Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: scstill on March 13, 2021, 08:43:33 am
-
is it correct to size the PT for max tube voltage and current?
For example, for two 6L6GC in push pull should the PT HV be sized for 450v (or 500v) and 232ma (210 for plate and 22 for screen)? plus the 12ax7 tubes at max spec
-
It may be helpful if you provided a bit of background, explain what you’re hoping to achieve and what the context is.
-
to build a Supro 1696TN
will include 12ax7 tremelo and reverb
-
More info needed, eg idle HT VDC, OT spec, power output.
Doesn't ClassicTone or somewhere have specs for that model available?
-
this might get you started;
from Dougs library;
Supro_1690t.pdf (el34world.com) (https://el34world.com/charts/Schematics/files/Supro/Supro_1690t.pdf)
-
^+1
The 1690T schematic would be a good start. Note the 310 volts AC supplied to the rectifier plates. A 5U4 rectifier should result in about 400 to 410 bolts DC B+ supply to the 6L6s. RCA datasheet for 6L6 GB indicates a pair in AB1 operation wants about 90 ma. at idle, and 150 ma. under full load. A Hammond 272FX looks like it would be a good fit.
https://www.hammfg.com/files/parts/pdf/272FX.pdf (https://www.hammfg.com/files/parts/pdf/272FX.pdf)
-
The tubes I am going with are 6L6GC (not 6L6GB) with AB1 spec of 470v 210ma max 116ma idle (Tung Sol)
To get 450vdc at the plate I would need 780vct with a 50v drop on 5U4
And will need max 237ma (210ma at plates, 22ma at screens and 5ma for the four 12ax7)
Not sure if Hammond 274BX (750vct 201ma $87) has enough current
Or should I spend a little more for the 278X (800vct 230ma $108)
-
I've built a couple of t-bolts - both rectifier tube powered amps - with PTs spec'd as follows
HT: 325-0-325VAC @ 180mA
Heater: 6.3VAC @ 4A
Rectifier: 5VAC @ 3A
With a 5AR4 and 22uF reservoir cap, I got 2 x cathode-biased 5881s idling at 400VDC with a shared 220R (227R) 10W cathode resistor at 29.3VDC. The screens were sitting at 395VDC. The PI node was 349VDC and the preamp node was 246VDC. (With a shared 270R (265R) cathode resistor, the voltages changed a bit. Plates 412, Screens 406, cathode node 32).
With a 5U4G, I got 379 plates (Ip 56.9mA on each tube), 375 screens (4.3mA Ig2 on each tube), and 27.8V across the 227R. The PI node was 313 and the pre-amp node was 195.
So I'd say, if you want to use 6L6GC with a 5U4G and get 'em cooking flat out at 400V with a 200R cathode resistor, maybe go for 350-0-350
-
The tubes I am going with are 6L6GC (not 6L6GB) with AB1 spec of 470v 210ma max 116ma idle (Tung Sol)
To get 450vdc at the plate I would need 780vct with a 50v drop on 5U4
And will need max 237ma (210ma at plates, 22ma at screens and 5ma for the four 12ax7)
Not sure if Hammond 274BX (750vct 201ma $87) has enough current
Or should I spend a little more for the 278X (800vct 230ma $108)
The 274BX has a no-load voltage of 791 VAC to center tap. On start-up, the DC voltage will exceed the 450 volt rating of the electrolytic caps in the power supply. Peak start-up voltage would be about 560 volts DC. The transformer will supply 750 volts under a 201 ma. load. The 5U4 does not have a straight 50 volt drop. The 50 volt drop specification is for a 275 ma. load. Under a 122 ma. load the 5U4 will drop about 25 volts. The 274DX with a 5U4 will provide about 485 volts DC under a 122 ma. load.
The 6L6GC operating points that you are referencing are for fixed bias operation. Do you want to build a fixed biased output section? Cathode bias operation operating points are closer to the operating points that I referenced earlier.
