Parasitics - always.

[heymymy]

Hey, it seems that 50% of the time I build a 5F6-A, I get crazy parasitics.  Can anybody send me some guts pictures of amps they've built so I can see what good lead dress is.  I don't understand this at all... it's like I'm cursed with these things.  Thanks guys!
I keep reading Gerald Weber's books over and over about how to stop them but it seems I'm just not getting it.

[Pleximan]

Crazy parasitics 50% of the time on a new build you say? 
Could be positive feedback

Try swap the OT wires around. The CT stays where it is and the plate wires on pin 3 are swapped and don't read GW books

Cheers

[Danskman]

what kind of "parasitics" do you talk about? Buzz, hum, scratch, boing?? I know of parasictics in some Tweed amps, like the Tweed Deluxe and the Bassman 59; it seems it can be due to the close proximity between the 68K input grid resistors and the plate voltage of the same pre-amp tube.
On my TD, turning the guitar volume pot generates scratches in the amp: in my Bman, i got some slight buzz when I touch the metal part of the axe.
The best corrective action is to mount the input grid resistors on the input jacks, or better, near the 12AY7 sockets.
HTH, BR and HY,
danskman

[heymymy]

Well, I'm guessing it's parasitic, I don't have a scope... ok, here's one issue... when I crank this one build it goes crazy when you turn the bass all the way up with the volumes jumped and you play a B on the low E string... just vibrates all over the place sounding like a blown speaker, but the speaker is fine, I pulled it out and checked it.  Also, the same amp, when I play sometimes there's this little static noise
when I play percussively, like the a slapback almost... but instead of echo it sounds like a loose solder joint, but the chopstick test reveals no loose joints or wires.  

[Leevi]

Some tips that might help:
- use shielded wires around the PI, the wire coming from the tone stack (treble pot) is critical
- try grid resistors in the power tubes
- try to ground the power tubes in different point
- use twisted wiring where possible
/Leevi

[JayB]

Well, I'm guessing it's parasitic, I don't have a scope... ok, here's one issue... when I crank this one build it goes crazy when you turn the bass all the way up with the volumes jumped and you play a B on the low E string... just vibrates all over the place sounding like a blown speaker, but the speaker is fine, I pulled it out and checked it.  Also, the same amp, when I play sometimes there's this little static noise
when I play percussively, like the a slapback almost... but instead of echo it sounds like a loose solder joint, but the chopstick test reveals no loose joints or wires.  

Tube maybe?  I built a combo amp that would go crazy on A, couldn't figure it out. The first preamp tube had a piece of glass inside it that would rattle like crazy on A.

[jojokeo]

There's a couple of things you're describing and thinking they are related?
"when I crank this one build it goes crazy when you turn the bass all the way up with the volumes jumped and you play a B on the low E string... just vibrates all over the place sounding like a blown speaker" - easily could be fret buzz or a cab/grill/tube rattle, or sometimes specific note issues can be attributed to a bad solder joint. I've also heard of improperly grounded output jacks and/or speaker ground causing this too.

"when I play sometimes there's this little static noise when I play percussively" - This may be a ground issue w/ your guitar and it's electronics, jack, or cable. If it doesn't do this on any other amp and only this one, then you could install a small .1/100v cap on your input jack's ground to chassis ground keeping the cap's wires as short as possible if it's an insulated type.

"it sounds like a loose solder joint, but the chopstick test reveals no loose joints or wires" - Chopsticking doesn't always reveal bad solder joints. Better to re-touch as many as you can immediately before messing w/ anything else. It only takes minutes to do and can only help, not hurt. You'd be surprised how easy it is for them to occur especially if you're not constantly cleaning your iron tip of the frass that accumulates and if any wires move while cooling you will get micro fractures in the joint and they can happen easily and be unseen. With repeated cycling of heat & cold, solder joints expand and contract more than you realize. Any weak joints wil become more problematic over time and Murphey's Law states that your amp will act up in the middle of a gig rather than while practicing in your home.

Lastly, parasitics are most always due do to lead dress and component layout issues in high gain amps unless it's just a pooly conceived desgin in the first place and don't confuse ground loops, cab rattles, cone rub, and tube microphonics w/ parsitic issues.

Good luck on fixing it.

[67polara]

I don't mean to sound stupid, but what the heck is a Parasitic?

