Footprint for a vacuum tube

Hi Experts,
For a retro project, I need to draw the footprint for a 14 pin vacuum tube. In KiCad footprint editor I found the “Footprint Wizard” for circular pad array.
Experimenting with the available “CPA_14” footprint in KiCad as a template, I found that though I can accommodate the required 14 pins, however, the maximum “center diameter” can only be set to 6.5 mm. This is not sufficient for the tube in question for the project.

I would need a center diameter = 16.65mm for proper alignment of all 14 pin pads along the perimeter of the circle.

My request, could someone point me to documentation explaining how to use this footprint wizard for circular pad arrays, please?

Thanks for your help in advance.
Best regards

1 Like

The pads can not be bigger then the outline.
So first enlarge the outline.

On the left you can select between “Pages” of parameters.

2 Likes

Hallo paulvdh,
That was a quick response indeed.
With your help, I could prepare the footprint of the 14 pad vacuum tube pretty quickly.
This wizard is really a very handy tool. Compliments to whoever took the trouble to create this tool for the community.
Thanks again and best regards

2 Likes

Perhaps you’ve downloaded or made a 3D model of a tube. If not, attached is a step model but, most likely it’s dimensions are not what you want.

It’s easy to change so, if you specify the dimensions you want, I’ll change them and re-post it…
Tube1.step (45.6 KB)

Current dim’s are:

O.D. = 16.65 mm
O.D. of Pin Pattern Circle = 2x6.79 = 13.58 mm
O.D. Pins = 0.80mm
Height = 65mm
Pin Length = 5mm


1 Like

Hi BlackCoffee,
I was dumbfounded with joy after reading your offer to create a 3D model for the tube. I am more than glad to accept your kind offer to create the 3D model.

The tube in question is ZM1020 (a Nixie tube for numerical digits display) and like most other tubes also a vintage component these days. The tube has been progressively displaced from the market by the 7-Segment LED and LCD units for 30- 35 years. The ZM1020 is no more in production - at least in central Europe.

Accordingly, it is very difficult to get the parts and even datasheet leave aside footprints and 3D models. After getting some the hint from paulvdh about correct use of Footprint wizard in KiCad, I was able to make the footprint but I don’t yet have any 3D model. I also couldn’t find any on the Internet - but to be frank - I was actually expecting that.

Therefore, if you could make one that would be just fantastic.
The only information I have about the dimensions of the tube can be taken from the image attached below.

The tube has 13 pins (10 Cathodes, 1 Anode, and 2 NCs). Since the data in the attached figures
doesn’t give any information about the angle between two pins from the centre of the circle, I created a footprint with 14 pins (the wizard placed them equidistant from each other on the circumference of the circle on which the pins are located) and I will neglect the pin 14 (located between pin 13 and 1).

The measurement unit in the figures is mm and the tube pins have a diameter of 1mm.
In case you need any additional information to prepare the 3D model please do let me know.

Thanks a lot for your efforts for creating a §D model for ZM1020 in advance.
Best regards

Will most likely be done tomorrow morning (US Pacific Time)… perhaps before

Good on you for dabbling with retro-tech. Glowing gas is great. But… strictly speaking a nixie is not a vacuum tube, as it’s gas filled.

Now you’ve encouraged me to look for something to do with my nixies…

1 Like

Hi kenyapcomau,

But… strictly speaking a nixie is not a vacuum tube, as it’s gas filled.

Yes, of course, you are absolutly right!!
Sorry for my carelessness.
Best regards

EDIT: Played with transparency - looks fine until exporting. Also flattened top and greyed per the drawing.

“strictly speaking”… 3D models are strictly for graphics. Thus, from my standpoint (and effort) no need for internal items…

This may not be what you want (let me know, I’ll change it tomorrow). I figured you’re not going to solder the tube and will probably have a Base/Socket for the Tube. So, I added a base and it’s pins are what go into the Thru-Holes on PCB. (tube and base are combined into one model)

Make your circular PAD pettern then, select #14 and delete it.

Files (Updated):
Tube2b.step (77.0 KB)
Tube2b.wrl (501.0 KB)

1 Like

Hi BlackCoffee,
Thank you so much for making the 3D model - as I feel - in record time indeed. I gave you the details around 12:30 a.m. CET and went to bed and when I got up in the morning the model was already there. I am delighted. You have done a great job. Thanks.

“strictly speaking”… 3D models are strictly for graphics. Thus, from my standpoint (and effort) no need for internal items…

Yes, I understand. The details about internals were just an attempt on my part to tell you how I deduced where the pins could be located on the circle’s circumference because the datasheet (or what I had) was vague about it.

