Can someone remind me why we don't use radial ecaps?

[Mike_J]

So I am reading Blencowes writing about grounding. Says what I have been hearing for ages. Ecap grounds should be close to the stage they serve. Of course my 5f6a has the old dog house where the grounds are a considerable distance from the stages they serve. Why don't we use radial ecaps for that purpose? Would have hell to pay to get a 220K resistor across their leads but otherwise would seem to fit well on a Hoffman board.

[acheld]

Nothing wrong with radial caps per se; there is no reason not to use them when they fit your layout scheme.

That said, if you were to use them in your doghouse, you'd have both your HT and your ground circuits on the same side of your board (I'm assuming you would lay them down on their side), and while this could be handled with care, what would be the point?

[thetragichero]

i'll often use smaller ones for a bias circuit but axial caps fit turret/eyelet layouts more securely (i fear that, when bending the leads 90 degrees there's too much stress on the leads than having the weight distributed over a longer area. not backed by science but i'm sticking with it)
radial caps are just fine for pcb builds

[Mike_J]

Nothing wrong with radial caps per se; there is no reason not to use them when they fit your layout scheme.

That said, if you were to use them in your doghouse, you'd have both your HT and your ground circuits on the same side of your board (I'm assuming you would lay them down on their side), and while this could be handled with care, what would be the point?

Wouldn't use them in the dog house. Would install them on the main circuit board close to the part of the circuit they would be serving. Should be less room for noise interaction ostensibly. Did a measurement of my 5f6a chassis and they would fit standing up.


Thank you for replying.

[Mike_J]

i'll often use smaller ones for a bias circuit but axial caps fit turret/eyelet layouts more securely (i fear that, when bending the leads 90 degrees there's too much stress on the leads than having the weight distributed over a longer area. not backed by science but i'm sticking with it)
radial caps are just fine for pcb builds

Like I mentioned in the last post the cap could be installed standing up as long as the chassis is deep enough. In the case of the '59 Bassman chassis it is. Should have given thought to it before now. Will go forward with the grounding plan I have using the ecaps that are in the dog house. Have other amps I will rebuild with a chassis deep enough to use them. Will try it then.


Thanks
Mike

[T Wilcox]

After doing a few builds with electrolytic and film radials I have come to hate them( radials) from a maintenance standpoint. In a proven design they would be perfectly fine but for a guy who likes to try out different component values and be able to take voltage reading throughout they have been quite the obstacle.
To swap a radial wima cap on a pcb you will need to access the opposite side of the board to desolder the component and unless you design a test point on your pcb forget getting voltage readings.
I stocked up on a bunch of Wima caps a few months back, now I cant wait to use these up and find a better axial solution for my prototype builds
As far as performance goes I cannot tell any difference and the radials are almost always less expensive

Todd

[drew]

Wouldn't use them in the dog house. Would install them on the main circuit board close to the part of the circuit they would be serving. Should be less room for noise interaction ostensibly. Did a measurement of my 5f6a chassis and they would fit standing up.

If you want to see layouts that creatively incorporate radial electrolytics, look up username xtian on ampgarage.com

[scstill]

I do builds with point to point wiring.
Radials are less expensive than axials
and they fit terminal strips better.
Here is a pic of my filter layout from a 5E3
Ground is the second from the left terminal

[2deaf]

Radials work fine in the doghouse.  Axials may get phased out, so you better have a contingency plan for the use of radials.

 

[2deaf]

Snap-in capacitors will also fit in a doghouse and the trend in available e-cap's seems to be headed that way.


 

[Mike_J]

Radials work fine in the doghouse.  Axials may get phased out, so you better have a contingency plan for the use of radials.

The dog house is staying and so will the caps in it. Only difference is there won't be any wires coming out of the dog house because nothing in it will be connected to anything. Will just be there to look good.


