3D printed cassegrain reflector.


Colin G4EML
 

Following on from the recent discussions on a sub reflector for the 18” dishes I decided to see if I could 3D print one.

Based on Neil’s hyperbola figures I drew the curve and rotated it to form the solid.

I printed it in PLA with a layer size of 0.1 mm to get it as smooth as possible. Then sanded out the very slight ridges with wet and dry paper while spinning it in the lathe.

Then sprayed with Nickel screening paint to make it conductive.

That may have been enough but the Nickel paint had a resistance of about 4 ohms measured across the diameter. So I decided to try Copper plating it.

6 hours at 300mA in copper sulphate gave a good coating and reduced the resistance to as close to zero as my meter can measure. The finish is Matt but it does burnish up to shining copper.

I have not tested it yet but I am very happy with the result so far, and a happy few hours was had remembering my electroplating school science experiments.

Colin G4EML


Gedas
 

Hi Colin. I was wondering if you would not mind sharing your .STL file for the reflector? TU 73

Gedas, W8BYA EN70JT

Gallery at http://w8bya.com
Light travels faster than sound....
This is why some people appear bright until you hear them speak.
On 12/22/2021 6:24 PM, Colin G4EML wrote:

Following on from the recent discussions on a sub reflector for the 18” dishes I decided to see if I could 3D print one. 

Based on Neil’s hyperbola figures I drew the curve and rotated it to form the solid. 

I printed it in PLA with a layer size of 0.1 mm to get it as smooth as possible. Then sanded out the very slight ridges with wet and dry paper while spinning it in the lathe. 

Then sprayed with Nickel screening paint to make it conductive. 

That may have been enough but the Nickel paint had a resistance of about 4 ohms measured across the diameter. So I decided to try Copper plating it. 

6 hours at 300mA in copper sulphate gave a good coating and reduced the resistance to as close to zero as my meter can measure. The finish is Matt  but it does burnish up to shining copper. 

I have not tested it yet but I am very happy with the result so far, and a happy few hours was had remembering my electroplating school science experiments. 

Colin G4EML



 






Barry Chambers
 

Colin

what f/D have you chosen for your virtual dish please?

73 Barry G8AGN

On 22/12/2021 23:24, Colin G4EML wrote:
Following on from the recent discussions on a sub reflector for the 18” dishes I decided to see if I could 3D print one.

Based on Neil’s hyperbola figures I drew the curve and rotated it to form the solid.

I printed it in PLA with a layer size of 0.1 mm to get it as smooth as possible. Then sanded out the very slight ridges with wet and dry paper while spinning it in the lathe.

Then sprayed with Nickel screening paint to make it conductive.

That may have been enough but the Nickel paint had a resistance of about 4 ohms measured across the diameter. So I decided to try Copper plating it.

6 hours at 300mA in copper sulphate gave a good coating and reduced the resistance to as close to zero as my meter can measure. The finish is Matt but it does burnish up to shining copper.

I have not tested it yet but I am very happy with the result so far, and a happy few hours was had remembering my electroplating school science experiments.

Colin G4EML






--
73
Barry, G8AGN


Colin G4EML
 

Hi Barry,

I can’t answer that as I didn’t design it and don’t fully understand it. I just used the figures from Neil’s post on the 30th November.

Colin.

On 23 Dec 2021, at 09:53, Barry Chambers <b.chambers@sheffield.ac.uk> wrote:

Colin

what f/D have you chosen for your virtual dish please?

73 Barry G8AGN


On 22/12/2021 23:24, Colin G4EML wrote:
Following on from the recent discussions on a sub reflector for the 18” dishes I decided to see if I could 3D print one.

Based on Neil’s hyperbola figures I drew the curve and rotated it to form the solid.

I printed it in PLA with a layer size of 0.1 mm to get it as smooth as possible. Then sanded out the very slight ridges with wet and dry paper while spinning it in the lathe.

Then sprayed with Nickel screening paint to make it conductive.

That may have been enough but the Nickel paint had a resistance of about 4 ohms measured across the diameter. So I decided to try Copper plating it.

6 hours at 300mA in copper sulphate gave a good coating and reduced the resistance to as close to zero as my meter can measure. The finish is Matt but it does burnish up to shining copper.

I have not tested it yet but I am very happy with the result so far, and a happy few hours was had remembering my electroplating school science experiments.

Colin G4EML






--
73
Barry, G8AGN


Andy G4JNT
 

Interested in your copper plating plastic.  What & where the nickel spray?



On Thu, 23 Dec 2021 at 10:45, Colin G4EML <colin@...> wrote:
Hi Barry,

I can’t answer that as I didn’t design it and don’t fully understand it. I just used the figures from Neil’s post on the 30th November.

Colin.



On 23 Dec 2021, at 09:53, Barry Chambers <b.chambers@...> wrote:

Colin

what f/D have you chosen for your virtual dish please?

73 Barry G8AGN


> On 22/12/2021 23:24, Colin G4EML wrote:
> Following on from the recent discussions on a sub reflector for the 18” dishes I decided to see if I could 3D print one.
>
> Based on Neil’s hyperbola figures I drew the curve and rotated it to form the solid.
>
> I printed it in PLA with a layer size of 0.1 mm to get it as smooth as possible. Then sanded out the very slight ridges with wet and dry paper while spinning it in the lathe.
>
> Then sprayed with Nickel screening paint to make it conductive.
>
> That may have been enough but the Nickel paint had a resistance of about 4 ohms measured across the diameter. So I decided to try Copper plating it.
>
> 6 hours at 300mA in copper sulphate gave a good coating and reduced the resistance to as close to zero as my meter can measure. The finish is Matt  but it does burnish up to shining copper.
>
> I have not tested it yet but I am very happy with the result so far, and a happy few hours was had remembering my electroplating school science experiments.
>
> Colin G4EML
>
>
>

>
>
>
--
73
Barry, G8AGN













Colin G4EML
 

Hi Andy, 

It is nickel coating designed for screening plastic enclosures. Farnell part number 2811914.  Unfortunately it is quite expensive but it produces a very conductive surface of around 0.5 ohms per square. 
I have heard people have had success with the cheaper graphite paints.  

