I redrew the relay boards for this amp. The pinouts are the same as those I used in the Skyliner so these are tested and I know they work but I wanted a slightly smaller board.
The relay system is one of the first things to go in. For this reason, I have a mixture of small pads which are part of the coil and switch circuit and then signal pads which are based on eyelets and so, much easier to install the wires later, when the other boards go in. These relays, like those in the Skyliner as well, have a load resistor and pin outs for LEDs on the front panel and the pedal. The current draw of two LEDs and the coil are less then 100mA so the traces at 2mm, should provide a negligibly cool conductivity. The signal traces at 1mm if we rate them at 100mA, should see no more than 3 deg Celsius difference. Well within operational limits.
Again, they are using 35um copper plate and I will be improving the current handling of the copper by applying a layer of solder over it and then, because of the small distance between the traces a layer of lacquer.
I can think of another way I might have cast the PAB board so that there would be space to lay the CC resistors flat, but I am going with the vertically mounted resistors to keep all the eyelets accessible on the inner side of the board (relative to the chassis).
For anyone who's interested, I will bundle all these layouts up in printable files in the opening post when I've completed them. Here's the preview to be getting on with...
New 183 build
Moderators: pompeiisneaks, Colossal
-
Stephen1966
- Posts: 1017
- Joined: Sun Nov 03, 2019 1:53 pm
- Location: Czech Republic
- Contact:
1 others liked this
Re: New 183 build
You do not have the required permissions to view the files attached to this post.
Stephen
www.primatone.eu
www.primatone.eu
-
Stephen1966
- Posts: 1017
- Joined: Sun Nov 03, 2019 1:53 pm
- Location: Czech Republic
- Contact:
Re: New 183 build
As soon as I posted the attachment above, I realised I hadn't left enough of a pad for the B+ and these wires carrying the power supply are better installed after the relay boards are bolted down. I've introduced an eyelet on the B+ pad. This will be a lot cleaner to install.
You do not have the required permissions to view the files attached to this post.
Stephen
www.primatone.eu
www.primatone.eu
-
Stephen1966
- Posts: 1017
- Joined: Sun Nov 03, 2019 1:53 pm
- Location: Czech Republic
- Contact:
1 others liked this
Re: New 183 build
This is the new layout FET board:
The Kemet 47uF cap has a rated lifecycle of 2000 hr but it takes up a bit of real estate because of its size. A couple of lessons were learned when I made the Skyliner FET. First, I wanted a board that would arrange the FET in and out eyelets at the top of the board and the Gnd at the bottom. This allows for neater and shorter, cable runs. The board dimensions now comply with other standard designs: L, 54mm x H, 44mm, 46mm between mounting bolt centres. Also, I like the precision trimpots for biasing. In practice, these proved very accurate and I think if you are dealing with anything outside the normal B+ these take care of that. I am also including test points from which I will mount short stubs of wire which will give the probes something other than the components to hang onto. If you use Martin's biasing techniques, you only need three test points. There's also a TO-5, 4-pin transistor socket (UX1)... not strictly necessary but it allows for testing different transistors without soldering issues.
The 183 photos show electrolytics in place of the tantalum caps. I don't know if this would make any real difference but I know the tantalum caps are fine and that's what the design here uses.
The design of the FET here, includes a front mounted Level pot, and the relay posted above. Basically, the Skyliner design. It will become clear when I post the amp layout...
The Kemet 47uF cap has a rated lifecycle of 2000 hr but it takes up a bit of real estate because of its size. A couple of lessons were learned when I made the Skyliner FET. First, I wanted a board that would arrange the FET in and out eyelets at the top of the board and the Gnd at the bottom. This allows for neater and shorter, cable runs. The board dimensions now comply with other standard designs: L, 54mm x H, 44mm, 46mm between mounting bolt centres. Also, I like the precision trimpots for biasing. In practice, these proved very accurate and I think if you are dealing with anything outside the normal B+ these take care of that. I am also including test points from which I will mount short stubs of wire which will give the probes something other than the components to hang onto. If you use Martin's biasing techniques, you only need three test points. There's also a TO-5, 4-pin transistor socket (UX1)... not strictly necessary but it allows for testing different transistors without soldering issues.
