Why Power transformer has 15 more volts after rebuild?

[RockinRocket]

I rebuilt my 18 watt lite again this time in a new chassis with nos parts, Phires /Mustards and used the same tranies, jacks, sockets, switches, tubes ect. Now Im seeing different volts from the transformer. Same wall volts of 118.5 - 119 for these tests . My power transformer now pushes out 15 more volts.


Here are my volts prier to the rebuild for the 32/ 32 can capacitor. With multiple testings on different days with the same wall volts I was getting 345-325 with the 1.5 k across the cap. After the rebuild Im getting 360-340 with the 1.5 k across the cap.

I don't know if this matters but I reversed the connections to the first cap can as my lead was a hair to short to the output transformer. Also the xicon resistors I used in the prior build for the plates were pretty much 100k close to tolerance. The Phires very from 98 to 103 k. Don't know if that matters.

Why is my power transformer pushing 15 more volts? I did use a 18 gauge mains compared to 16 gauge in the prior build.

Edit- I also didn't include a 220k to ground from the first filter cap like I did prior. That's got to be why I lose 15 volts?

[Leo_Gnardo]

I also didn't include a 220k to ground from the first filter cap like I did prior. That's got to be why I lose 15 volts?

That would account for some but not all of your voltage bonanza. How much? well it's easy enough to clip in that 220K and find out. Also have to wonder how healthy that 32+32uF is. If the voltage went up simply by swapping sections, perhaps one of those 32's is leaky.

[ToneMerc]

Edit- I also didn't include a 220k to ground from the first filter cap like I did prior. That's got to be why I lose 15 volts?

No bleeder resistor the second time around would certainly explain the 15V delta.

TM

[martin manning]

I'm betting Leo has it right. The section previously used as the first filter has been dealing with higher voltage and ripple current for some time, and may have deteriorated. I can't imagine that the 1.5 mA saved by omitting the bleeder would make that much difference.

[RockinRocket]

Now after playing for a few hours on it the volts are going back to how they were previously. Last night and this morning I was getting about 360 after rectification. Its at 350 for a consistent wall volt of 119.2

I am going to add a bleeder to see what it will do for volts. Im interested in that especially since Ingrid has a different bleeder value than the others.

Edit- I have took measurmnts of the 220k bleeder in and out of circuit and the difference wasn't measurable.

[JazzGuitarGimp]

I'm betting Leo has it right. The section previously used as the first filter has been dealing with higher voltage and ripple current for some time, and may have deteriorated. I can't imagine that the 1.5 mA saved by omitting the bleeder would make that much difference.

It's most likely the other way 'round: electrlytic caps have a memory. If you run a 350V lytic at 250V for its life, its rated voltage will decrease to 250. I have this on good authority from a BSEE whom I trust. His rule of ghumb is, don't overspec lytic voltages because they degrade to the circuit voltage anyway, so you're wating your money by overspecing. He uses the first voltage rating over what the cap will see in-circuit. So a 310V circuit would get a 350V cap.

I would guess the first cap (which used to be the second, lower-voltage-connected cap) is now pulling current because it isn't used to the higher voltage it now sees.

[RockinRocket]

I would guess the first cap (which used to be the second, lower-voltage-connected cap) is now pulling current because it isn't used to the higher voltage it now sees.

The 1.5k across the cap drops 20 volts. I wonder if that's enough to mess with its memory.

I keep checking the wall volts so I can have an accurate comparison. I'm about 5-7 volts off of what it was prior however the wall volts are 1 - 1.5 volts more. Maybe the can is getting used to the new voltages. The amps got about 4 hours on it now.

[IMG769]http://i1225.photobucket.com/albums/ee383/superchampable/22a0cfd3-11cb-4959-a3c8-2297a0fd570e.jpg[/img]

The Phiers,Iskras, Mustards, Lemco ceramic and CC screen resistor are the only tone changers and it sounds better than it did. Placebo effect?
Its a tad more noisy than before.. same grounding scheme as prior. The mains wire layout is new and copys a original 18watt. Maybe that has an effect. Using CC as a screen resistor is new this time and I bet that adds some noise as well.

[Structo]

That is strange.

I was going to suggest elevated wall voltage but it looks like you checked that.

[JMFahey]

I wouldn't overthink it, voltage difference is nil, around 4%, absolutely irrelevant.

[RockinRocket]

ya I don't know they are much closer to normal now.