Steptoes amp.

[Paul Barker]

Yes Alex, been there done that worn the T shirt.

More than 60% of the 813's i ever bought second hand are low on dissipation. I suspect they are pulled from equipment in which they awere caned to within an inch of their life. You got the best of what I had, all I have left is the dross.

The one from Steptoe was nos but demonstrates the problems encountered shipping large valves.

Darren and I both experienced 1/3rd failures from shipping when we bought 212's and they were shipped to us.

Welcome to the club. Soul destroying I know.

[Alex Kitic]

Yes Alex, been there done that worn the T shirt.

...

The one from Steptoe was nos but demonstrates the problems encountered shipping large valves.

Darren and I both experienced 1/3rd failures from shipping when we bought 212's and they were shipped to us.

Welcome to the club. Soul destroying I know.

This means it's final? No cure...

I was hoping that the violet arcing in the base area was not due to some misalignment of elements, but gassing of a very old tube (70 years, after all).

What do you think of my idea to heat up the two troublesome tubes for an hour and check what happens with them?

[Paul Barker]

I would say a worn out tube is worn out.

You can read up on rejuvenation if you like, google it you'll eventually find the formula. But it won't last.

[Alex Kitic]

I would say a worn out tube is worn out.
.

I guess this means that there is nothing to loose and perhaps something to gain.

I will put the two troublesome NU tubes on "just heating" for 30 minutes, and than apply B+.

If the one that arcs stops arcing and starts working normally, this will be a momentous happening... and we shall know whether you can really degas an old NOS TT filament tube by doing so.

As for the other one, maybe it will "get a little bit better"... in that case it could be used as a pair for the NU which is OK.

[Steptoe]

Dear Alex, sorry to hear your news but those valves are younger than me and if I was ever asked to perform again I would most definitely arc violetly. I know there is a bundle more of the same vintage and I'll try to find a couple more for you. The only problem is my filing system, which is worse than non existent, but I'll do my best. With best wishes, Steptoe. PS. In return. could you explain to me how to measure the inductance of a transformer? I have a very good multimeter but no other gear.

[Mike H]

Yes Alex, been there done that worn the T shirt.

This means it's final? No cure...

I was hoping that the violet arcing in the base area was not due to some misalignment of elements, but gassing of a very old tube (70 years, after all).

Most of my 300B's went like this ~ new ones! Some had been bought with credit card and went gassy before I'd finished paying them off!

[Alex Kitic]

PS. In return. could you explain to me how to measure the inductance of a transformer? I have a very good multimeter but no other gear.

The easiest way to measure the inductance of a transformer or choke is using an LCR meter - special DMM that measures only resistance, capacitance, and inductance. Besides large table units like the one Paul has shown on some pictures, there are some small hand held ones, and they can be relatively inexpensive yet efficient. Some DMMs can measure inductance as well, and they are not necessarily expensive.

Otherwise, a circuit can be built to test the capacity to present a resistance to AC at a given frequency and current, enabling to calculate the inductance based on measurements. Still, I think that for most versions of the circuit you would need to have several other instruments. I guess an LCR meter, or DMM with L function (usually up to 20H), is a much better idea.

An LCR meter will be the first instrument I am going to buy as soon as I get a job.

[Alex Kitic]

PeakTech 3730

This is an example of a cheap DMM that can measure inductance. It was available in a local store, but although it was relatively cheap, I was not particularly motivated to buy one (money, money...).

3725, slightly more expensive

[Paul Barker]

i wrote a google doc for calculating inductance in DC conditions, for SE output transformers and chokes.

Any questions?

Make a Pi filter, measure AC at the first cap and then at the second cap put the figures in the spreadsheet. the cap you need to input in the spreadsheet is the second cap. The formula is interested in the inductor and the second cap. Do this at any current you may want to use the device at. Set current with a resistive load at last cap (ohms law tp work out value), or if you know how use a CCS.

The link.

[Alex Kitic]

The results: what happens after cooking old tubes...

1.
The 813 NU with violet arc problems: after one hour of cooking just on filaments, no difference. It did not stop arcing. This problem is definitely due to gasses present in the envelope. Now that I've done some thinking, I have experienced this with a 1619, it's the same buuzzz-bzzztt sound they make: only the 813 makes a louder sound, and the 1619 is metal so one cannot see the arcs inside.