The 183 photos show electrolytics in place of the tantalum caps. I don't know if this would make any real difference but I know the tantalum caps are fine and that's what the design here uses.
The design of the FET here, includes a front mounted Level pot, and the relay posted above. Basically, the Skyliner design. It will become clear when I post the amp layout...
You do not have the required permissions to view the files attached to this post.
Stephen
www.primatone.eu
www.primatone.eu
Re: New 183 build
when you mentioned (only) 2000h ..and if you read again datasheet .. usually they provide this info on the max tempwrature.. in 20-30 degrees temp, C will last hundread thousands .. and +h..Stephen1966 wrote: ↑Sun Nov 13, 2022 5:30 pm
The Kemet 47uF cap has a rated lifecycle of 2000 hr but
-
Stephen1966
- Posts: 1017
- Joined: Sun Nov 03, 2019 1:53 pm
- Location: Czech Republic
- Contact:
Re: New 183 build
True, the 2000h is at +125 deg C. It's also rated for 5000h at +105 deg C but it's probably going to be much longer (many years) at the temperatures it's going to be operating at. Another plus is the low ESR - a redundancy, given the heavily filtered supply voltage it will be receiving but all a plus when it comes to the idea of durability.bepone wrote: ↑Sun Nov 13, 2022 11:12 pmwhen you mentioned (only) 2000h ..and if you read again datasheet .. usually they provide this info on the max tempwrature.. in 20-30 degrees temp, C will last hundread thousands .. and +h..Stephen1966 wrote: ↑Sun Nov 13, 2022 5:30 pm
The Kemet 47uF cap has a rated lifecycle of 2000 hr but
https://cz.mouser.com/ProductDetail/?qs ... wF6Q%3D%3D
Stephen
www.primatone.eu
www.primatone.eu
-
Stephen1966
- Posts: 1017
- Joined: Sun Nov 03, 2019 1:53 pm
- Location: Czech Republic
- Contact:
Re: New 183 build
Compared to the complexity of the FET board the Bias and Rectifier board are a doddle 
I will use hook up wire under this board - offcuts of the PT cables for the high voltages. Board is 80mm x 50mm.
I will use hook up wire under this board - offcuts of the PT cables for the high voltages. Board is 80mm x 50mm.
You do not have the required permissions to view the files attached to this post.
Stephen
www.primatone.eu
www.primatone.eu
-
ijedouglas
- Posts: 776
- Joined: Fri Sep 22, 2017 9:07 pm
- Location: Southern California
Re: New 183 build
Looking good!
You may want to drop the snubber caps C38/C39/C40/C41 and use a fast recovery diode. I have had really good results with https://www.mouser.com/ProductDetail/625-RGP10M-E3-73. I guess you could leave the board as-is and then you can test as well
You may want to drop the snubber caps C38/C39/C40/C41 and use a fast recovery diode. I have had really good results with https://www.mouser.com/ProductDetail/625-RGP10M-E3-73. I guess you could leave the board as-is and then you can test as well
Ian
-
Stephen1966
- Posts: 1017
- Joined: Sun Nov 03, 2019 1:53 pm
- Location: Czech Republic
- Contact:
2 others liked this
Re: New 183 build
Thanks Ian, I will give these a try...ijedouglas wrote: ↑Mon Nov 14, 2022 3:03 pm Looking good!
You may want to drop the snubber caps C38/C39/C40/C41 and use a fast recovery diode. I have had really good results with https://www.mouser.com/ProductDetail/625-RGP10M-E3-73. I guess you could leave the board as-is and then you can test as well
Comparing the recovery (switching) times of the 1N4007 https://eu.mouser.com/ProductDetail/?qs ... iCkg%3D%3D and the RGP10M-E3/73 https://www.mouser.com/ProductDetail/625-RGP10M-E3-73 it's 1.5 us and 500 ns, respectively. The latter has a negligibly higher forward voltage of 1.3V compared to 1.1V for the 1N4007 so that is hardly worth consideration. Better surge current handling on the RGP10M-E3/73 with 30A compared to 27A with the 1N4007. In all other respects, looking at the datasheets, they appear to conform with each other.