The radials will be on the circuit board. You can see the diameter of the ecaps in the picture below. The first two 100uF @ 350VDC caps would filter the first stage. The ground end of the ecap would go to the grounds for the power tube cathode resistors and then on to the PT bolt. The second pair of 47uF @ 350VDC ecaps will filter the screen supply. The screen supply would be grounded at the little 430K ohm resistor very close to it which was designed to be part of a voltage divider system to provide DC under the AC on the heater supply. Will remove the two resistors in the voltage divider for now and connect a wire to the ground under the 100VDC radial ecap for the voltage divider but won't connect it to ground. Will wrap it around a turret so it is easily available should I decide to use it. Will also run a wire from the positive terminal of the same ecap between the power tube sockets and along the back corner of the chassis. It will be long enough so I can install a humdinger pot, should it be deemed necessary, between the rectifier socket and the last power tube socket. Until the resistors are reinstalled and the ground wire connected to ground from the negative side of the little ecap it will be doing nothing.


The body of the caps are 1-1/4" high and I have 1-5/8" of room so pretty sure they would fit. Would install the screen caps a little closer to the edge of the board but for now was just testing it for feasibility. Pretty sure it is feasible. Need to remove some turrets and reinstall turrets where they would be needed for this system.


Reason for doing this is I have read for years it is best to keep the ground end of the ecap close to the grounding point in the circuit for the components it will be filtering. No way I can think of that I could get any closer.


Would still run an F&T 20uF @ 500VDC under the pots at the end of the board to filter the pi. Would connect the ground of that cap at the end of the buss wire where presence pot is grounded then connect another ground wire from that point to all the ground points for the bias supply then on to the PT bolt grounding point. With the input jacks isolated this would be the grounding point for V1,V2 and V3.

[Mike_J]

Would like to provide a little more space for these ecaps. Keep in mind the one ohm cathode resistors would be connected under the board as would the wire from the ground side of the two stage one ecaps. They would connect to the one ohm cathode resistor closest to the preamp. The one ohm resistor ground towards the end of the board would go to the PT ground and if I remember correctly connect at the same point as the PT CT. That would take care of stage one of the filter supply.


The 100K resistor that helps form the voltage divider won't be there initially. That turret will be sitting there all by its lonesome until I connect the ground side of the two screen stage resistors to it and then it heads on to the PT bolt. My bigger concern is the supply side (positive side of the ecap) for the screen resistors and the 430K resistor that would help form the voltage divider. These turrets would be in a much better place as far as placement of these ecaps are concerned if they were moved a little bit closer to the ground side. How much can I get away with?

[Mike_J]

Think it is time to do some reconfiguring. If things don't work out all the holes will be there to switch it back to the way the turrets are structured now. Have plenty of turrets. The rest would just be time.

[Mike_J]

Took wires off a few turrets and the solution to the spacing problem became clear as can be. The three wires in the attached picture are for the two screen resistors and the entry resistor into the voltage divider circuit for the heaters (maybe). The three turrets above them are for the one ohm cathode resistors (outside turrets) and the second resistor and ground wire for that little ecap (maybe) would connect to the middle turret.


Looking at the picture below everything from the first row from the bottom to the little ecap gets moved back one step. Will work with the plan above and make the space I need hopefully.

[Bieworm]

I always build my own turretboards.  That way it's easier to implement the turrets for the radial caps in an ideal arrangement. The radial caps I use only for the last 2 or 3 tubes right at the stages nodes. For most of my builds the 450V caps are just fine. And the rafial caps for that are 21mm high when placed vertically  do they even fit in a deluxe reverb chassis. The first filter caps are axial 500V types.

