to do a full "noise reduction" rework. Going to replace the transformer, shield the V+ inputs, swap out the input and both output lines with shielded wire. See where that goes.

More to come possibly tomorrow 

@Larry--- Anxiously awaiting the Mutron project! Sounded KILLER!!!

[armdnrdy]

@Larry--- Anxiously awaiting the Mutron project! Sounded KILLER!!!

I've been working on it all day! 

I had to redo things that I had completed in advance......wiring diagrams, parts BOMs, etc, because a few things changed at the end. Ticking issue fix (board revision) Custom rate pot work around (board revision) (BOM revision)

This is like work! 

[Govmnt_Lacky]

I've been working on it all day!  

I had to redo things that I had completed in advance......wiring diagrams, parts BOMs, etc, because a few things changed at the end. Ticking issue fix (board revision) Custom rate pot work around (board revision) (BOM revision)

This is like work!  

Well, I just want you to know that I appreciate all your hard work!!!!!!!!!!  

[PRR]

24VAC CT is 12VAC each side.

IMHO this is not enough to power a 15V regulator.

Read through R.G.'s points. Yes 13VAC is 19V peak, but then you have losses. At this scale, perhaps large losses. He has you down to 16.67V DC, you see 16.5V DC, reasonable agreement. You don't say what regulator, but a '7815 has 2V dropout at 1 Amp. 16.5V-2V= 14.5V.

Very likely the "15V" is 15V with 14.5V dips in it 120 times a second.

Another half-volt input fixes it, *today*. But go to a club where the "120V" is 110V, it'll buzz again.

A lower dropout regulator buys a little more headroom, but still may go buzzy if the light-board or pizza-broiler pulls the club voltage down to 108V. (And I've worked in a place where 108V was normal.)

A rule-o-thumb for little supplies is that the AC Volts RMS should be a bit +more+ than the regulated volts required. For 15V DC out, use 15VAC (or 30VAC CT).

Alternatively: assess if your dingus can live just as well on 12V DC. A hasty glance at the Echoflanger suggests that it might. Or might not.

[Seljer]

If you don't have a oscilloscope, you can get some preliminary measurements on how much 50hz (or in this case 100hz because its rectified) ripple there is on the supply with the AC voltage measurement mode on your multimeter. It wouldn't stand up in a specifcation sheet but it'd be good enough to verify if its the source of your problem.

[Govmnt_Lacky]

Thanks all for the responses.

Taking it one step at a time. Replaced the transformer with a 24VAC output (no CT). After rectification, it should feed the LM7815 with enough voltage to keep the 15V threshold BUT, should stay well below the 30-35V ceiling 

Will test and post results later tonight or tomorrow 

QUESTION: Will there be a problem with using the series wired transformer vice the CT version? I am not sure whether or not the series V- will effect the circuit. Just wondering if this could create a whole new set of issues.

I am wiring it like this:

110VAC Hot ------------- Transformer ------------- PCB V+ Input

110VAC Neutral --------- Transformer ------------ PCB V- (Gnd) Input

Safety GND ------------- Pedal Chasis

Anyone see any potential problems?

[R.G.]

QUESTION: Will there be a problem with using the series wired transformer vice the CT version? I am not sure whether or not the series V- will effect the circuit. Just wondering if this could create a whole new set of issues.

I am wiring it like this:

110VAC Hot ------------- Transformer ------------- PCB V+ Input

110VAC Neutral --------- Transformer ------------ PCB V- (Gnd) Input

Safety GND ------------- Pedal Chasis

Anyone see any potential problems?

If you really wire it like that, yes. I'm guessing that there's a bridge rectifier and filter in there between the transformer and PCB pads. Otherwise the AC will cook whatever's on the PCB.

[Govmnt_Lacky]

If you really wire it like that, yes. I'm guessing that there's a bridge rectifier and filter in there between the transformer and PCB pads. Otherwise the AC will cook whatever's on the PCB.