Tony

[bigsbybender]

I don't mean to sound stupid, but what the heck is a Parasitic?

Tony

A shortened term for "Parasitic Oscillation". These are noises picked up from within the circuitry. It's often caused by wires crossing each other that shouldn't. One example might be a the large output of an anode (plate) lead crossing the small signal grid wire. Thus the strong signal from the plate can cause noise to induce in the grid.... This causes an oscillation (much like if you stick your the microphone from your PA in front of the main speakers).

Parasitic oscillations can happen in any amp, but amps with powerful and/or several gain stages certainly are the most prone to this problem.


j.

[67polara]

Thank you for the answer.  Now, I am not a stupid as I was LOL.

Tony

[heymymy]

Ok, so the plate and grid wires need to not cross.  which ones should be short as possible? 

also, I should shield the wire going to the treble control from the tone stack.  makes sense.

is that why when you turn the treble knob all the way up it hums louder than the other controls?

[jojokeo]

There are a number of things in Weber's books that are valid, like keeping grid wires as short as possible as they can act like an antenna & these should be shielded early in the circuit where they are more sensitive therefore vulnerable to RF also. Plate wires carry high voltages and high signal strength & are not as sensitive. Another thing he mentions is not to have the circuit cross back over itself. Things like this are good "rules of thumb" to follow.

"is that why when you turn the treble knob all the way up it hums louder than the other controls?"
No, you're hearing all of the available high frequencies that are in the signal along w/ other things like power supply, heater, and transformer humm as these are all being accentuated.

[Leevi]

The wire from the the treble pot is critical especially if it's placed close to the wires going out from the PI
and that's why it is better to shield. In this case there is probably not question about an antenna effect
because the signal level is high which means that the wire is not that sensitive.
The shielding prevent the crosstalk (I'm not sure about term) in general.
/Leevi

[heymymy]

ok, I understand, but how come old fender amps, like the super I have, is a wiring disaster as far as plates and grids laying on each other and all the same length... and that works fine(mostly).  crazy.  I always find horrible layout (according to what we're saying is good layout, which I agree with completely) in vintage amps. 

[TIMBO]

Hi Hey, I've done a couple of builds and have been lucky not to have these problems. I think you are going in the right direction by looking at other builds to see how to layout your build. Short runs are best and long runs are shielded.Transformers close and right angled to each other, with a short run from OT to power tubes with preamp tubes at the other end. My board builds are a bit different to the norm as you will see in the pictures but it seems to work. I have converted to using a turret board for the next build. The layout is as the drawing, which means positioning the components near to the tube connection ( this is a complicated build so i am expecting the worst) the tone stacks are on a PCB, with connecting wires to the main board. I building this one on a piece of ply to start with because i am expecting to make a lot of changes and have a lot of problems, who knows Hope this helps.

[TIMBO]

These are some pics of my LITTLE BOOGA project .Thanks

[phsyconoodler]

Quote:"- try grid resistors in the power tubes"

  The original 5F6A bassman did not have grid stoppers and they almost always need them.I have a done a couple of bassman builds and they both needed at least 1.5k grid stoppers.
  Also you need to follow the layout really closely and not deviate from the positioning of the wires and where they cross or don't cross.Other than that they are pretty flawless.
  I have an amp building friend who likes to do builds his own way and changes layouts to make them look neater.He has all kinds of troubles.Fender's layouts were time-proven to work.Hoffman's layouts work too but in my opinion they are more complicated.
  Simple is best in my book.

[Leevi]

Some practices I normally  use:

- place the first tube as close as possible to the inputs
- use twisted heater wires
- if you use resistors in the filament "ground" the resistors to a positive voltage (e.g. to cathode biased cathode)
- use for longer signal wirings shielded wires
- place the filter caps close to their target (plate)
- use bus grounding for the preamp (solder the bus to the pots)
- use star grounding for the power amp
- place the PT as far as possible from the amp circuit
- place the power tube grid stoppers as close as possible to the tube pins
- in general keep the wirings short
- pots are very important, use only new ones
- select the chassis according to the amp (1.5mm aluminium works)
- use screws instead of rivets

... but still sometimes I have to put lot of effort on troubleshooting because
of a mistake somewhere.