Colin. G4EML.


On 23 Dec 2021, at 10:48, Andy G4JNT <andy.g4jnt@...> wrote:

Interested in your copper plating plastic.  What & where the nickel spray?



On Thu, 23 Dec 2021 at 10:45, Colin G4EML <colin@...> wrote:
Hi Barry,

I can’t answer that as I didn’t design it and don’t fully understand it. I just used the figures from Neil’s post on the 30th November.

Colin.



On 23 Dec 2021, at 09:53, Barry Chambers <b.chambers@...> wrote:

Colin

what f/D have you chosen for your virtual dish please?

73 Barry G8AGN


> On 22/12/2021 23:24, Colin G4EML wrote:
> Following on from the recent discussions on a sub reflector for the 18” dishes I decided to see if I could 3D print one.
>
> Based on Neil’s hyperbola figures I drew the curve and rotated it to form the solid.
>
> I printed it in PLA with a layer size of 0.1 mm to get it as smooth as possible. Then sanded out the very slight ridges with wet and dry paper while spinning it in the lathe.
>
> Then sprayed with Nickel screening paint to make it conductive.
>
> That may have been enough but the Nickel paint had a resistance of about 4 ohms measured across the diameter. So I decided to try Copper plating it.
>
> 6 hours at 300mA in copper sulphate gave a good coating and reduced the resistance to as close to zero as my meter can measure. The finish is Matt  but it does burnish up to shining copper.
>
> I have not tested it yet but I am very happy with the result so far, and a happy few hours was had remembering my electroplating school science experiments.
>
> Colin G4EML
>
>
>

>
>
>
--
73
Barry, G8AGN













Neil Smith G4DBN
 
Edited

Plating seems like a great solution if the initial coat adheres well to the plastic and the surface finish is better than lambda/10 or so. Others have tried using burnished aluminium foil and had decent results, but the foil might need replacement now and then.
 
The hyperbola I modelled was based on having the feed phase centre pretty much at the face of the dish, and used a 48 mm diameter subreflector, with a space attenuation at the dish edge of around 3.7 dB, so an edge taper at the subreflector of around 8 dB to get the dish illumination to about 12 dB taper at the edge.  Michel F5DZK is working on a non-hyperbolic subreflector curve to see if the gain can be optimised by changing the taper over the dish surface.
To get that illumination of the subreflector needs a horn with a gain around 20dB, using a Pickett-Potter dual-mode horn with an aperture around 4.5 wavelengths, but it needs some testing with a range of gains to get it optimised.  Paul W1GHZ mentioned that he might need to modify the space attenuation expression in the spreadsheet, as it is showing 6.02 dB space attenuation, which would be right if the feed was at the focus of the parabola, but the distance from feed to sub to dish centre is around 210 mm, but feed to sub to dish edge is 315 mm or so and I think that leads to more like 3.7dB space attenuation at the dish edge.  The equivalent f/d is about 1.7 according to the sheet with the hyperbola having an inter-focal length of 111 mm, apex distance to parabola focus about 11 mm, diameter 48 mm, a=44.64 and b=32.82.
 
Looking at the modelled pattern of a Pickett-Potter and matching the pattern to get -8dB at an angle of arctan(24/111) = +-12.2 degrees, something around 4 wavelengths aperture seemed about right, but for experimentation, I think a range of horns of different dimensions would make sense to get the best compromise feedhorn gain.
 
It's perfectly feasible to use a section of guide with a much wider feedhorn closer to a subreflector, I just wanted to see how it would look with the PCB just behind the dish and the feedhorn poking through the dish face a little.  Others have used 2mm ID/4mm OD tubing to extend from the coupler body to a feed much closer to a subreflector with a much higher hyperbolic eccentricity.
 
The big issue here is the accuracy of the Edmund dish surface, which is probably not up to spec for 122 GHz anyway, so a fully optimised Cassegrain will perhaps fail to deliver the 55dBi or so of gain that it ought to provide. My plans for next year are to try making a cast aluminium parabola with a very precisely machined surface, to see if that will work more efficiently than using spun dishes, but I'm limited to about 13 inches diameter on my machines.
 
I wonder if a resin 3D printer might produce a more precise surface, but whether the nickel paint would adhere to the resin, I have no idea.  I have some PETG filament as well as PLA, so I might try to emulate Colin's excellent work with electroforming of a reflector once I get this Prusa fully tuned up.
 
Another possible approach might be to go full-on electroforming, and make a negative of the hyperbolic form, plate it thickly, then melt away the PLA, leaving a copper shell which could be glued to a mount. Only problem would be the initial layer of nickel.  Perhaps carbon colloid or something would work as a removable substrate?  Sounds like a pile of fun to be had anyway.


Barry Chambers
 

Neil, Colin

another route for the subreflector is to start with the spherical profile provided by the bottom of an aerosol spray can and to copper plate this but using an appropriate electrode shape so as to build up the copper layer preferentially at the centre to achieve the hyperbolic curve. I haven't done the maths but as a first try, the electrode could be a flat circular plate of smaller diameter than the aerosol can bottom and the variation with radius of the spacing between the sphere and the flat plate will give differential plating.

This process could be done in reverse by repeated dipping the end of a piece of copper bar into a plating bath (operating in reverse) and using an immersed ring electrode.