[thetragichero]

IF i were to use radial caps on a turret board, i would have them lying  down with some sort of adhesive between the body and the board (lot of folks use rtv silicone, i use gorilla hot glue. just 'flapping in the wind' like that seems like asking for em to break at the leads

[Mike_J]

IF i were to use radial caps on a turret board, i would have them lying  down with some sort of adhesive between the body and the board (lot of folks use rtv silicone, i use gorilla hot glue. just 'flapping in the wind' like that seems like asking for em to break at the leads

I could certainly do that. I have a gorilla two side foam tape that is unmerciful in how much it sticks. Would probably consider using that since I know it works and it is already paid for. Need more space for the laydown method though. The diameter of these are 18mm and height is 31.5 mm for the 100uF caps and the 47uF caps are same diameter but 25mm high. Suppose I better look to see if they have a single cap at 47uF and 22uF that would have enough voltage to do the job. The two caps in each location would still give me 700VDC which I like in the first two stages.

[Mike_J]

Parts are ordered. Will see what needs to be done when they come in. May need to stay vertical. If that is the case vertical terminal strips with the lugs removed could be used as enough support to keep the leads from breaking. A radial cap is likely to bend in only one direction. Pretty sure a vertical terminal strip would provide adequate support to keep it sound. Can possibly do a laydown and a standup for both stages. Don't see that it is likely there will be enough room for four laydown caps. We will see when they arrive and all components are placed in the chassis what needs to be done.


Thanks
Mike

[Mike_J]

IF i were to use radial caps on a turret board, i would have them lying  down with some sort of adhesive between the body and the board (lot of folks use rtv silicone, i use gorilla hot glue. just 'flapping in the wind' like that seems like asking for em to break at the leads

From the ounce of prevention is worth a pound of cure camp I think it may be advisable to support them with the vertical strip I discussed above. However they only weight .44 oz which can't be more than the .1uF orange drop 6ps caps I installed. The 6ps is also a radial cap although not an electrolytic. Lead thickness of the 6ps is pretty good whereas with the ecap thickness I do not know. Will compare when the ecaps arrive. The 6ps caps also have a stress relief bend built into them the ecaps don't have one as I see it. Direction of the cap would probably be a factor as well I would think.

[Mike_J]

Have a picture below of how the radial ecaps will fit in the amp. Will use a vertical strip to help support the two 100uF @ 350VDC caps because they are about 1-1/4" tall. The 47uF caps are only an inch tall. Probably will leave them alone unless the leads are looking very puny.


Do like the short runs to the associated grounds. Would be pretty hard to get them any shorter. Will make all the needed attachments, including the 220K-3W MO resistors across each cap next.


Thanks
Mike

[EL34]

It's very easy to design and have made custom boards with layouts for radials.

Dog house boards can be made with the spacings set up for radials

DIYLC can do all of that
https://el34world.com/Forum/index.php?board=26.0

[Mike_J]

It's very easy to design and have made custom boards with layouts for radials.

Dog house boards can be made with the spacings set up for radials

DIYLC can do all of that
https://el34world.com/Forum/index.php?board=26.0

I am looking very forward to using your services for my future boards. Seems like I am making a career out of the current rebuild I am on. Hopefully the things I am learning will transfer on to other projects. Going to spend the weekend reading Aiken and Valve Wizard and am actually looking forward to it.


Constructed the ten amps I have one after the other without as much knowledge or thought going into them as I can probably give them now. Probably will all need new boards.


Thanks
Mike

[Mike_J]

It's very easy to design and have made custom boards with layouts for radials.

Dog house boards can be made with the spacings set up for radials

DIYLC can do all of that
https://el34world.com/Forum/index.php?board=26.0

The reason I wanted to use radials is because the Valve Wizard said short distances for ground wires is best. Seems to be working for stages one and two. The LTP and preamp ecaps are going to be the F&Ts I bought from you and will be installed under the pots. The wires from the dog house to the circuit board are very long. Doubt that Leo would be using them today. So in my opinion a combination of radials ands axials is probably the answer.

[Mike_J]

Finally received the first and second stage ecaps. Put them in their places to see how the fit looks. Don't personally see anything I am over concerned about. Three more 22nF @ 500VDC caps are on their way. They are the same diameter just quite a bit shorter.