-
To get 450vdc at the plate I would need 780vct with a 50v drop on 5U4
I will echo what pdf said as to what you want to achieve. What drives the desire for 450V? If its just the tube spec I would downplay the importance. If you are going for maximum clean volume - ok. But most folks building Valcos are not looking for that. 350-400v is typically where the big Valcos run; 300-350 for the 6973 and 6V6 models. Or at least that has been my experience repairing them. I would focus on making sure you have the amperage covered. I have liked the Classic Tone iron, but they are going out of business. I think it is worth looking at Musical Power Supplies - Musical Matt on ebay. I think his trannies are good - fairly priced - and made in the USA
-
hammond 274AX or the classictone 40-18060 if you can find one.
http://www.triodeelectronics.com/40-18060.html
http://www.hammondmfg.com/pdf/EDB274AX.pdf
--pete
-
The tubes I am going with are 6L6GC (not 6L6GB) with AB1 spec of 470v 210ma max 116ma idle (Tung Sol)
To get 450vdc at the plate I would need 780vct with a 50v drop on 5U4...
Do you really want to build the 1696? Because that will probably have a HT of around 400V, not 450V.
Cathode bias operating conditions will be rather different than those you mention above.
What OT are you planning to use?
I guess that the 1696 primary impedance will be somewhere in the range 5k to 10k. Lower impedances may not get the best results from cathode bias.
-
The tubes I am going with are 6L6GC (not 6L6GB) with AB1 spec of 470v 210ma max 116ma idle (Tung Sol)
To get 450vdc at the plate I would need 780vct with a 50v drop on 5U4
And will need max 237ma (210ma at plates, 22ma at screens and 5ma for the four 12ax7)
Not sure if Hammond 274BX (750vct 201ma $87) has enough current
Or should I spend a little more for the 278X (800vct 230ma $108)
Under a 122 ma. load the 5U4 will drop about 25 volts. The 274DX with a 5U4 will provide about 485 volts DC under a 122 ma. load.
I misspoke. Looking at the 5U4 plate curves, with the 274BX, the B+ should come in at around 435 DC volts under a 122 ma. load.
-
Do you really want to build the 1696? Because that will probably have a HT of around 400V, not 450V.
Cathode bias operating conditions will be rather different than those you mention above.
What OT are you planning to use?
I guess that the 1696 primary impedance will be somewhere in the range 5k to 10k. Lower impedances may not get the best results from cathode bias.
Thinking of using Hammond 1460J for OT (40w, 4000ohm Primary to 4/8/16 secondary, $47)?
Allows switched flexibility to use other speaker cabinets. The speaker in cabinet will be '64 Jensen 15" 8ohm
https://www.hammfg.com/files/parts/pdf/1760J.pdf Does the primary Need to be closer to 5k?
On the bias, aren't most Supros including the 1696tn Cathode biased? So it seems that I should stay with that.
Since the Thunderbolt has a Classictone PT at 700v/125ma; maybe I should stay with that range in the 1696.
Why the 1696...
I like the idea of control panel on the top rather than in the back like the Tbolt. And I think the concept of a front and back slanted cabinet and a split amp chassis is pretty unusual. I bought the speaker and decided to build the cabinet and amp around it. This was by first handmade cabinet and first bend your own chassis. So far it has come out pretty decent, (but I would use lighter gauge aluminum next time). Now I am in the process of gathering the parts. So far I have a matched pair of Mullard 6L6CG. This post was to get forum advice on the pricey xfmrs.
I also would like any advice on cabling between the top and lower chassis units
-
I think you have a very cool and ambitious project - I applaud your patience.
On the bias, aren't most Supros including the 1696tn Cathode biased? So it seems that I should stay with that.
Since the Thunderbolt has a Classictone PT at 700v/125ma; maybe I should stay with that range in the 1696.
Yes on both in my opinion. I think that bias question was due to your 450V question, and there is no reason to go over 350-400.
Cabling between the pre and power amps - have you searched for a used cable? A lot of tube organs used them - many plug into octal tube sockets - maybe others here have direct experience? For new, I'm not sure if you can buy or need to make up a cable.
-
...Thinking of using Hammond 1460J for OT (40w, 4000ohm Primary to 4/8/16 secondary, $47)?
Allows switched flexibility to use other speaker cabinets. The speaker in cabinet will be '64 Jensen 15" 8ohm
https://www.hammfg.com/files/parts/pdf/1760J.pdf Does the primary Need to be closer to 5k?
...
Given the rest of the circuit, 4k is too low for optimal results; because it will facilitate anode current to increase significantly with signal, there will be a significant degree of bias shift with signal level.
6k6 or 8k are classic common choices for OT primary impedance in this type of scenario.
-
Looks like a cool project. I agree with those who say "why 470volts?" Even with the 6L6Gcs, (a pair of which which I'm trying to bias right now on my own bench), you need wiggle room, and 470 is so high that you won't have much...or any.