73 Barry, G8AGN

On 23/12/2021 11:13, Neil Smith G4DBN wrote:
Plating seems like a great solution if the initial coat adheres well to the plastic and the surface finish is better than lambda/10 or so. Others have tried using burnished aluminium foil and had decent results, but the foil might need replacement now and then.

The hyperbola I modelled was based on having the feed phase centre pretty much at the face of the dish, and used a 48 mm diameter subreflector, with a space attenuation at the dish edge of around 3.7 dB, so an edge taper at the subreflector of around 8 dB to get the dish illumination to about 12 dB taper at the edge.  Michel F5DZK is working on a non-hyperbolic subreflector curve to see if the gain can be optimised by changing the taper over the dish surface.
To get that illumination of the subreflector needs a horn with a gain around 20dB, using a Pickett-Potter dual-mode horn with an aperture around 4.5 wavelengths, but it needs some testing with a range of gains to get it optimised.  Paul W1GHZ mentioned that he might need to modify the space attenuation expression in the spreadsheet, as it is showing 6.02 dB space attenuation, which would be right if the feed was at the focus of the parabola, but the distance from feed to sub to dish centre is around 210 mm, but feed to sub to dish edge is 315 mm or so and I think that leads to more like 3.7dB space attenuation at the dish edge.  The space attenuation should be  and.but the equivalent f/d is about 1.7 according to the sheet with the hyperbola having an inter-focal length of 111 mm, apex distance to parabola focus about 11 mm, diameter 48 mm, a=44.64 and b=32.82.

Looking at the modelled pattern of a Pickett-Potter and matching the pattern to get -8dB at an angle of arctan(24/111) = +-12.2 degrees, something around 4 wavelengths aperture seemed about right, but for experimentation, I think a range of horns of different dimensions would make sense to get the best compromise feedhorn gain.

It's perfectly feasible to use a section of guide with a much wider feedhorn closer to a subreflector, I just wanted to see how it would look with the PCB just behind the dish and the feedhorn poking through the dish face a little.  Others have used 2mm ID/4mm OD tubing to extend from the coupler body to a feed much closer to a subreflector with a much higher hyperbolic eccentricity.

The big issue here is the accuracy of the Edmund dish surface, which is probably not up to spec for 122 GHz anyway, so a fully optimised Cassegrain will perhaps fail to deliver the 55dBi or so of gain that it ought to provide. My plans for next year are to try making a cast aluminium parabola with a very precisely machined surface, to see if that will work more efficiently than using spun dishes, but I'm limited to about 13 inches diameter on my machines.

I wonder is a resin 3D printer might produce a more precise surface, but whether the nickel paint would adhere to the resin, I have no idea.  I have some PETG filament as well as PLA, so I might try to emulate Colin's excellent work with electroforming of a reflector once I get this Prusa fully tuned up.

Another possible approach might be to go full-on electroforming, and make a negative of the hyperbolic form, plate it thickly, then melt away the PLA, leaving a copper shell which could be glued to a mount. Only problem would be the initial layer of nickel.  Perhaps carbon colloid or something would work as a removable substrate?  Sounds like a pile of fun to be had anyway.

-- 
73
Barry, G8AGN


Colin G4EML
 

The surface finish is an extremely fine crystalline structure. It is difficult to estimate the size of the crystals but I am pretty sure it is much less than lambda/10. The photo I posted makes the surface look rougher than it actually is. If necessary it could probably be burnished to provide a smoother surface. It seems to be very well adhered to the plastic, on one of my test pieces I tried to remove it and it took quite a bit of effort with a scalpel.

I did do an experiment plating at a much higher current which increased the rate of plating but that resulted in much larger crystals and a rougher surface. 

A resin printer might produce a better finish but it is still built up in layers so there will still be some variations. Wet sanding the PLA produced a very good surface for painting and removed the layer lines. Maybe a filling primer would help with that. Obviously sanding will change the shape but I was careful not to remove too much. 

I was wondering if the Nickel paint on its own would make a usable reflector. When I get the dish mount finished I will set up a test range and do some comparisons. 


Colin G4EML







On 23 Dec 2021, at 11:13, Neil Smith G4DBN <neil@...> wrote:

[Edited Message Follows]
[Reason: oops, cut/paste error fixed]

Plating seems like a great solution if the initial coat adheres well to the plastic and the surface finish is better than lambda/10 or so. Others have tried using burnished aluminium foil and had decent results, but the foil might need replacement now and then.
 
The hyperbola I modelled was based on having the feed phase centre pretty much at the face of the dish, and used a 48 mm diameter subreflector, with a space attenuation at the dish edge of around 3.7 dB, so an edge taper at the subreflector of around 8 dB to get the dish illumination to about 12 dB taper at the edge.  Michel F5DZK is working on a non-hyperbolic subreflector curve to see if the gain can be optimised by changing the taper over the dish surface.
To get that illumination of the subreflector needs a horn with a gain around 20dB, using a Pickett-Potter dual-mode horn with an aperture around 4.5 wavelengths, but it needs some testing with a range of gains to get it optimised.  Paul W1GHZ mentioned that he might need to modify the space attenuation expression in the spreadsheet, as it is showing 6.02 dB space attenuation, which would be right if the feed was at the focus of the parabola, but the distance from feed to sub to dish centre is around 210 mm, but feed to sub to dish edge is 315 mm or so and I think that leads to more like 3.7dB space attenuation at the dish edge.  The equivalent f/d is about 1.7 according to the sheet with the hyperbola having an inter-focal length of 111 mm, apex distance to parabola focus about 11 mm, diameter 48 mm, a=44.64 and b=32.82.
 