This may not be what you want (let me know, I’ll change it tomorrow)

In a way, you are right. The anode and cathode as drawn by you inside the tube as plates are in fact ‘wire-formed’ numerical digits (0-9) in the case of Nixie tubes. So, yes, if you can replace the plate electrodes with digits (as shown in the datasheet figure), that would be just great. Then the model would really be representing a Nixie tube.

Of course, showing one or two digits would be more than sufficient.
I have no idea if that - what I am asking for- can be done easily or how much of your precious time it would consume but if you could do it, I am pretty sure that besides me many other community members would benefit from your work.

Played with transparency - looks fine until exporting. Also flattened top and greyed per the drawing.

Your model looks really fine. I really appreciate the efforts you spent on flattening and greying the top.

I figured you’re not going to solder the tube and will probably have a Base/Socket for the Tube.

Your assumption is correct. I will be using a socket for the tube. These tubes are too costly and too delicate to be soldered directly on the PCB. Adding base/socket to the model is a nice act of “forward-thinking” on your part. Many thanks.
I am not sure how many different socket types are available for this tube. The one I will be using looks as shown below:

But, you don’t have to change the model with respect to the base/socket. What you have made for the base/socket is good enough for me. If you could only replace the plate electrodes with digits, please, then that would be more than sufficient.

Thank you once again for your great help.
Best regards

Hi BlackCoffee,

Just wanted to give you a brief feedback that your 3D model for ZM1022 with base/socket matches perfectly with the footprint used in the pcbnew. The combined 3D-View of the PCB looks nice. Though, here the tube is looks almost opaque irrespective of the viewing angle. One can’t see the plate electrodes inside the tube.

Best regards

Re: the Anode/Cathode - numbers would be done via ‘Piping’ to look real. I’m too lazy (piping 3D is time consuming). I looked at several of these tubes and see some show numbers in vertical some horiz. I simply did a view from the top (in space) and left the Plates as was.

Re: the Base - added the attach pattern. I did not make the contacts as shown in your photo (wire blades).
If using the Base you show, you’ll need a large diameter hole in PCB for it. If intending to rest it on PCB with the blades going through holes, you’ll need to dial the hole diam’s in such that they don’t touch. That may be a problem - you’ll figure that out…
Also, the dimensions in your photo indicate Tube’s glass will interfere with mount screws (without hardware in-hand, I can’t assess it and probably wouldn’t bother for my own project). Other things I didn’t bother to do: Raise the tube on a perimeter-ring for screw-head clearance, Radius’ing the mounts top edges…

Again, the graphic with transparency will not show in Kicad/Step/Wrl so, ‘Solid’ is what you get (along with photo of transparency).

Files:
Tube2c.wrl (601.3 KB)
Tube2c.step (122.2 KB)

1 Like

I’ve been lurking in the shadows on this thread, but this mention prompted me to speak up. This specific footprint that you showed for the socket has what appear to be pins much larger (in one dimension) that 1mm. Unfortunately it appears that this socket is designed for wires (either soldered on or using quick disconnect lugs). I hope that elsewhere in the datasheet is a drawing giving the dimensions for the pins and the pin layout, but I would be surprised if the data sheet gives a suggested footprint.
Make sure that the footprint you design is sized for the socket that you will be using, not the actual tube.

Finishing up - my previous comments still apply.

I measured some Blades and below are the files… Hole, Mounting…etc … you’ll figure it out…

Tube3.step (300.5 KB)
Tube3.wrl (615.1 KB)

Before deleting the files (I don’t keep user files) I threw a Number on the top surface (it’s offset a very small amount - only way to do it without bleeding colors…

Tube3a.wrl (704.8 KB)

Hi BlackCoffee,
Thank you so much for spending so much of your time for preparing the different 3D-models.
You did intuitively the right thing by choosing the top-view for number display. The ZM1020 Nixie tube has top-view display indeed.

I fully agree with your all your comments about working with socket with wire blades instead of round pins. Therefore, I think I go for socket like you have implemented in your first 3D-model. It may take some time to find the right vendor but I think it is doable.

By implementing the three socket versions you have done much more than I was expecting. It is just wonderful. So, kindly accept my sincere thanks for all the work you did to help me. I appreciate it very much and by no means take such help for granted.

Best regards

1 Like

Yes, you are right. The socket diagram I sent to Blackcoffee requires its pin-blades either to be connected by wires or with quick disconnect lugs. This of course is a totally different assembly option that was not on my radar - but as have to accept, devil in the details.
Therefore, as I mentioned in my reply to BlackCoffee I will now have to look for a socket with round pins and not pin-blades. But I am optimistic.
Thanks for your constructive comment.
Best regards

1 Like