Cableing between top and bottom - I've built a few projects like that, and it sorta makes everything exponentially more tricky when pulling the amp in and out of the cabinet to tweak. I learned to use as much shielded cable as possible - particularly for signal lines of course; find a good quality 4-strand shielded cable, or 6 strand if needed.
I also learned to keep the on/off switch as far from the tone stack as possible. I ran two separate cables - one for AC power and one for the tone stack and controls. With good shielding, you can afford to add a little more length to the cables so that, when you pull the amp out to tweak on the bench, you don't have to also remove the controls and on/off switch. Instead, you can just lay the cabinet on its side on the bench and plug it into the chassis from there.
Did I say plug in? I should mention that I made the controls pluggable (is that a word?). I used shielded RCA plugs for signal. I connected the AC wiring using an octo-plug. This is something I learned from seeing how Hammond connected all the stuff in their organs - lotsa octo plugs. It may seem a bit low-brow compared to today's technology, but it works well and you can avoid noise issues fairly easily when done right. I feel confident you will want to have it plug-friendly once you start tweaking.
Good luck with the project.
-
I did a couple split chassis back in the '60s. And I currently have one split chassis, preamp up top and power amp down below. I made an umbilical cable that was hard wired into the preamp and connected to the power amp via an octal plug/socket. I used RG-316 shielded cable for the signal and a shielded twisted pair (STP) for the filaments. I used another STP for the B+ and ground. Take a look...
http://sluckeyamps.com/vox/vox.htm
And my Magnatone has a split chassis with the power supply down below and the rest of the amp up top. Shielded cable not needed...
http://sluckeyamps.com/magnatone/magnatone.htm
Oh yeah. Don't be tempted to build a cable with male connectors on each end! :icon_biggrin:
-
As alternate to the old Octal: PC motherboard power connectors are often rated for hundreds of volts and more current than you need. Get an extender cable and cut it in half.
https://www.newegg.com/p/N82E16812198007?Description=24-pin%20extension
https://www.amazon.com/Cable-Matters-Motherboard-Power-Extension/dp/B01DV1Z66C
(https://images-na.ssl-images-amazon.com/images/I/61c-8wRXa%2BL._AC_SL1500_.jpg)
-
Given the rest of the circuit, 4k is too low for optimal results; because it will facilitate anode current to increase significantly with signal, there will be a significant degree of bias shift with signal level.
6k6 or 8k are classic common choices for OT primary impedance in this type of scenario.
If 6000 to 8000 is common then why does the Tbolt schem (on this site) show an OT 35w, 5000:8 and Classictone have a direct replacement OT (40-18061)for which is 35w, 5000:8. They claim it was built identical to original. It is quite pricey at $85 but I understand its now scarce. And it doesn't offer output impedence choice.
Hammond offers alot of choices (at decent value?) but not in perfect combination. 6k to 2/4/8 40w; 4k to 4/8/6 40w $47; 4200 to 2/4/8 60w (which would be 8400 connected to 4/8/16 right?). BTW - is there tonal differenece between paper and bobbin?? https://www.hammfg.com/electronics/transformers/classic/1750
They also have a 20w at 6600 to 4/8/16 but would this be too small? I know many people have said the Silvertone 1484 (also 6L6GC) OT is too small but I have one of those and it sounds really good although I have not cranked it because of ear pain. :smiley:
-
On the split chassis cable....
1) there is a heater(2), HV and ground, can these be bundled? with their own individual shields?
2) there is a push pull signal(2) should these be bundled with their own shield? Could an RCA cable work here?
3) there is an RCA reverb cable, diverting from original design as I had a tank laying around and wanted to build my first reverb circuit
did I miss anything??
-
I did a couple split chassis back in the '60s. And I currently have one split chassis, preamp up top and power amp down below. I made an umbilical cable that was hard wired into the preamp and connected to the power amp via an octal plug/socket. I used RG-316 shielded cable for the signal and a shielded twisted pair (STP) for the filaments. I used another STP for the B+ and ground. Take a look...
http://sluckeyamps.com/vox/vox.htm
And my Magnatone has a split chassis with the power supply down below and the rest of the amp up top. Shielded cable not needed...
http://sluckeyamps.com/magnatone/magnatone.htm
Oh yeah. Don't be tempted to build a cable with male connectors on each end! :icon_biggrin:
Greatly impressed by your website, Mr. Luckey. I especially like the TV cart conversion and the Intruders.