Looking at the modelled pattern of a Pickett-Potter and matching the pattern to get -8dB at an angle of arctan(24/111) = +-12.2 degrees, something around 4 wavelengths aperture seemed about right, but for experimentation, I think a range of horns of different dimensions would make sense to get the best compromise feedhorn gain.
 
It's perfectly feasible to use a section of guide with a much wider feedhorn closer to a subreflector, I just wanted to see how it would look with the PCB just behind the dish and the feedhorn poking through the dish face a little.  Others have used 2mm ID/4mm OD tubing to extend from the coupler body to a feed much closer to a subreflector with a much higher hyperbolic eccentricity.
 
The big issue here is the accuracy of the Edmund dish surface, which is probably not up to spec for 122 GHz anyway, so a fully optimised Cassegrain will perhaps fail to deliver the 55dBi or so of gain that it ought to provide. My plans for next year are to try making a cast aluminium parabola with a very precisely machined surface, to see if that will work more efficiently than using spun dishes, but I'm limited to about 13 inches diameter on my machines.
 
I wonder if a resin 3D printer might produce a more precise surface, but whether the nickel paint would adhere to the resin, I have no idea.  I have some PETG filament as well as PLA, so I might try to emulate Colin's excellent work with electroforming of a reflector once I get this Prusa fully tuned up.
 
Another possible approach might be to go full-on electroforming, and make a negative of the hyperbolic form, plate it thickly, then melt away the PLA, leaving a copper shell which could be glued to a mount. Only problem would be the initial layer of nickel.  Perhaps carbon colloid or something would work as a removable substrate?  Sounds like a pile of fun to be had anyway.


Neil Smith G4DBN
 

That sounds like even MORE fun to be had.  I did start wondering idly about a truly sparkly bling subreflector made by machining and polishing a negative of the hyperbola in aluminium, then using the anaerobic zincate/gold/nickel/alkaline copper plating process to electroform a positive.  A quick boil in caustic soda to dissolve the aluminium mandrel, and I'd be left with a gold-plated subreflector that would look *awfully* nice.

Gold Solutions Plating https://gsplating.co.uk/ do some fancy solutions and charge controllers that are supposed to offer a mirror finish even on plastics. I was looking at much smaller parts though, using electroforming on an aluminium positive mandrel of corrugated horns, with the zincate+gold+copper process. Their electroforming kits use a graphite colloid paint as the base layer on the plastic, but I don't think they sell it separately. The old way was to use a well-thinned cellulose laquer to paint the object, dust it with fine bronze powder while it was still wet, then plate using acidic copper solution. I did see someone trying zinc cold galvanising paint (Galvafroid or Rust-Anode) with fine copper dust added, but there are some tricky issues with copper over zinc.

I wonder how many non-radio folks imagine that these are the sort of things we get up to in the pursuit of the hobby?

Neil G4DBN

On 23/12/2021 11:50, Barry Chambers wrote:

Neil, Colin

another route for the subreflector is to start with the spherical profile provided by the bottom of an aerosol spray can and to copper plate this but using an appropriate electrode shape so as to build up the copper layer preferentially at the centre to achieve the hyperbolic curve. I haven't done the maths but as a first try, the electrode could be a flat circular plate of smaller diameter than the aerosol can bottom and the variation with radius of the spacing between the sphere and the flat plate will give differential plating.

This process could be done in reverse by repeated dipping the end of a piece of copper bar into a plating bath (operating in reverse) and using an immersed ring electrode.

73 Barry, G8AGN


Paul G8AQA
 

Even more fun would be making a good strong negative in steel and then explosive forming a sheet of aluminium.  Assuming one survived the experience one could make another using the same mould.

Paul G8AQA

On 23/12/2021 12:34, Neil Smith G4DBN wrote:
That sounds like even MORE fun to be had.  I did start wondering idly about a truly sparkly bling subreflector made by machining and polishing a negative of the hyperbola in aluminium, then using the anaerobic zincate/gold/nickel/alkaline copper plating process to electroform a positive.  A quick boil in caustic soda to dissolve the aluminium mandrel, and I'd be left with a gold-plated subreflector that would look *awfully* nice.

Gold Solutions Plating https://gsplating.co.uk/ do some fancy solutions and charge controllers that are supposed to offer a mirror finish even on plastics. I was looking at much smaller parts though, using electroforming on an aluminium positive mandrel of corrugated horns, with the zincate+gold+copper process. Their electroforming kits use a graphite colloid paint as the base layer on the plastic, but I don't think they sell it separately. The old way was to use a well-thinned cellulose laquer to paint the object, dust it with fine bronze powder while it was still wet, then plate using acidic copper solution. I did see someone trying zinc cold galvanising paint (Galvafroid or Rust-Anode) with fine copper dust added, but there are some tricky issues with copper over zinc.

I wonder how many non-radio folks imagine that these are the sort of things we get up to in the pursuit of the hobby?

Neil G4DBN



On 23/12/2021 11:50, Barry Chambers wrote:

Neil, Colin

another route for the subreflector is to start with the spherical profile provided by the bottom of an aerosol spray can and to copper plate this but using an appropriate electrode shape so as to build up the copper layer preferentially at the centre to achieve the hyperbolic curve. I haven't done the maths but as a first try, the electrode could be a flat circular plate of smaller diameter than the aerosol can bottom and the variation with radius of the spacing between the sphere and the flat plate will give differential plating.

This process could be done in reverse by repeated dipping the end of a piece of copper bar into a plating bath (operating in reverse) and using an immersed ring electrode.