If there is an element of switching noise to these, it makes sense that the faster RGP10M is going to sound smoother (less discernible). I am curious about the effect of the snubber caps on the faster diodes though. I just placed an order for Merlin's updated book on power supplies, I'm hoping there is some maths in there to help me figure this one out.
An easy mod to perform though. Maybe there are genuine reasons why the faster diodes sound better without the snubbers. If anyone has any thoughts on this, I would appreciate them.
Stephen
www.primatone.eu
www.primatone.eu
-
Stephen1966
- Posts: 1017
- Joined: Sun Nov 03, 2019 1:53 pm
- Location: Czech Republic
- Contact:
Re: New 183 build
Last of the boards, here...
This is the filter board for the relay supply. I went with the same sized board as for the Rectifier/bias board (80mm x 50mm) as my caps are a bit stockier than usual.
https://eu.mouser.com/ProductDetail/Vis ... wjkQ%3D%3D
Again, durability - long service life and low ESR.
Can't emphasize enough that MrD really went for durability in his component choices. Sometimes, regardless of what the datasheets say, this is a deciding factor. He built his amps old-school. Built them to last. No hint of "planned obsolescence" here!
This is the filter board for the relay supply. I went with the same sized board as for the Rectifier/bias board (80mm x 50mm) as my caps are a bit stockier than usual.
https://eu.mouser.com/ProductDetail/Vis ... wjkQ%3D%3D
Again, durability - long service life and low ESR.
Can't emphasize enough that MrD really went for durability in his component choices. Sometimes, regardless of what the datasheets say, this is a deciding factor. He built his amps old-school. Built them to last. No hint of "planned obsolescence" here!
You do not have the required permissions to view the files attached to this post.
Stephen
www.primatone.eu
www.primatone.eu
-
Stephen1966
- Posts: 1017
- Joined: Sun Nov 03, 2019 1:53 pm
- Location: Czech Republic
- Contact:
3 others liked this
Re: New 183 build
Season's greetings friends. Here's wishing you all the best for the Christmas season and the new year.
I've been MIA for a while but that doesn't mean work on the amp stopped. Au contraire, the chassis is 99% ready, the boards are ready to be populated and I have the layout/build layers all mapped out. This evening though, I turned my attention to "shaving the pots."
The pots in question are:
Bass - 350k
Treble - 270k
(OD) Drive - 272k
(OD) Ratio a.k.a. Volume - 285k
I had toyed with all kinds of modifying methods. In the end, I chose a rudimentary set of tools and a softly softly approach. By far, the easiest to modify are the CTS pots. The platen is easily removed from the case to reveal the carbon track.
The disassembled parts look like this:
Here is the linear trace:
And here is the 10% "audio" taper:
I think it's important to understand the structure of the trace in the audio pots particularly. I had done some test pieces where I removed a scintilla of carbon across the whole trace, but actually, in the final pieces, I found that removing only from the overprinted carbon track (seen left of centre) I could get to the target resistance without really affecting the taper. The linear carbon track is printed in a couple (or more) even layers, but the audio print is scalloped over a relatively low resistance layer. The linear trace needs to be thinned across its length but the audio trace only requires thinning across the heavier part of the trace.
It's here that I found "shaving the pots" to be a misnomer, actually all we need is some fine 3200 grit sanding cloth and IPA. With the DMM attached up to the outer pins, I made very light passes with the sanding cloth and kept a watchful eye on the resistance readings.
I found it's important to clean the track with IPA between passes as this can amount to 5k-10k of extra resistance after cleaning. The cotton bud in this shot shows the amount of carbon removed in a single operation. It's very easy to overshoot if you don't clean between passes.
Anyway, mission accomplished. All the pots are within 1k of the target values, and if it weren't for the scratches left on the case lugs, you would never know they have been modified. It's really a delicate operation and very easy to lift the resistance by a 100k in the Bass pot. All the other pots in the amp are close enough and I was lucky enough to get a 2k2 presence pot which won't require any modification at all.
I've been MIA for a while but that doesn't mean work on the amp stopped. Au contraire, the chassis is 99% ready, the boards are ready to be populated and I have the layout/build layers all mapped out. This evening though, I turned my attention to "shaving the pots."