Not really 

V+ from Transformer (24VAC) ----------> 1N4001 diode -----------> 470uF cap to V- ------------> Regulator Input

V- from Transformer ----------> PCB V-

Essentially, I removed the CT transformer entirely and only re-used 1 of the PCB V+ input pads with one of the Transformer leads and sent the other transformer lead to V-

Problems? Do tell please 

[R.G.]

OK, there's a rectifier and a filter in there, just not a full wave rectifier.  It's OK.

I assumed that a FWB was there, because they're cheap, easy to get, and give you twice the filtering out of your filter cap compared to a half wave rectifier. Have I mentioned Power Supplies Basics at Geofex? ( see http://www.geofex.com/Article_Folders/Power-supplies/powersup.htm)
But this works. The series diode rectifiers the incoming AC, it's filtered in the cap, and passed to the PCB.

Reading back over the thread, you feed a 15V regulator with this. The regulator will knock down ripple by about 40db (i.e. by a factor of 100) as long as the voltage feeding it is at least 17V (if it's a 7815).

This 24Vac/diode/filter puts out about (24*1.15*1.414)-0.7 = 38Vdc with no load, sagging to 33.2Vdc at about 60ma. That's peak. The filter cap runs down between peaks by dV= I*dT/C where I is the load current. If you have 90ma of load current, the voltage sags by dV= 0.09*0.015/.00047F = 2.8V, so the minimum is between 35.2 and 30.4. Good. Your regulator won't run out of voltage.

The regulator will reduce the 2.8V of ripple to about 0.028V. Not too bad.

You're probably OK.

[Govmnt_Lacky]

OK, there's a rectifier and a filter in there, just not a full wave rectifier.  It's OK.

I assumed that a FWB was there, because they're cheap, easy to get, and give you twice the filtering out of your filter cap compared to a half wave rectifier. Have I mentioned Power Supplies Basics at Geofex? ( see http://www.geofex.com/Article_Folders/Power-supplies/powersup.htm)
But this works. The series diode rectifiers the incoming AC, it's filtered in the cap, and passed to the PCB.

Reading back over the thread, you feed a 15V regulator with this. The regulator will knock down ripple by about 40db (i.e. by a factor of 100) as long as the voltage feeding it is at least 17V (if it's a 7815).

This 24Vac/diode/filter puts out about (24*1.15*1.414)-0.7 = 38Vdc with no load, sagging to 33.2Vdc at about 60ma. That's peak. The filter cap runs down between peaks by dV= I*dT/C where I is the load current. If you have 90ma of load current, the voltage sags by dV= 0.09*0.015/.00047F = 2.8V, so the minimum is between 35.2 and 30.4. Good. Your regulator won't run out of voltage.

The regulator will reduce the 2.8V of ripple to about 0.028V. Not too bad.

You're probably OK.

LM7815 Maximum Voltage input = 35VDC   

Would there be any SAFE way to trim that input voltage to the regulator down just a bit to be on the safe side? I ASSUME that this circuit has quite a draw to it (if you have seen it you would know why) but, I would still like to be safe.

Thoughts?

[Govmnt_Lacky]

OK.....

Tried running the series transformer like I mentioned above (one output to D1 and the other to GND). Still getting the hum. I upped the filter cap after the rectifier diode and before the regulator (C38) to 1000uF. Still get the humming.

Checked all of my grounds. All good.

I know that the problem lies somewhere in the transformer portion of the circuit. When I apply 24VDC directly to the regulator input, the circuit is quiet as can be. (injected the 24VDC at the connection point of C38 and the regulator input)

So... should I try a full rectifier bridge type circuit? Is there any way to filter the incoming mains power to try to get rid of the hum?

BTW: I was also getting the hum with a CT transformer wired identical to the schematic below.

EDIT: See schematic below 

[armdnrdy]

I think that an LC filter might be the next thing to try.

Let's tame that wild schematic!

[Govmnt_Lacky]

I think that an LC filter might be the next thing to try.

Thanks Larry.... might need some direction on implementing the LC filter. Thoughts??

Are there any other ideas on how to get rid of this hum??

EDIT: Another item to note. C41 on the schematic IS NOT on the PCB. It is not on any of the original units either. Just throwing that in for consideration.