73 Barry, G8AGN









Virus-free. www.avg.com


cricri
 

see that : ( the silver paint seems interesting .. no ?? )


Xtian / F1VL

Le 23/12/2021 à 13:34, Neil Smith G4DBN a écrit :
That sounds like even MORE fun to be had.  I did start wondering idly about a truly sparkly bling subreflector made by machining and polishing a negative of the hyperbola in aluminium, then using the anaerobic zincate/gold/nickel/alkaline copper plating process to electroform a positive.  A quick boil in caustic soda to dissolve the aluminium mandrel, and I'd be left with a gold-plated subreflector that would look *awfully* nice.

Gold Solutions Plating https://gsplating.co.uk/ do some fancy solutions and charge controllers that are supposed to offer a mirror finish even on plastics. I was looking at much smaller parts though, using electroforming on an aluminium positive mandrel of corrugated horns, with the zincate+gold+copper process. Their electroforming kits use a graphite colloid paint as the base layer on the plastic, but I don't think they sell it separately. The old way was to use a well-thinned cellulose laquer to paint the object, dust it with fine bronze powder while it was still wet, then plate using acidic copper solution. I did see someone trying zinc cold galvanising paint (Galvafroid or Rust-Anode) with fine copper dust added, but there are some tricky issues with copper over zinc.

I wonder how many non-radio folks imagine that these are the sort of things we get up to in the pursuit of the hobby?

Neil G4DBN



On 23/12/2021 11:50, Barry Chambers wrote:

Neil, Colin

another route for the subreflector is to start with the spherical profile provided by the bottom of an aerosol spray can and to copper plate this but using an appropriate electrode shape so as to build up the copper layer preferentially at the centre to achieve the hyperbolic curve. I haven't done the maths but as a first try, the electrode could be a flat circular plate of smaller diameter than the aerosol can bottom and the variation with radius of the spacing between the sphere and the flat plate will give differential plating.

This process could be done in reverse by repeated dipping the end of a piece of copper bar into a plating bath (operating in reverse) and using an immersed ring electrode.

73 Barry, G8AGN









Neil Smith G4DBN
 

On 23/12/2021 13:59, cricri wrote:
see that : ( the silver paint seems interesting .. no ?? )
Slightly expensive at £252.55 for the aerosol from RS online, but you could coat DOZENS of subreflectors.

The smaller jar 842UR-12ML is £75. https://docs.rs-online.com/696e/A700000006482818.pdf

The "What to do if you swallow this stuff" instructions are great fun.  Especially section 4.3 and section 11.  Sounds tasty. Perhaps not to be used in food preparation areas and definitely not as a sandwich filling.

Neil G4DBN


Neil Douglas
 

All,

 

Paul, there is a process called electro hydraulic forming which discharges an arc ( with a bang) from charged HV capacitors into a liquid on top of the sheet on metal on top of a die.

 

Regular hydraulic forming is possible using a die and plate with the hydraulic pressure applied to the top side of the plate using one of the hand pumps used for bearing pullers.

 

NeilD

G4SHJ

 

 

From: UKMicrowaves@groups.io <UKMicrowaves@groups.io> On Behalf Of Paul G8AQA via groups.io
Sent: 23 December 2021 12:42
To: UKMicrowaves@groups.io
Subject: Re: [UKMicrowaves] 3D printed cassegrain reflector.

 

Even more fun would be making a good strong negative in steel and then explosive forming a sheet of aluminium.  Assuming one survived the experience one could make another using the same mould.

Paul G8AQA

On 23/12/2021 12:34, Neil Smith G4DBN wrote:

That sounds like even MORE fun to be had.  I did start wondering idly about a truly sparkly bling subreflector made by machining and polishing a negative of the hyperbola in aluminium, then using the anaerobic zincate/gold/nickel/alkaline copper plating process to electroform a positive.  A quick boil in caustic soda to dissolve the aluminium mandrel, and I'd be left with a gold-plated subreflector that would look *awfully* nice.

Gold Solutions Plating https://gsplating.co.uk/ do some fancy solutions and charge controllers that are supposed to offer a mirror finish even on plastics. I was looking at much smaller parts though, using electroforming on an aluminium positive mandrel of corrugated horns, with the zincate+gold+copper process. Their electroforming kits use a graphite colloid paint as the base layer on the plastic, but I don't think they sell it separately. The old way was to use a well-thinned cellulose laquer to paint the object, dust it with fine bronze powder while it was still wet, then plate using acidic copper solution. I did see someone trying zinc cold galvanising paint (Galvafroid or Rust-Anode) with fine copper dust added, but there are some tricky issues with copper over zinc.

I wonder how many non-radio folks imagine that these are the sort of things we get up to in the pursuit of the hobby?

Neil G4DBN



On 23/12/2021 11:50, Barry Chambers wrote:


Neil, Colin

another route for the subreflector is to start with the spherical profile provided by the bottom of an aerosol spray can and to copper plate this but using an appropriate electrode shape so as to build up the copper layer preferentially at the centre to achieve the hyperbolic curve. I haven't done the maths but as a first try, the electrode could be a flat circular plate of smaller diameter than the aerosol can bottom and the variation with radius of the spacing between the sphere and the flat plate will give differential plating.

This process could be done in reverse by repeated dipping the end of a piece of copper bar into a plating bath (operating in reverse) and using an immersed ring electrode.

73 Barry, G8AGN






 

 

Virus-free. www.avg.com


R Hopkins
 

I've been copper plating the inside of plastic boxes for years.
I use Ambersil Cold Galvanizing Spray from CPC.

Roger GW4NOS

On Thu, 23 Dec 2021 10:48:29 +0000, you wrote:

Interested in your copper plating plastic. What & where the nickel spray?