The pots in question are:
Bass - 350k
Treble - 270k
(OD) Drive - 272k
(OD) Ratio a.k.a. Volume - 285k
I had toyed with all kinds of modifying methods. In the end, I chose a rudimentary set of tools and a softly softly approach. By far, the easiest to modify are the CTS pots. The platen is easily removed from the case to reveal the carbon track.
The disassembled parts look like this:
Here is the linear trace:
And here is the 10% "audio" taper:
I think it's important to understand the structure of the trace in the audio pots particularly. I had done some test pieces where I removed a scintilla of carbon across the whole trace, but actually, in the final pieces, I found that removing only from the overprinted carbon track (seen left of centre) I could get to the target resistance without really affecting the taper. The linear carbon track is printed in a couple (or more) even layers, but the audio print is scalloped over a relatively low resistance layer. The linear trace needs to be thinned across its length but the audio trace only requires thinning across the heavier part of the trace.
It's here that I found "shaving the pots" to be a misnomer, actually all we need is some fine 3200 grit sanding cloth and IPA. With the DMM attached up to the outer pins, I made very light passes with the sanding cloth and kept a watchful eye on the resistance readings.
I found it's important to clean the track with IPA between passes as this can amount to 5k-10k of extra resistance after cleaning. The cotton bud in this shot shows the amount of carbon removed in a single operation. It's very easy to overshoot if you don't clean between passes.
Anyway, mission accomplished. All the pots are within 1k of the target values, and if it weren't for the scratches left on the case lugs, you would never know they have been modified. It's really a delicate operation and very easy to lift the resistance by a 100k in the Bass pot. All the other pots in the amp are close enough and I was lucky enough to get a 2k2 presence pot which won't require any modification at all.
You do not have the required permissions to view the files attached to this post.
Stephen
www.primatone.eu
www.primatone.eu
Re: New 183 build
IPA? As in Sierra Nevada or Sculpin? Or do you mean isopropyl alcohol?Stephen1966 wrote: ↑Mon Dec 26, 2022 9:57 pm
I found it's important to clean the track with IPA between passes as this can amount to 5k-10k of extra resistance after cleaning. The cotton bud in this shot shows the amount of carbon removed in a single operation. It's very easy to overshoot if you don't clean between passes.
cheers,
Jack Briggs
Briggs Guitars
Briggs Guitars
-
Stephen1966
- Posts: 1017
- Joined: Sun Nov 03, 2019 1:53 pm
- Location: Czech Republic
- Contact:
Re: New 183 build
You forgot India Pale Alejabguit wrote: ↑Tue Dec 27, 2022 12:09 pmIPA? As in Sierra Nevada or Sculpin? Or do you mean isopropyl alcohol?Stephen1966 wrote: ↑Mon Dec 26, 2022 9:57 pm
I found it's important to clean the track with IPA between passes as this can amount to 5k-10k of extra resistance after cleaning. The cotton bud in this shot shows the amount of carbon removed in a single operation. It's very easy to overshoot if you don't clean between passes.
cheers,
But no, plain old boring isopropyl alcohol. 
Stephen
www.primatone.eu
www.primatone.eu
Re: New 183 build
No - that's what I was referring to...Stephen1966 wrote: ↑Tue Dec 27, 2022 2:52 pm
You forgot India Pale Ale
cheers,
Jack Briggs
Briggs Guitars
Briggs Guitars
-
martin manning
- Posts: 14308
- Joined: Sun Jul 06, 2008 12:43 am
- Location: 39°06' N 84°30' W
2 others liked this
Re: New 183 build
A fish? Snowy mountains? What is he talking about? My go-to is Bell's Two-Hearted, a Michigan river and Hemingway short story reference.
-
pompeiisneaks
- Site Admin
- Posts: 4244
- Joined: Sat Jan 14, 2017 4:36 pm
- Location: Washington State, USA
- Contact:
Re: New 183 build
Out here on the west coast I love the bale breaker IPA's there are two specifically that rock, topcutter and bottomcutter... mmm GOODmartin manning wrote: ↑Wed Dec 28, 2022 12:28 pm A fish? Snowy mountains? What is he talking about? My go-to is Bell's Two-Hearted, a Michigan river and Hemingway short story reference.
~Phil
tUber Nerd!