1979 Fentder Twin, Output hum.

[catstrat]

Hi Paul P. I'm still don't know if the one on my bench is a 100 watt or a 35 watt. I always thought the only twins that had the ultra linear taps were the 135 watters. Guess i'm wrong. With 500 + volts, you may actually have the 135 watt model. Every schematic of a silverface twin i've looked at show the U/L taps. Well dummy me, i just looked on the back by the speaker outs and sure 'nuff, it says 135. I thought most twins were just 100 watt unless specified on the front panel. I was wrong.I don't work on that many twins. In this thread a lot of people chimed in on the bias circuit and it sounds like the amp you have has a bias issue as well. If you haven't done it yet, i suggest reading through this thread and there has been a lot of info from the guys here, far more than what i can offer. They give you a different take on how bias circuits are wired and maybe one of them can help you. Good luck and i hope you get it going.

[potatofarmer]

Thanks, catstrat. Actually my bias circuit is fine at the moment, tubes are dissipating somewhere in the low 50%'s at idle and it's putting out a clean 100W into a dummy load before clipping. At least it was before I clipped the NFB and triode-strapped two of the tubes. I should check that again... I'll probably go to a balance & adjust at some point, but for now "if it ain't broke..."

I only brought up the hum balance control because I thought it was odd that (somehow?) the heater winding is so asymmetrical. I'm no humdinger expert, but on the few I've used on amps usually somewhere close to 50% rotation is the null point for hum, or at least somewhere in the 40-60% range. My point being that I suspect your amp is similar, and that putting the humdinger back in (assuming a previous owner didn't fry it - worth checking out) might get you lower hum than the two 100 ohm resistors.

But uh... back to the point g-one mentioned in post #2, the low output and low voltages indicate something else is wrong. The schematic calls for a B+ of 500V (520V in standby) with a 120V wall voltage, so you should be seeing something in that neighborhood on your plates/screens. Unless fixing your bias issues took care of that too. I'm torn on what to do with mine long-term - really don't like using 600V caps, but I like putting >500V on 500V caps even less, so I've either got to try to cram some totem-poles into the doghouse or knock the voltage down with a zener. Someday, at least.

Anyway: how's the output power now?

[catstrat]

Hi Everybody. It seems i've made some statements here that are just incorrect. I said earlier, every twin schematic i've looked at has ultra linear taps. That simply isn't true. I guess that's my brain playing tricks. After reviewing the schematics it's the 135 watt twin, the 135 watt has the U/L taps. So i stand corrected. I've tried various bias configurations and i'm going to stick with the original bias balance circuit because that is the quietest one. I'm still only getting ---62 volts at pin 5 but i'm going to see if i can't get that more inline with the schematic which shows --52 volts i believe. I'm really not too worried about the o/p tube plate voltage. This guy plays old time honky tonk type music and they don't really play that loud. If i can get a good 80 to 100 watts out of it, and have a relatively clean signal with the volume around 4 or 5, i'm going to call it fixed. By the way Paul p, you might want to check the bias balance pot in your amp. Mine had the solder lugs not making good connection to the conductive surfaces of the pot because they were loose. Good luck and happy fixing. Thanks.

[Tom Phillips]

...I'm still only getting ---62 volts at pin 5 but i'm going to see if i can't get that more inline with the schematic which shows --52 volts i believe...

You don't necessarily want to set the bias voltage to that -52 V reference value on the schematic. Remember that you set the bias voltage to obtain the plate current that's required to produce the plate dissipation power (Watts) for the plate voltage and tube type in your amp. Negative 62 volts at pin 5 may be just right for the exact tubes and plate voltage in your amp. If you change it to -52 V then your plate current and plate dissipation will go up. Maybe too far up. You need to measure the actual plate current and voltage as you make the adjustment. (Measure current either directly or indirectly such as by installing 1 Ohm resistors in series with the cathode.
Hope that makes sense.
Tom

[catstrat]

Hi Tom. Finding out what the plate is actually drawing current wise is the best way to do it. The bias voltage on the schematic is more of a reference than anything else, or so i'm told. The safe way to do it without a bias probe of some kind is putting a 1 ohm resistor in the tubes cathode circuit and measure across the resistor. I've done it using the transformer shunt method also, but honestly, that scares the patootie out of me. Some people use the O-scope way, but i hear that's not the best way either. We really need to know what current the plates are pulling at idle. If my plates are at 460 volts then the ma draw should be, according to my Dan Torres cheat sheet, 26 ma to 31 ma per tube. Bias is confusing for me sometimes and i have to stop and think. Thanks all.

[g1]

I'm still a little baffled as to why the plate voltage is so much lower than stock. Especially considering the bias voltage being more neg.than stock, probably due to higher modern line voltages. I'd expect the main B+ to be higher than stock, rather than lower.