Andy
www.g4jnt.com



On Thu, 23 Dec 2021 at 10:45, Colin G4EML <colin@durbridge.me.uk> wrote:

Hi Barry,

I can’t answer that as I didn’t design it and don’t fully understand it. I
just used the figures from Neil’s post on the 30th November.

Colin.



On 23 Dec 2021, at 09:53, Barry Chambers <b.chambers@sheffield.ac.uk>
wrote:

Colin

what f/D have you chosen for your virtual dish please?

73 Barry G8AGN


On 22/12/2021 23:24, Colin G4EML wrote:
Following on from the recent discussions on a sub reflector for the 18”
dishes I decided to see if I could 3D print one.

Based on Neil’s hyperbola figures I drew the curve and rotated it to
form the solid.

I printed it in PLA with a layer size of 0.1 mm to get it as smooth as
possible. Then sanded out the very slight ridges with wet and dry paper
while spinning it in the lathe.

Then sprayed with Nickel screening paint to make it conductive.

That may have been enough but the Nickel paint had a resistance of about
4 ohms measured across the diameter. So I decided to try Copper plating it.

6 hours at 300mA in copper sulphate gave a good coating and reduced the
resistance to as close to zero as my meter can measure. The finish is Matt
but it does burnish up to shining copper.

I have not tested it yet but I am very happy with the result so far, and
a happy few hours was had remembering my electroplating school science
experiments.

Colin G4EML






--
73
Barry, G8AGN















Andy G4JNT
 

Folks - 
All this new stuff about plating and metal sprays.   It's just asking to be written up
Next (or the one after) DNotes in RadCom will be mentioning some of this.   Are there aspects of electroplating other than dipping in copper sulphate and passing a few tens of milliamps per cm^2 that are important?



On Thu, 23 Dec 2021 at 15:41, R Hopkins via groups.io <gw4nos=btinternet.com@groups.io> wrote:
I've been copper plating the inside of plastic boxes for years.
I use Ambersil Cold Galvanizing Spray from CPC.

Roger      GW4NOS

On Thu, 23 Dec 2021 10:48:29 +0000, you wrote:

>Interested in your copper plating plastic.  What & where the nickel spray?
>
>Andy
>www.g4jnt.com
>
>
>
>On Thu, 23 Dec 2021 at 10:45, Colin G4EML <colin@...> wrote:
>
>> Hi Barry,
>>
>> I can’t answer that as I didn’t design it and don’t fully understand it. I
>> just used the figures from Neil’s post on the 30th November.
>>
>> Colin.
>>
>>
>>
>> On 23 Dec 2021, at 09:53, Barry Chambers <b.chambers@...>
>> wrote:
>>
>> Colin
>>
>> what f/D have you chosen for your virtual dish please?
>>
>> 73 Barry G8AGN
>>
>>
>> > On 22/12/2021 23:24, Colin G4EML wrote:
>> > Following on from the recent discussions on a sub reflector for the 18”
>> dishes I decided to see if I could 3D print one.
>> >
>> > Based on Neil’s hyperbola figures I drew the curve and rotated it to
>> form the solid.
>> >
>> > I printed it in PLA with a layer size of 0.1 mm to get it as smooth as
>> possible. Then sanded out the very slight ridges with wet and dry paper
>> while spinning it in the lathe.
>> >
>> > Then sprayed with Nickel screening paint to make it conductive.
>> >
>> > That may have been enough but the Nickel paint had a resistance of about
>> 4 ohms measured across the diameter. So I decided to try Copper plating it.
>> >
>> > 6 hours at 300mA in copper sulphate gave a good coating and reduced the
>> resistance to as close to zero as my meter can measure. The finish is Matt
>> but it does burnish up to shining copper.
>> >
>> > I have not tested it yet but I am very happy with the result so far, and
>> a happy few hours was had remembering my electroplating school science
>> experiments.
>> >
>> > Colin G4EML
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> --
>> 73
>> Barry, G8AGN
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>
>
>
>








Colin G4EML
 

Hi Andy,

 

I am far from an expert but all I did was make a saturated solution of copper sulphate in a glass jam jar.

Suspend a piece of copper pipe connected to the positive supply in the solution with the item to be plated connected to the negative supply.

I used a current limited power supply but you could just as easily use a variable voltage supply and just adjust to get the required current. It only needs a few volts.

 

Copper Sulphate is toxic and a skin irritant so the necessary precautions need to be taken.

 

Rogers idea of cold galvanizing spray contains zinc and sounds a lot cheaper than the Nickel spray.  I think anything conductive is probably suitable.

 

Colin.

 

 

 

 

 

From: Andy G4JNT
Sent: 23 December 2021 16:33
To: UK Microwaves groups.io
Subject: Re: [UKMicrowaves] 3D printed cassegrain reflector.

 

Folks - 

All this new stuff about plating and metal sprays.   It's just asking to be written up

Next (or the one after) DNotes in RadCom will be mentioning some of this.   Are there aspects of electroplating other than dipping in copper sulphate and passing a few tens of milliamps per cm^2 that are important?

 

 

 

On Thu, 23 Dec 2021 at 15:41, R Hopkins via groups.io <gw4nos=btinternet.com@groups.io> wrote:

I've been copper plating the inside of plastic boxes for years.
I use Ambersil Cold Galvanizing Spray from CPC.

