It's really starting to look like a tube tester. And works too! Part 4

[@PCBWay]

Comprehensive and inspirational sharing!💪

[@mamoynas]

Another easy feature you can install is the "Life test", that test simly reduces the heater voltage by 10% and if the transconductance stays within 20% of original value then that means that the tube's Cathode is not very aged and has many hours of "Life" ahead. (the percentages 10% and 20% are taken out of my Hickok tubetester manual). You could also add a small speaker through a cap at the anode to be able to hear if a tube is microphonic by tapping it

[@electronicsoldandnew]

👍

[@DavidTipton101]

That's a lot of work Manuel but you seem to be enjoying it 🙂

[@electronicsoldandnew]

Most fun I’ve had in a while … 😊

[@DerekHerbst747]

Nice work! I wouldn't have gone to as much detail but then again I'm a lazy sod. 😂

[@electronicsoldandnew]

We all are 😊

[@berndb.5097]

Hey Manuel, your tube tester looks really good. Perfectly adequate.👌 Thank you for your clever idea.👍👍

[@electronicsoldandnew]

My pleasure

[@19janiboy96]

This is a great series

[@electronicsoldandnew]

Thanks

[@dougietamson]

Very nice project. After adding an extra base socket or two, it would be a good idea to add small ferrite beads to the wires used for any pins that will connect as grid electrodes. Keep up the good work.

[@electronicsoldandnew]

Good idea. 👍

[@MrGeschwint]

I’m always inspired by the support instruments you build and use, also the repairs. Not a tube guy really, but I had to make a load to make amplifier repairs easier. So the passive load you made and use was an inspiration to my own. Best thing I ever built. Keep up the good work.

[@electronicsoldandnew]

Thanks

[@johnpettiford4861]

It is really quite good. I'll look forward to your further thought on the project.

[@electronicsoldandnew]

👍

[@xyredmax]

This is evolving nicely, I think quite a few of us will be using the final power supply board

[@electronicsoldandnew]

👍

[@martijnappeldoorn8686]

The tube socket (or base) used on el41, ecc40 and other ‘40’ series tubes is called ‘rimlock’ or B8A.

[@electronicsoldandnew]

That’s it. Couldn’t remember the term. Thanks.

[@t1d100]

Samtec makes great connectors and may have what you need to make your various tube programming plugs. Or, maybe DIY a PCB, in some manner.

[@radio-ged4626]

On the valve tester I have lined up for building I'm using opto-couplers for the meter battery supply rather than relays. To select the pinout assignment for the valves I'm using banana plugs and leads with the supply leads labelled as Plate, Grid, heaters etc and the pin numbers on the sockets of the banana plugs. Loving your project so far, it all seems quite well regulated and stable for reliable results.

[@electronicsoldandnew]

👍

[@mostlyoldthings]

Superb Manuel, the project is really coming together. An inspiration to us all! Thank you

[@electronicsoldandnew]

My pleasure

[@dl7majstefan753]

Very nice project dr Manuel! One hint for the mA-Meter: If you connect it in the ground path of B1+ (and not on the B+-side), you have a common negative ground with the B1+ and B2+ meters and need only one battery for these three meters (as i understand.....).

[@electronicsoldandnew]

Yes, that would work, but then I would be measuring the cathode current, which will have a screen current component as well. Not sure if that will be accurate enough for power tubes that have screen current. Am I thinking correctly?

[@dl7majstefan753]

@@electronicsoldandnew Yes, you´re thinking correct. I don´t know the difference of cathode and screen grid current, but does it really hurt? The screen resistors are always rather big so i assume only some mA of screen current. But in case of an internal fault inside the tube this may be decisive.

[@electronicsoldandnew]

I know how a tube should behave in a normal application (like power amp), and in that case, the screen current is about 5% of the anode current. However, in the Funke test setup, I’m not sure as there are no tube charts for these conditions. Do if I want to use their cards to determine the tube health, I’m a little stuck to their test conditions too. I’ll do some more thinking 🤔

[@dl7majstefan753]

@@electronicsoldandnew Yes, Manuel. It´s important whether you want to build just a "tube tester", where we can accept some percent of error (and we know how to estimate the error) or a real tube checker to measure the parameters. I´am very interested in this project

[@electronicsoldandnew]

👍

[@robnic52]

Brilliant project, great to see it progressing and your musings over the design. I was about to ask about short testing, usually the first stage before letting the magic smoke out of your tester but your parting comments addressed that. Also feeding in audio and RF signals and finding microphonic tubes? This is going beyond what most testers can do so this is a bit tongue in cheek. Loving this series.