/Leevi

[JayB]

These are some pics of my LITTLE BOOGA project .Thanks

Is that copper traces I see under that board? That's pretty slick looking.

[TIMBO]

Hi JayB, I've found using the PCB copper clad board easier and its something i can get from the local electronic store.Valve amp building parts aren't easy to get in australia. I find using it keeps the connecting wires under the board and keeps the components free from clutter.It works for me. Thanks

[JayB]

Hi JayB, I've found using the PCB copper clad board easier and its something i can get from the local electronic store.Valve amp building parts aren't easy to get in australia. I find using it keeps the connecting wires under the board and keeps the components free from clutter.It works for me. Thanks

You must go through a lot of etch. I keep thinking about doing that myself. Currently, I just use some buss wire to do all the under board wiring. Mainly grounds and B+. I just line every thing up like a hoffman layout using eyelets then use turrets for everything else.

How do you do the layout of the underside?

[jojokeo]

How do you do the layout of the underside?

It's like making a small circuit board for an effect box or similar. All you have to do is draw out the layout like you normally would and instead of running a real wire on top of the board, you "draw" it to the underside like a negative for a picture if this makes sense to you?

What I do to keep things simple is draw out the layout in actual sized dimensions using actual parts if needed so you know things will fit perfectly. (I do this on every new build anyway whether using a turret or eyelet board). For making your traces and part end points, draw the traces a little darker going over all points/lines a couple of times. Do this so that you can see through the paper when looking at it by holding it up in the air or to a light source from underneath. Depending on how good you get - you can lay another sheet of paper over it and retrace the circuit but from the backside this time. Now you have a negative of your board. Or you can just use the original piece and poke a small hole into every part's end point as you would before drilling out a board for turret or eyelets. Lay the paper circuit down onto the board material face down (so that it's like a negative) and mark every hole w/ a sharpie exactly where you made the little holes in the paper. No look at your layout through the underside of your drawing by holding it up to a light and redraw your lines lightly w/ whatever like a pencil in case you make a mistake. Now you should have your circuit drawn on the copper board just like looking through the paper.

Next step: Get etching transfer dots & rub them onto every hole you marked. I use a chopstick or whatever for this. Then use either etching transfer lines, printed circuit tape or you can also use automotive pinstriping tape for the lines. I don't recommend using only a sharpie as it's not thick enough and your lines will be crappy (uneven in copper thickness and the edges too) when done etching.

Now all you have to do is clean the board and drill it out after the etching process. I like to use a small hole punch to keep the drill bit centered for all solder tab holes or you'll maybe be going through the sides of some of them which isn't good and looks shoddy too.

[heymymy]

TIMBO, - that's some super nice layout work... I think I've just been too messy, or maybe I rush a bit to finish when I really should take it slow and do it right the first time. 

For what it's worth, I just saw a trick where the wires on the underside of the board were just dabbed with a little hot glue to keep them in place, I thought that was genius, figured I'd mention it to you guys too if you hadn't already known about it.

[jojokeo]

ok, I understand, but how come old fender amps, like the super I have, is a wiring disaster as far as plates and grids laying on each other and all the same length... and that works fine(mostly).  crazy.  I always find horrible layout (according to what we're saying is good layout, which I agree with completely) in vintage amps.  

My own feelings on this are that "many" (but not all) of the older vintage designs you're thinking of aren't considered "high gain" designs. You can get away w/ much more on these w/out issues but in more modern high and very high gain amps, things can ramp up quickly on the sensitivity scale exponentially.

Another thing to consider is that even w/ Fender, his boards & layouts were done from a manufacturing point of view for production of many per day by semi-skilled women mostly and also a servicing point of view so that they could be easily repaired and/or replaced. There were compromises and things done then that you wouldn't necessarily do today unless you too were obsessed w/ production like he was. Time is money and he was very frugal in this sense in everything his company did. This isn't to say he sacrificed quality though but he knew & strived how to get the both.