Roger      GW4NOS

On Thu, 23 Dec 2021 10:48:29 +0000, you wrote:

>Interested in your copper plating plastic.  What & where the nickel spray?
>
>Andy
>www.g4jnt.com
>
>
>
>On Thu, 23 Dec 2021 at 10:45, Colin G4EML <colin@...> wrote:
>
>> Hi Barry,
>>
>> I can’t answer that as I didn’t design it and don’t fully understand it. I
>> just used the figures from Neil’s post on the 30th November.
>>
>> Colin.
>>
>>
>>
>> On 23 Dec 2021, at 09:53, Barry Chambers <b.chambers@...>
>> wrote:
>>
>> Colin
>>
>> what f/D have you chosen for your virtual dish please?
>>
>> 73 Barry G8AGN
>>
>>
>> > On 22/12/2021 23:24, Colin G4EML wrote:
>> > Following on from the recent discussions on a sub reflector for the 18”
>> dishes I decided to see if I could 3D print one.
>> >
>> > Based on Neil’s hyperbola figures I drew the curve and rotated it to
>> form the solid.
>> >
>> > I printed it in PLA with a layer size of 0.1 mm to get it as smooth as
>> possible. Then sanded out the very slight ridges with wet and dry paper
>> while spinning it in the lathe.
>> >
>> > Then sprayed with Nickel screening paint to make it conductive.
>> >
>> > That may have been enough but the Nickel paint had a resistance of about
>> 4 ohms measured across the diameter. So I decided to try Copper plating it.
>> >
>> > 6 hours at 300mA in copper sulphate gave a good coating and reduced the
>> resistance to as close to zero as my meter can measure. The finish is Matt
>> but it does burnish up to shining copper.
>> >
>> > I have not tested it yet but I am very happy with the result so far, and
>> a happy few hours was had remembering my electroplating school science
>> experiments.
>> >
>> > Colin G4EML
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> --
>> 73
>> Barry, G8AGN
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>
>
>
>






 


Andy G4JNT
 

I have some CuSO4, a jar of it waiting for a use.   Will stick a nickel spray can on the next Farnell order   



On Thu, 23 Dec 2021 at 17:47, Colin G4EML <colin@...> wrote:

Hi Andy,

 

I am far from an expert but all I did was make a saturated solution of copper sulphate in a glass jam jar.

Suspend a piece of copper pipe connected to the positive supply in the solution with the item to be plated connected to the negative supply.

I used a current limited power supply but you could just as easily use a variable voltage supply and just adjust to get the required current. It only needs a few volts.

 

Copper Sulphate is toxic and a skin irritant so the necessary precautions need to be taken.

 

Rogers idea of cold galvanizing spray contains zinc and sounds a lot cheaper than the Nickel spray.  I think anything conductive is probably suitable.

 

Colin.

 

 

 

 

 

From: Andy G4JNT
Sent: 23 December 2021 16:33
To: UK Microwaves groups.io
Subject: Re: [UKMicrowaves] 3D printed cassegrain reflector.

 

Folks - 

All this new stuff about plating and metal sprays.   It's just asking to be written up

Next (or the one after) DNotes in RadCom will be mentioning some of this.   Are there aspects of electroplating other than dipping in copper sulphate and passing a few tens of milliamps per cm^2 that are important?

 

 

 

On Thu, 23 Dec 2021 at 15:41, R Hopkins via groups.io <gw4nos=btinternet.com@groups.io> wrote:

I've been copper plating the inside of plastic boxes for years.
I use Ambersil Cold Galvanizing Spray from CPC.

Roger      GW4NOS

On Thu, 23 Dec 2021 10:48:29 +0000, you wrote:

>Interested in your copper plating plastic.  What & where the nickel spray?
>
>Andy
>www.g4jnt.com
>
>
>
>On Thu, 23 Dec 2021 at 10:45, Colin G4EML <colin@...> wrote:
>
>> Hi Barry,
>>
>> I can’t answer that as I didn’t design it and don’t fully understand it. I
>> just used the figures from Neil’s post on the 30th November.
>>
>> Colin.
>>
>>
>>
>> On 23 Dec 2021, at 09:53, Barry Chambers <b.chambers@...>
>> wrote:
>>
>> Colin
>>
>> what f/D have you chosen for your virtual dish please?
>>
>> 73 Barry G8AGN
>>
>>
>> > On 22/12/2021 23:24, Colin G4EML wrote:
>> > Following on from the recent discussions on a sub reflector for the 18”
>> dishes I decided to see if I could 3D print one.
>> >
>> > Based on Neil’s hyperbola figures I drew the curve and rotated it to
>> form the solid.
>> >
>> > I printed it in PLA with a layer size of 0.1 mm to get it as smooth as
>> possible. Then sanded out the very slight ridges with wet and dry paper
>> while spinning it in the lathe.
>> >
>> > Then sprayed with Nickel screening paint to make it conductive.
>> >
>> > That may have been enough but the Nickel paint had a resistance of about
>> 4 ohms measured across the diameter. So I decided to try Copper plating it.
>> >
>> > 6 hours at 300mA in copper sulphate gave a good coating and reduced the
>> resistance to as close to zero as my meter can measure. The finish is Matt
>> but it does burnish up to shining copper.
>> >
>> > I have not tested it yet but I am very happy with the result so far, and
>> a happy few hours was had remembering my electroplating school science
>> experiments.
>> >
>> > Colin G4EML
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> --
>> 73
>> Barry, G8AGN
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>
>
>
>






 


Neil Smith G4DBN
 

On 23/12/2021 16:33, Andy G4JNT wrote:
Folks - 
All this new stuff about plating and metal sprays.   It's just asking to be written up
Next (or the one after) DNotes in RadCom will be mentioning some of this.   Are there aspects of electroplating other than dipping in copper sulphate and passing a few tens of milliamps per cm^2 that are important?

My well-thumbed copy of The Canning Handbook has about 90 pages dedicated to the processes of copper plating and copper electroforming, along with quite a few chapters covering equipment.  One thing I had forgotten is that the copper foils for PCBs are made by electrodeposition on to slowly-rotating polished cylinders, usually of titanium, but I think stainless steel was also used.  That is part of the reason why the substrate side is matt and the upper surface (which is in contact with the cylinder) is shiny.  As the cylinder rotates , the foil is then wound on to a takeup roller above the tank.

