845 power stage.

pre65 · Post by **pre65** » Thu Apr 10, 2008 5:01 pm

Hi-i have had discussions with Nick and it seems my best choice,given my output transformers,and my wish to use my existing 300b amp as a driver,is the 845 valve.

The power supply i outlined before used 3 100uf caps before the choke and 3 220uf caps (both 400v) after the choke.

Now i can bid on 12 Rifa-Evox 400v 220uf caps on E-bay,so would it be OK to use that value in both positions ?

Post by **Nick** » Thu Apr 10, 2008 5:08 pm

A first cap of 70uf should be ok with the 6d22s rectifiers, so yes, they should be ok.

pre65 · Post by **pre65** » Thu Apr 10, 2008 6:02 pm

Turning my attention to the 845 filament heating arrangements,most circuits i have seen use a 10v transformer,bridge and cap.

Now i happen to have a couple of 50va 9v dual secondary toroids.Rapid say that they are 2.78 Amps but is that total,or per secondary ?

The 845 heaters are rated at 3.25 Amps so if the secondary's are 2.78 Amps each i can combine them and use low voltage drop diodes in the bridge, that will do the job (hopefully) and i will have saved some more beer vouchers.

Post by **Nick** » Thu Apr 10, 2008 6:25 pm

50/9 = 5 and a bit, so both secondaries in parallel will produce enough current, but you may struggle getting 10v. for the fill from that though.

pre65 · Post by **pre65** » Thu Apr 10, 2008 6:59 pm

When i get the valves i can test to see how many volts i can get,unless you know the hot resistance of the 845 filament so i can use a resistor.

I have several modules i made up for another project using IR50SQ100 ,1N5820 and UF5401 so hopefully (fingers crossed) one of them will just do the job.

pre65 · Post by **pre65** » Thu Apr 10, 2008 8:47 pm

Right-time for a little brain work.

10V @ 3.25 A = 3.08 ohms = 32.5W

I used 5 off 15 ohm 5 or 6W resistors in parallel which gives (in theory) 3 ohms

Using the Variac which is still suffering from past violent abuse,and only goes to about 230 max now my 9V toroid and one of my modules gets 9.4V DC

So in reality,is this enough for a 10v filament ?

Post by **Nick** » Thu Apr 10, 2008 9:39 pm

It will work, but I would try for within 5%. Thorated fillaments are a bit picky.

pre65 · Post by **pre65** » Fri Apr 11, 2008 11:08 am

Bingo !

Trying it direct from the mains,once my "load" had warmed up a bit i was getting 9.7V !!!

5% = 0.5V so within that margin.

Bluebell do a couple of Hammond dedicated 10v filament transformers but i will stick with what i have for the time being.

Post by **Nick** » Fri Apr 11, 2008 12:23 pm

Ok, that will do for the moment.

pre65 · Post by **pre65** » Fri Apr 11, 2008 1:00 pm

Now getting down to spending a bit of money.

Been looking at new 845's and the most reasonable option seems to be
"chinabest2006" on e-bay

http://cgi.ebay.co.uk/ws/eBayISAPI.dll? ... :IT&ih=008

and Jumbo sockets ? Bluebell have Chinese ones @ £9.00 each-are these OK for what i'm doing ?

Post by **Nick** » Fri Apr 11, 2008 1:08 pm

If they are the same as the Chineese Jubmo sockets I bought from Phillip some years ago, yes, thay are perfectly fine.

pre65 · Post by **pre65** » Fri Apr 11, 2008 2:08 pm

Now i'm getting out of my depth !

I need to understand how to sort out the cathode resistors,if i drive the 845 grid from the 300b anode via a cap (what size and quality ?) would something like a 100 ohm humbuck pot and a 1 ohm 1W resistor be suitable ?

If i use interstage as you suggested Nick,would the cathode arrangement stay the same or would i need to bias the grid with another power supply ?

Some 845/211 circuits i have seen don't seem to have any cathode resistors at all,just the 10v feed

Post by **Nick** » Fri Apr 11, 2008 5:29 pm

I would stick to simple and stable cathode bias.

To work out what valve needs a bit of looking at load lines, I will take a look later and see what would work, or anyone else can suggest what they will.

845 shouldn't need A2 so notmal bias should be fine, either cap coupled with a grid leak to 0v, or interstage with other end of the secondary to 0v.

pre65 · Post by **pre65** » Fri Apr 11, 2008 7:16 pm

Hi-i'm trying (very trying

) to understand what's going on.

I have printed the load lines,and drawn a line between 65ma and 1000V which seems to be the "typical operation" data.

Now i have got MJ out and reading about cathode bias.I sort of understand the general idea but i cant make sense of any values.

He seems to suggest that the cathode voltage should be a known value for each valve but it does not appear on the spec sheet (not my copy anyway).

I recently "made" a new clutch for my Tri-BSA scrambler from bit of clutch from BSA,Suzuki and Kawasaki without too much trouble but valve theory ?

Post by **Nick** » Fri Apr 11, 2008 8:12 pm

I can feel some drawing coming on next weekend

.

Ok, given you have about 900v B+, I would think aiming for a cathode bias of about -110v. So thats the cathode at 110v, that leaves 790v across the valve, that puts the op point at something like 80ma. 80ma at 790v is a diss of 63w, so thats making life very esy for the 845, and I think a 6k5 load will give you about 15w of outout from a 220v P/P signal, which should be well within the realm of a 300b to provide.

So to get the cathode at 110v with 80ma, thats a resistor of 1375, so 1k3 with a bit more from the hum pot is about right. 110v at 80ma is dissapating 9w, so I would use a 50W 1k Wirewould, and a 330 or 300 10W resistor in series.

There may be better points, I don't know what your TX's are gapped for, but that would work.

Now, it needs 220v P/P drive, or 80v RMS into the grid, The 300v will need about 25v to drive it, so you need about a gain of 25 from your driver to give full power from 1v RMS in. Don't know what your 300b driver is.

audio-talk

845 power stage.

#1 845 power stage.

#2

#3

#4

#5

#6

#7

#8

#9

#10

#11

#12

#13

#14

#15