[@electronicsoldandnew]

👍

[@retro_tech]

Really great work! Really looking forward to the rest of the project. About the sockets & connectors for tube selection: maybe an edge connector might work? You could then create a PCB with an edge connection per tube type.

[@electronicsoldandnew]

👍

[@6A8G]

Great build good sir! Possible to use an opto-isolator instead of the battery relays?

[@electronicsoldandnew]

I tried the opto-isolators and got really poor results.

[@ralphj4012]

Recoms (DC to DC converters) can be a pain on breadboards because of capacitance. You also need to draw some reasonable milliamps from the outputs so (for 1 mA meters) you would need something like an 820 Ohm parallel resistor across each converter output. Also, 1uF ceramics at the input and output and a small output smoothing cap (100uF ish). Yes, ripple also becomes an issue, especially at low currents, so you may end up with LC filters. Your battery solution is a good workaround. Don't be put off by those converters though, there is a Recom design guide, if you have a year free to study it. The best multipole virtually zero insertion / removal force connectors are Lemo, but you don't want to know how much they are.

[@electronicsoldandnew]

I did some of the research in that design guide, and it was indeed long 😊 I like these idolaters, and will continue to use them. Just sorry they did not work here without excessive circuitry to avoid their interaction with each other.

[@t1d100]

Excellent

[@electronicsoldandnew]

Thanks

[@BjornV78]

Looking good, almost done :-) For the selection of type of sockets, i don't know if there are large current draws, but maybe a 15 pin female VGA panel connector, and then with 15 pin male VGA connectors wired like a loopback plug making a "dongle" for each socket type. If you had needed only 8 wires, you could use Rj45 connectors, a female Rj45 panel connector, and for each socket type a male Rj45 connector as a "dongle". And if you need more then 15 pins, you can use a printer connector (LPT1), those are 25 pins.

[@electronicsoldandnew]

👍 exactly what I had in mind. Thanks

[@thehifiworkshop1452]

Nice work! I really like the fact that it has a digital readout. Some thoughts: For connecting other tubes, you could use some safety-banana plugs and sockets on top. For the heater supply, I use a bridge rectifier along with a LM 350. Really looking forward to build this one!

[@electronicsoldandnew]

👍

[@dxff5843]

Manuel, great series on creating a tube tester! This will be my next project! You inspired me already to built your dim bulb tester and the condensator tester. Please, could you specify the mosfets you used? Looking forward to your next video on the diy tube tester.

[@electronicsoldandnew]

I will have all that info for you when the birds arrive and I can test everything again properly. Don’t want smoke to spoil my viewers’ lives 😊

[@mackfisher4487]

Manuel, I think you're having altogether too much fun building the Madeira tube tester. Have you considered what you're doing to the used tube tester market? I can see the eBay prices dropping as you describe your concept. Kidding aside I like the use of pressboard for your Beta test must be that engineer thought process.

[@electronicsoldandnew]

It’s a nice change having prices fall due to my projects 😊 I did a restoration series in the ELV magazine on a Grundig 2147, and their ebay prices more than doubled. Fortunately I’d bought 4 of them 😊

[@mobilfone2234]

That is the problem, tube testers became unaffordable. I'm desperatly waiting for this project as this all I would need ;-)

[@electronicsoldandnew]

👍

[@mackfisher4487]

@@electronicsoldandnew Your mother didn't raise a fool, I know it's against copyright laws but if an English translation appeared in my mailbox I wouldn't object.

[@electronicsoldandnew]

I’m one issue away from completing the series, and hope to make an ebook of the 10 articles when they’re done. Should come to over 100 pages 😊 and I’ll gladly send you one when complete.

[@vasilification]

I wanted to comment on the switching needed to connect the filament, B1, B2 and grid to the correct pins. Since you are dealing with a small subset of tubes there may be a simple switching network you can use. A small 5x9 switching table for each tube of interest can be constructed. For example if the filament connections for all tubes of interest is 4 different connections, a 4 position 2 deck rotary switch could manage it. Some of those small relays you showed may be needed since may not be a good idea to route high current or high voltage through a rotary switch. It would need to be detailed out for B1, B2, G etc to see if this is a workable solution. Changing wires or headers my be too burdensome to make it a useful tool. Best Wishes

[@electronicsoldandnew]

👍

[@terencebarfield5864]

I am still wanting the valve tube tester schematic and details how to built your valve tube testers

[@electronicsoldandnew]

You don’t seem to read my replies. There have been two so far.