There are different surface effects depending on whether the copper solution is acidified or alkaline. I use air agitation like when etching PCBs at home to avoid uneven buildup from localised depletion of copper ions. There's a lot of Art involved with electroforming to ensure that fine slots are filled and don't have pockets of electrolyte in the middle, and there are fancy things like adding protective rings around sharp corners to prevent excessive local buildup and burning/crystal growth at points and edges.  If you use a smallish current, agitate a bit and ideally have a bit of sulphuric acid in the electrolyte, it works well enough.  I used to electroform leaves and scrumpled-up fine wire, but the surface finish was always pretty rough, and that was in the days before having battery acid made you a terrorist unless you have an EPP licence! I found some ancient lab notebooks from the 1970s and I see that I used something called "Aquadag", as a conductive coating on leaves for electroforming. I think it was just a graphite colloidal suspension. It was messy anyway.

The fancy modern electrolytes have brighteners that reduce the size of the copper crystals that tend to form as the layer thickens. For what I'm doing with corrugated horns, that's not important, I want a nice thick and homogeneous buildup on my aluminium mandrels.  Surface finish is irrelevant as I'll be machining it off, whereas Colin's experiment needs a decent surface finish to reduce scattering loss.

I've got a 5kg drum of hydrated copper sulphate crystals in the workshop and lots of wires and suspicious-looking buckets and chemical gloves. If I have any success with the electroformed horns, I'll write that up. Meanwhile, I'll be watching developments with interest.

In slightly related news, Michel F1CLQ has carried out loss tests on some drawn copper  tube that has around 0,7µm measured interior surface roughness. A 10cm length of 2mm ID tube has 1.9 dB measured loss at 122 GHz.  The modelling done by Michel F6DZK showed a graph of loss versus surface roughness that the experiment was trying to verify.

That waveguide loss is the reason I was looking at putting the feedhorn as close to the PCB of the transceivers as possible, and using a polished 2.00 mm reamed hole to minimise scattering losses, leading to the design parameters that Colin has emulated.


Barry Chambers
 

Neil

what virtual f/D are you assuming for your subreflector and horn feed?

I am presently  doing some modelling using openEMS and f/D = 0.7. For ease of fabrication, I am using a W2IMU feed since this only needs a length of metal rod and two sizes of drill.

73 Barry, G8AGN



On 23/12/2021 11:13, Neil Smith G4DBN wrote:

[Edited Message Follows]
[Reason: oops, cut/paste error fixed]

Plating seems like a great solution if the initial coat adheres well to the plastic and the surface finish is better than lambda/10 or so. Others have tried using burnished aluminium foil and had decent results, but the foil might need replacement now and then.
 
The hyperbola I modelled was based on having the feed phase centre pretty much at the face of the dish, and used a 48 mm diameter subreflector, with a space attenuation at the dish edge of around 3.7 dB, so an edge taper at the subreflector of around 8 dB to get the dish illumination to about 12 dB taper at the edge.  Michel F5DZK is working on a non-hyperbolic subreflector curve to see if the gain can be optimised by changing the taper over the dish surface.
To get that illumination of the subreflector needs a horn with a gain around 20dB, using a Pickett-Potter dual-mode horn with an aperture around 4.5 wavelengths, but it needs some testing with a range of gains to get it optimised.  Paul W1GHZ mentioned that he might need to modify the space attenuation expression in the spreadsheet, as it is showing 6.02 dB space attenuation, which would be right if the feed was at the focus of the parabola, but the distance from feed to sub to dish centre is around 210 mm, but feed to sub to dish edge is 315 mm or so and I think that leads to more like 3.7dB space attenuation at the dish edge.  The equivalent f/d is about 1.7 according to the sheet with the hyperbola having an inter-focal length of 111 mm, apex distance to parabola focus about 11 mm, diameter 48 mm, a=44.64 and b=32.82.
 
Looking at the modelled pattern of a Pickett-Potter and matching the pattern to get -8dB at an angle of arctan(24/111) = +-12.2 degrees, something around 4 wavelengths aperture seemed about right, but for experimentation, I think a range of horns of different dimensions would make sense to get the best compromise feedhorn gain.
 
It's perfectly feasible to use a section of guide with a much wider feedhorn closer to a subreflector, I just wanted to see how it would look with the PCB just behind the dish and the feedhorn poking through the dish face a little.  Others have used 2mm ID/4mm OD tubing to extend from the coupler body to a feed much closer to a subreflector with a much higher hyperbolic eccentricity.
 
The big issue here is the accuracy of the Edmund dish surface, which is probably not up to spec for 122 GHz anyway, so a fully optimised Cassegrain will perhaps fail to deliver the 55dBi or so of gain that it ought to provide. My plans for next year are to try making a cast aluminium parabola with a very precisely machined surface, to see if that will work more efficiently than using spun dishes, but I'm limited to about 13 inches diameter on my machines.
 
I wonder if a resin 3D printer might produce a more precise surface, but whether the nickel paint would adhere to the resin, I have no idea.  I have some PETG filament as well as PLA, so I might try to emulate Colin's excellent work with electroforming of a reflector once I get this Prusa fully tuned up.
 
Another possible approach might be to go full-on electroforming, and make a negative of the hyperbolic form, plate it thickly, then melt away the PLA, leaving a copper shell which could be glued to a mount. Only problem would be the initial layer of nickel.  Perhaps carbon colloid or something would work as a removable substrate?  Sounds like a pile of fun to be had anyway.

-- 
73
Barry, G8AGN