[TIMBO]

Jojokeo, You are right again. I've looked at a lot of the old style turret/eyelet boards and noted that components were not in any kind of order and two or three wires were stuffed into one eyelet. This was done that way to save on turrets/eyelets i'm guessing. I do my boards in such a way that each component has its own holder and connection to the next component is done under board ( soldering the turret to the trace also holds it in place because the PCB is about 1.5mm thick) this is for ease of removing parts for tweaking (the turrets not only have a hole through the centre like normal but have a notch at the top as well, so rather than bending the leads into the hole i can just sit them in the grove and removal is easy. The board layout should be done in such a way that it keeps all wire runs as short and as direct as possible. Thanks

[jojokeo]

Whether or not I use a tagboard lug, eyelet hole or turret w/ one part or several - it seems that many times it comes down to size of the board, chassis, and design complexity get everything to fit right. I've re-used boards on many occasions so that I don't have to remake them if I don't have to. I'll fit whatever design I want into what I've got to work with. It's challenging and fun to me. Here's a few examples of gutting everything for something else.

[JayB]

How do you do the layout of the underside?

It's like making a small circuit board for an effect box or similar. All you have to do is draw out the layout like you normally would and instead of running a real wire on top of the board, you "draw" it to the underside like a negative for a picture if this makes sense to you?

What I do to keep things simple is draw out the layout in actual sized dimensions using actual parts if needed so you know things will fit perfectly. (I do this on every new build anyway whether using a turret or eyelet board). For making your traces and part end points, draw the traces a little darker going over all points/lines a couple of times. Do this so that you can see through the paper when looking at it by holding it up in the air or to a light source from underneath. Depending on how good you get - you can lay another sheet of paper over it and retrace the circuit but from the backside this time. Now you have a negative of your board. Or you can just use the original piece and poke a small hole into every part's end point as you would before drilling out a board for turret or eyelets. Lay the paper circuit down onto the board material face down (so that it's like a negative) and mark every hole w/ a sharpie exactly where you made the little holes in the paper. No look at your layout through the underside of your drawing by holding it up to a light and redraw your lines lightly w/ whatever like a pencil in case you make a mistake. Now you should have your circuit drawn on the copper board just like looking through the paper.

Next step: Get etching transfer dots & rub them onto every hole you marked. I use a chopstick or whatever for this. Then use either etching transfer lines, printed circuit tape or you can also use automotive pinstriping tape for the lines. I don't recommend using only a sharpie as it's not thick enough and your lines will be crappy (uneven in copper thickness and the edges too) when done etching.

Now all you have to do is clean the board and drill it out after the etching process. I like to use a small hole punch to keep the drill bit centered for all solder tab holes or you'll maybe be going through the sides of some of them which isn't good and looks shoddy too.

That's how I do my layouts now. I just print that out and tape it to the board, then center punch every hole. I would imagine circuit tape is better?

[jojokeo]

Yes the tape is much faster and better to use and w/ an exacto knife it's just "roll & go". Sorry about the long explanation of circuitboard making but maybe it can help some others?

[JayB]

Yes the tape is much faster and better to use and w/ an exacto knife it's just "roll & go". Sorry about the long explanation of circuitboard making but maybe it can help some others?

I'm sure it would, the more info the better. 

I always wondered how they did it in the old days. I use presensitized boards for relay boards and foot pedal effects but I can only get that in small sizes. Which works great for small projects like that.

[heymymy]

Hey, so I've found something I think is cool on how to find or track down issues... see what you guys think.  I read that you can short the grid of each preamp tube to ground to isolate which stage of the pre amp is problematic with a jumper wire.  Anybody use this method?

[sluckey]

I've used that method a lot. However, USE CAUTION! You must know what you are doing. The rule is, it's safe to ground the grid of a stage ONLY if there is normally zero volts dc on that grid. That covers a lot of preamp stages. But there are exceptions and if you break the rule you may break the amp. Two examples come to mind without even thinking hard.

1. Many popular amps use a cathode follower to drive the tone stack. Most times the grid of the CF is direct coupled to the plate of the previous stage. If you put a ground jumper on the grid of such a CF, you will probably burn up the plate load resistor of the previous stage.

2. Many circuits will use a bootstrapped bias circuit. The Fender Princeton Reverb split load PI is one example. The LTP PI circuit is another example. Both of these circuits operate the grid at a positive dc voltage level. You probably wont burn up anything by grounding the grids on either of these examples, but upsetting the bias on these type circuits will likely void the results of what you're trying to achieve.

Bottom line. It's a great tool but know what you are doing. If you use shotgun troubleshooting techniques with a ground clip, you may have to deal with collateral damage.