[@erikdenhouter]

Using 4 Schottky diodes for a bridge rectifier would give you an extra .6 V. You can buy 2x 1/2 W transformers the size of a box of matches for a few Euro. They have double windings so you can choose output: double voltage or double current. If you buy two, you will have 4 windings for 4 meters. E.g. Conrad partnumber: 1092849 - 62 2 x 6V 41mA € 5,07. Loaded with only a few mA that is more than enough.

[@electronicsoldandnew]

I misunderstood your previous comment. I thought you were referring to the heater supply. Sorry. Yes, I see your point about using small transformers for the meters. Good one 👍

[@erikdenhouter]

@@electronicsoldandnew Yes sorry for that, I edited my comment to many times to make any sense of it 😅 The first sentence is for the heater, the second for the meters.

[@electronicsoldandnew]

👍

[@yveslesage8525]

I feel happy that you gave up the idea of using resistors to lower the heater supply. I would have used TO220 LD1085 (rated 3A) or LD1084 (rated 5A) as Low Drop Out Regulators . Is there any mistake on the value of the Variable Resistor for adjusting the out voltage (given as 47k on your schematics) ? A 1k pot (+270 series resistor) would have the job done much more smoother.

[@electronicsoldandnew]

That is a mistake. I used a 2k2 pot. Apologies.

[@alancordwell9759]

I don't suppose the supply voltage for the meters is that critical; you could wind three separate windings on your transformer that would give 9 volts each when rectified and smoothed; it would only need 1N914s and a few microfarads if you only need a milliamp or two.

[@electronicsoldandnew]

I’ve thought of that but don’t have enough space on the transformer 😊 I was tempted to have a specific one wound, but that will make it more difficult to reproduce.

[@ebones6957]

Love it!

[@electronicsoldandnew]

👍

[@chumbuddy100]

Manuel, a D-Sub 15 should do the trick.

[@electronicsoldandnew]

That’s exactly it 😊 I did some testing last week to see if the plugging and unplugging would work well and it does. I have some in order already.

[@petercalum9767]

These FUNKE Cards, can they be downloaded ? Thanks, fine video

[@electronicsoldandnew]

I found a dull set on the web. Will try and find the link and post it when the project gets to its final stage.

[@peteselectronicprojects6522]

Look for - tuberocker funke

[@electronicsoldandnew]

That’s exactly where I count these 😊 thanks

[@petercalum9767]

Nice project. - Can Q1 survieve a short of B1+ ?

[@electronicsoldandnew]

It should as the 10 Ohm resistor will drop more than 0.6V and the current limiting transistor brings the voltage down.

[@clementeelectronicacomohob3205]

Hello good evening: Mr. M. Caldeira Sorry for my boldness but I want to build a valve tester and I am finding out about the Operation when it comes to co-testing the valves. Please, I would like to know the files that you left for the valve data cards. How do I know which card number each one is on? For example, the EL84 is on card 148. When searching for them, there is no index of them. . Thank you very much and I love your project, it is spectacular. Greetings.

[@electronicsoldandnew]

See email I sent you

[@mauryfeskanich4523]

When you mentioned using an AC signal to measure mu and S values, I wondered if you had considered adding a signal generator input, and oscilloscope (or other?) outputs to actually see the graph? Not sure if this would be practical, but I wanted to share the thought. My background is software, and I’ve always found data visualization to be very informative. Anyway, thanks for all your videos, I’m really enjoying them! Try not to have too much fun! 😄

[@electronicsoldandnew]

That idea is still in the mix, but not yet fully formed 😊

[@mauryfeskanich4523]

Not that I’m trying to influence you, but think of all the fun that would be! 😉

[@electronicsoldandnew]

😊

[@andylamont8111]

How about using a 15 pin “D” type serial connector and plug ?

[@electronicsoldandnew]

👍

[@sincerelyyours7538]

Very nice! Is it possible to measure the change in output current vs the change in input voltage directly (transconductance), or would you have to graph multiple readings in an X-Y plot to get an idea of what that is? The way your pots allowed you to change voltages suggested that might be possible, if I understood what you were demonstrating correctly. Looking forward to the final result!

[@electronicsoldandnew]

You can easily measure that by altering the grid voltage directly, but that is effectively a DC condition, which I’m not sure is going to be the same as injecting an AC voltage and measuring the altering current.

[@robtitheridge9708]

@@electronicsoldandnew you could use 50c/s from yore heater suply .

[@electronicsoldandnew]

The input is quite easy. I’m wondering how to read the output. I’ll probably need a small resistors to create the voltage drop that would be scaled and measured.

[@paulbennell3313]

Hey, it's coming along, you've been busy!