I think there is nothing that can be done about it. Even if heating at higher temperature could absorb some gas traces, more air will probably get into the envelope sooner or later.

2.
The 813 drawing low current: no change. The reason it is not drawing the current set at the CCS is the bias or cathode potential being too low, the dropping resistor is 390 ohms and in the end there isn't enough voltage for the regulator to function properly. With a lower value dropping resistor, it would draw the set current, but the cathode potential would remain low compared to a good tube. Thus it remains a worst-case scenario spare.

The good thing here is that with this system the user does not need to know that his tubes are gradually exhausting their emission (assuming both channel used and gradually worn). The tubes are not going to suffer a current pull they cannot handle, rather it is the regulator that will gradually pull less and less current; eventually the user becomes aware the amp is not working as it once did (lower power), and after replacing the tubes everything is fine again... said user would only have to find new tubes.

[Mike H]

For the low current one, maybe it's cathode has been stripped?

[Steptoe]

Dear Paul and Alex, thanks for that advice. While you are surely right Alex, because I will probably only want to measure a small number of inductors, I can't really justify buying another meter. Mine will just about measure anything except L (another example of Sod's law) so I will try Paul's magic formula. It is, however, a very long time since I used any algebra and my brain, since then, has lost a fair number of neurones so is well past it's best before date and it will come as no surprise that I can't get the same result as you from your example,even when you give the answer.
I have included my calculations below and would be truly grateful if you could explain where I am going astray;
vout/vin= 0.07/5= 0.14

w2=(2x3.14x50 )2= 94200 so 0.14xw2= 1/LC-1=13608

13608L= 1/C.1= 1/49 so L=1/ 13608x49. Where did I go so hopelessly wrong?
Please help. Best wishes, Steptoe

[Paul Barker]

Dear Paul and Alex, thanks for that advice. While you are surely right Alex, because I will probably only want to measure a small number of inductors, I can't really justify buying another meter. Mine will just about measure anything except L (another example of Sod's law) so I will try Paul's magic formula. It is, however, a very long time since I used any algebra and my brain, since then, has lost a fair number of neurones so is well past it's best before date and it will come as no surprise that I can't get the same result as you from your example,even when you give the answer.
I have included my calculations below and would be truly grateful if you could explain where I am going astray;
vout/vin= 0.07/5= 0.14

w2=(2x3.14x50 )2= 94200 so 0.14xw2= 1/LC-1=13608

13608L= 1/C.1= 1/49 so L=1/ 13608x49. Where did I go so hopelessly wrong?
Please help. Best wishes, Steptoe

In electrical formulas C is always in whole farrads.

To find farrads devide microfarrads by 1,000,000.

Also 2pi.50 squared is 98,596 not 94,200.

And again .07/5 = 0.014.


I tell you what it would be a lot quicker to just put your numbers in my spreadsheet. That is what is supposed to happen.

Where the box is under input field you put the numbers for your situation in there.

Don't touch the block at the bottom it contains the whole formula and is calculated after you input all the information above it.

[Alex Kitic]

Indeed Paul was very precise in his datasheet, and his calculation recommendation.

I can only suggest doing calculation either in a spreadsheet or on a piece of paper, making eventual lapsus calami more evident. Care should be taken to get all the values right in the proper units, F for capacity, H for inductance, A for current, etc.

I guess Steptoe wants to measure the inductance of transformer primaries, thus it is necessary to use the transformer in question as a choke (primary only) between two caps like Paul suggests.

A method I use is statistical comparison. A choke of unknown characteristics can be measured in terms of DC resistance and EI lamination size and stack height, and compared to manufacturer data for chokes of the same size and similar DCR.

While size is a variable, DCR is a function of wire length and thickness. Two chokes the same size and DCR will most probably have the same inductance. The only element that can cause a difference is the gap, but we assume that both manufacturers were following the same guidelines.

This gets more difficult when trying to estimate the inductance of an output transformer, particularly if not-standard in size...

This is where the DMM with L capabilities comes in handy... and usually it can do something the other one cannot, like higher capacity, or better resolution, or frequency...

[Paul Barker]

There is a grey background in the results block, now. Thanks whoever improved it?