I've been designing some pcb boards for both ODS and Dumblelator.
This is the file I used for reference:
And the attachment is the PCB design.
Could you guys check the dumblelator pcb for errors etc.? It's a 2 layer board: dark traces are bottom side and light traces are top side. Bottom side is in x-ray vision.
You do not have the required permissions to view the files attached to this post.
Last edited by erwin_ve on Wed Mar 20, 2019 3:17 pm, edited 2 times in total.
Seems we are seeing multiple PCB recently, which is cool. Not picking on this particular one. Wonder about wisdom of tube sockets directly attached however. Fixed more than a few dead duds in commercial designs. Better board material and heavy copper helps. Anyone have evidence to recommend this versus the obvious alternative of flying wires to a socket?
didit wrote: ↑Mon Mar 18, 2019 3:37 pm
Seems we are seeing multiple PCB recently, which is cool. Not picking on this particular one. Wonder about wisdom of tube sockets directly attached however. Fixed more than a few dead duds in commercial designs. Better board material and heavy copper helps. Anyone have evidence to recommend this versus the obvious alternative of flying wires to a socket?
Best .. Ian
Hi Ian,
real dumblelators also have the sockets on the pcb. I 've build several and never encountered problems.
didit wrote: ↑Mon Mar 18, 2019 3:37 pm
Seems we are seeing multiple PCB recently, which is cool. Not picking on this particular one. Wonder about wisdom of tube sockets directly attached however. Fixed more than a few dead duds in commercial designs. Better board material and heavy copper helps. Anyone have evidence to recommend this versus the obvious alternative of flying wires to a socket?
Best .. Ian
Hi Ian,
real dumblelators also have the sockets on the pcb. I 've build several and never encountered problems.
Yep, understood. A few Watts heater and at most a few more working Watts off the plate unlikely to be an issue.
For fairness most, though not all, issues past encountered were power tubes -- EL84s in particular -- in upside down orientation.
fred.violleau wrote: ↑Mon Mar 18, 2019 4:35 pm
1 question : on some D'lator V8 bypass caps is 5uf. Yours and others is 25uf. What sonic difference would it make?
Thanks,
Fred.
Fred.
at a guess, how much low end is let through. The 5 uF will filter out more low end.
erwin_ve wrote: ↑Fri Mar 15, 2019 9:38 pm
Could you guys check the dumblelator pcb for errors etc.?
Checked. No outright errors I caught. Might move recovery B+ pad up and away from "Out" output. Out of caution. And your labels might be clearer: "B+1", "B+2" vs "B+" twice; and "recovery in" rather than "grid".
fred.violleau wrote: ↑Mon Mar 18, 2019 4:35 pm
1 question : on some D'lator V8 bypass caps is 5uf. Yours and others is 25uf. What sonic difference would it make?
Thanks,
Fred.
Fred.
The cathode bypass cap in my example is 47uF!
As Marcus said in his reply above; it will set the amount of low end.
The difference between 5uF and 47uF is a the roll off point. At 5uF it starts to roll off at 120-130Hz, At 47uF it starts to roll off at 90-100Hz.
For the low E string you got 1dB more with the 47uF cap. It's a tone shaper in the pre-amps for most designs(check twin reverb and ODS differences on the cathode bypass cap.). In case of the dumblelator: all shaping has been done in the pre-amp sections of the ods; So basically you can make it hifi if you want( in the ideal world.)
Rob Robinette, Merlin Blencowe(valve wizard) and Randal Aiken have some useful website information on this topic.
Make sure you search for triode gain stage and load lines first and then cathode bypass cap.
When understanding this, you got a basic understanding how and why different values are chosen for the triode gain stage.
erwin_ve wrote: ↑Fri Mar 15, 2019 9:38 pm
Could you guys check the dumblelator pcb for errors etc.?
Checked. No outright errors I caught. Might move recovery B+ pad up and away from "Out" output. Out of caution. And your labels might be clearer: "B+1", "B+2" vs "B+" twice; and "recovery in" rather than "grid".
Best .. Ian
Thank you for checking Ian!
The labels are doable except for "recovery in"; that is taking up too much space, done this above and below.
The B+ pad vs Out; good catch, I have to rethink this; the brackets that hold the pcb in vertical mount are also close to the B+ pad.
Erwin
You do not have the required permissions to view the files attached to this post.
fred.violleau wrote: ↑Tue Mar 19, 2019 10:58 am
@Marcus, @Erwin, thanks for the replies! I read alot about hipass and lowpass filters and sometimes I need time for all the information to sink in
Fred.
Fred.
it's one of the things Dumble did to take 'fender' amps into 'Dumble' territory. If you look Fender used 25uf cathode bypass caps on a lot of amps whereas Dumble ODS amps tend to be 4.7 / 5uF, this tightens up the bottom end a lot and stop the flubiness you can get when driving a fender hard. Combine that with extra power filtering and you get an amp that holds together a lot better at volume.
erwin_ve wrote: ↑Tue Mar 19, 2019 10:20 am
The B+ pad vs Out; good catch, I have to rethink this; the brackets that hold the pcb in vertical mount are also close to the B+ pad.
Suggest simply move that B+ "wire" solder pad up to the point where the trace bends to come down to the bottom edge.
erwin_ve wrote: ↑Tue Mar 19, 2019 10:20 am
The B+ pad vs Out; good catch, I have to rethink this; the brackets that hold the pcb in vertical mount are also close to the B+ pad.
Suggest simply move that B+ "wire" solder pad up to the point where the trace bends to come down to the bottom edge.
Maybe; I'll do a paper model on that one with the mounting brackets, as soon as the mounting brackets arrive.