[@electronicsoldandnew]

Sure have 😊

[@holgers5216]

how about a voltage doubler on the DC side, so that should give you enough DC for the LM338?

[@electronicsoldandnew]

Voltage doubling is great, but ripple will shoot it with any sort of highish current. The previous comment by Erik about using a small transformer is better and quite cheap.

[@holgers5216]

@@electronicsoldandnew ripple should not be an issue if adequate filtering is used. I use a voltage doubler arrangement for DC filament supply, with a total current draw of around 900mA without any issues. I had it up to 1.2A, again no ripple issues, nice and steady and clean, it used for a tube preamp. Even if there was a small amount of ripple, for a tube tester setup such as this, it should not be an issue.

[@electronicsoldandnew]

True 👍

[@robtitheridge9708]

if you think yours looks like a rats nest you should see the inside of my taylor 45D admittedly all nicely harnest but an awfull lot of wire. interesting series cant wait to see the finished tester.

[@electronicsoldandnew]

I guess my wire harnessing skills will need refreshing 😊

[@BobAndersson]

Hi Manuel, Shoot me down in flames but a cheap and cheerful solution to your patch connector conundrum might use a number of 15 pin D-Sub connectors. Parts like Mouser No: 523-L717SDA15POL2 have a 600V rating so one for each valve (connection) type could be wired at the solder pin side of the connector, encapsulated in resin and then sutably labelled. Just select the D-Sub plug for the valve under test and plug it into a socket on the valve tester. Can you get away with just one socket on the tester for valve sockets with different pin counts? Would it be reliable enough after repeated mating/demating? I'm sure you can think of other questions as well. Maybe there's a more suitable plug/socket sytem out there that doesn't cost the earth? Just a thought...

[@electronicsoldandnew]

You are psychic 😊 What I have in mind is exactly something like that. Some years ago I was a partner in a software company, and the software licensing was activated by such a plug that you plugged into the series ports.

[@mikegLXIVMM]

10:24 Why not use a full-wave bridge instead of D14?

[@electronicsoldandnew]

Full wave will add another diode drop and make things worse.

[@KD0CAC]

How about power tubes for RF amps ? ;)

[@electronicsoldandnew]

I’m building this for the usual tubes that I find on the tube radios that I restore, but provided you know how to measure them, there’s nothing stopping you testing other tubes as well.

[@chrisclark6192]

Hi Manuel. This project is coming along nicely. I know you have made this tube tester for the tubes that you use frequently in the receivers that you specialise in. I was just wondering, is there any types of tubes that you can't test with this?. This series is turning out to be both very interesting and enjoyable to watch (as always). Cheers Chris.

[@electronicsoldandnew]

The power supply board is quite flexible as regards the voltages that you can get from it. The amount of current can also be increased by using a bigger transformer. The final hookups can be made to allow for other tubes as long as you know what characteristics to look for. When this is finished I hope to be able to advise on extending the range of tubes that can be tested.

[@radiotvrepair1059]

thank you very much for this tester can you give me the schematic diagram.

[@electronicsoldandnew]

It’s not finished yet. I’ll show all when the project is done.

[@radiotvrepair1059]

@@electronicsoldandnew ok thank you very much

[@heinz8770]

Hi Manuel ! Good job, but still a lot to do by switching the tubesockets. Why You do not use analogmeters ? There are very good looking and to find by e-bay, for about 14$ would be more matching for a tubetester :-) And You would not need all the batteries and relais, sorry, its only my mind ;-) I builed some testgear for repairing radios and so on and I used this meters and they look great an the set. And much easyer to connect. And sorry, maybe I am not mindfully listen Your video, but why You use DC-voltage for heaters ? In most radios and tv-sets the heatervoltage is AC and it is not so important if 6.0 or 6.5V especially for a short time testing. But still I wish You good success by this project ! By the way, I own a Funke W19 Tubetester and had to restore the most of the cards ;-)

[@electronicsoldandnew]

I chose these meters as a personal preference but any meter styles will do the job. The DC heaters were chosen as my mains voltage here tends to fluctuate quite a lot, and it was starting to annoy me 😊

[@heinz8770]

@@electronicsoldandnew Aha, ok, I understand ! So do the job well ! 🙂

[@electronicsoldandnew]

👍

[@danielsaturnino5715]

Posted 2 minutes ago and already it has 2 thumbs up!

[@electronicsoldandnew]

👍 Always nice 😊

[@radiotvrepair1059]

it's better to use analog meter and eliminate the batteries

[@electronicsoldandnew]

As I said, personal choice. Makes no difference as long as you’re comfortable with the functionality.