[catstrat]

You know G-one, i don't know what to make of the voltage being low. The rectifier diodes are ok. Usually you can have low plate voltage if the bias is set where the plates are pulling excessive amps but then you have other symptoms like red plating plus my variacs ampmeter would show that. But that doesn't seem to be the case here. I replaced some out of spec cathode caps and that fixed most of the early breakup i had earlier. Now i'm getting a pretty strong clean signal. I'll check the plate draw and see where i'm at. I may connect the heater balance circuit and do away with the 2 resistors. I have ran myself ragged in the past trying to fix a problem like this and i've found out if the customer comes over to get his amp and he plays through it first and he's happy with it, then all is well. If the customer is happy, then i'm happy. What a way to run a business!

[JoeM]

.... If my plates are at 460 volts then the ma draw should be, according to my Dan Torres cheat sheet, 26 ma to 31 ma per tube. Bias is confusing for me sometimes and i have to stop and think. Thanks all.

You cant tell what the current draw is based on the plate voltage.

[The Dude]

You cant tell what the current draw is based on the plate voltage.

I believe he's referring to bias recommendations at a given plate voltage. I don't believe he's inferring that tube current can be derived with plate voltage as the only supplied parameter.

[JoeM]

Oh, I see what you mean. For a certain dissipation based on plate voltage.

[catstrat]

Yes, i'm referring to the suggested plate ma according to Dan Torres. All honestly, i don't know enough about the subject to carry on a conversation about it. I generally go by what the tubes suggested Plate MA draw is at a specific plate voltage is. Then i use the old ear test to fine tune it. If possible , i'll have the customer sit by me while i set the bias, just like i do with gain mods. Sometimes if the customer is not available, i just have to wing it. I guess i'm pretty good at it because i haven't had any complaints yet. I would like to know more about mathmatically figuring out the plate disapation though. So if anyone feels in a professor mode, i'm all ears!

[g1]

The math for the plate dissipation is the easy part. The hard part is finding any kind of consensus on what the target % should actually be!
You said you know how to find the idle current (shunt method or with 1 ohm cathode resistors). For example we'll use the cathode resistor method. The 1 ohm resistor is in series with the cathode. All plate and screen current flow through the resistor. The screen current is small enough to just ignore. So for example you measure 20mVDC at idle across the 1 ohm resistor. I=E/R so 20mV divided by 1 ohm = 20mA. So you have 20mA idle current. You want to know the power, so you need to know the plate voltage, for example 460V. P=EI so 460V x 20mA equals 9.2 watts. That is your plate dissipation in watts. If you want to express it as a percentage, you need the max. plate power spec. for the particular tube. A modern 6L6 is rated at 30W max. The example is idling at 9.2W, 9.2 over 30 gives 30.67%. So the example 6l6 idling at 20mA with 460V at the plate is idling at about 31% plate dissipation.

[catstrat]

Thanks G-one. Easy Peesy. But i'll find a way to complicate it. LOL. It seems when we're dealing with bias, every one seems to have a different take on it and i've seen guys argue until the cows come home about it. I just find the way i like and what works for me and i go with it. The amp is ready for final test now with 1 ohm resistor in and we're going to find out shortly what it's doing.

[isaac]

I had one of these 135w UL twins, and I might add (really quickly...) mine sounded great, after i replaced a tube, and a couple other things... Anyway, when I got it, someone had already replaced the bias balance control with a regular bias supply, except with no adjustment. and the Hum balance had been gotten rid of for the regular 100 ohm resistors. But mine hummed like any normal amp, quietly sitting. No super loud humming that is annoying or anything. SO I don't have any experience with those balance controls, but mine was quiet without them. If it helps at all.

One thing I noticed is that you said that you tested all the tubes on your tester. I found that most guitar amps run tubes harder than a tester does, so that even though sometimes the tester says its ok, it could be bad in an amp. I tested a pair of EL84's once and they were fine, but I put them in a good amp, and they went nuts, weird noises, microphonics, the works. Same with a couple 12ax7's. etc... So "maybe" just maybe you have a bad tube, although it tested ok.

Oh well, hope my little mumbling might be able to help or at least inform a little bit.

[catstrat]

The humming twin is now fixed waiting to be picked up. As a last attempt to shut it up, i increased the value of the 3 20 mfd filter caps to 47 mfd. That seemed to have done the job. It looks like to me the amp just didn't have enough filtering in the first place. It has 2-220 350 volt and 3- 20 mfd cap plus the bias voltage filters. I guess as it got older something changed in the power supply which requires more filtering. I don't know but right now it's quiet as a mouse and sounds good to. Now i'm ready for some easy fixes.