-
I did a couple split chassis back in the '60s. And I currently have one split chassis, preamp up top and power amp down below. I made an umbilical cable that was hard wired into the preamp and connected to the power amp via an octal plug/socket. I used RG-316 shielded cable for the signal and a shielded twisted pair (STP) for the filaments. I used another STP for the B+ and ground. Take a look...
http://sluckeyamps.com/vox/vox.htm
very nice....
Was the reason to put the vox PI in the lower section to run one signal line and keep that ground close? Why did Supro put their's on top? I have read that PI could be grounded either to Pre or Output. Any preference?
-
I've always considered the PI as part of the power amp. There is an ext. input jack on the lower chassis that can accept a line level signal, such as a different guitar preamp. Another reason for putting the PI down under is you only have to run one signal line between the two chassis.
I have no idea why Supro did anything they did. Well I do know that price point was a BIG factor.
-
Getting back to this project. Hopefully it is still visible to you all.
I have a question about the cable between the split chassis. I plan to cable the 6.3v heater, and a single HV from the lower to the upper, can these three wires be run as a 3 connector in the same shielded cable?
In a separate shielded twisted pair cable I will run the 2 signal cables from the top chassis (PI) to the lower chassis
-
I would use three STP cables. Here's what I did in a similar split chassis. But I didn't need a STP cable for my signal. I hard wired the umbilical to the preamp and used an octal plug/socket for the power amp connection.
http://sluckeyamps.com/vox/v_10.jpg
http://sluckeyamps.com/vox/Vox.pdf
-
Sluckey, do you hear any difference with the EL84's set up in grid bias instead of the usual K bias?
Marshall, Traynor and a few others have amps with grid biased EL84's, but still pretty rare.
-
Haha! I can barely hear the difference between an E chord and a G chord. :l2:
-
Marshall, Traynor and a few others have amps with grid biased EL84's, but still pretty rare.
And typically plate/screen voltage near 400v, and recurring mention of EL84 flameout.
I'm not sure what the point is of trying to run EL84 above about 300v plate & screen. Even if current production EL84 tubes have a reputation for lasting 4 or 5 decades... :l2:
-
I would use three STP cables. Here's what I did in a similar split chassis. But I didn't need a STP cable for my signal. I hard wired the umbilical to the preamp and used an octal plug/socket for the power amp connection.
http://sluckeyamps.com/vox/v_10.jpg
http://sluckeyamps.com/vox/Vox.pdf
I am finally building this split chassis cable. How would you dress the shields from each of the three STP? Would you connect them together? Or keep them independent? Would you ground them at both ends of the cable or just at one end?
sneak peak at the amp progress
-
Cable details are in the pdf in your quote.
-
Cable details are in the pdf in your quote.
Yes indeed, Thank you, your drawings are incredibly professional, something to aspire to.
I think I made a mistake (holes are already cut) in that the plug is coming from the Power Supply so there is HV and heater voltages on the pins. I thought of patching the holes and reversing the cable, but not sure there is ever a reason to unplug.
-
My cable is hard wired to the upper chassis and has a male octal plug that connects to a standard octal socket on the lower chassis. No chance of being bitten when the cable is unplugged.
On another note... Looks like you have the upper chassis mounted to the cab with a couple wood blocks on each end of the chassis. Is that right? If so, will you have access to those screws once the chassis is full of stuff?
-
I am going to revise the cable to hard wire to the upper chassis (wish I would have thought about this earlier).
There should be enough room to remove the upper chassis after being wired. The alternative is to add screws on the outside. I was going for a clean look with minimum screws in the cabinet. This is going to be wired point to point as well so I don't have to allocate space for a board.
As another note, rather than tolex cover, I am going for a wood finish. But this is pine and even with pre-stain, it does not take stain very well. Gets blotchy, especially on the ends. Any suggestions for getting a nice uniform finish. Tried boiled linseed but that is a little too light.
-
Any suggestions for getting a nice uniform finish. Tried boiled linseed but that is a little too light.
I am not keen on working with pine as it is very difficult to get a uniform finish. On this cabinet, I had one piece of flamed maple for the top and both sides and the bottom are pine. I used water based analine dye and used amber. On the end grain wood showing in the dovetail, wet the endgrain with water first prior to applying the analine dye.
Then I used semi-gloss urethane.
With respect, Tubenit
-
this is the first cabinet hand made with finger joints by my son. I want to emphasize his craftsmanship.
have been experimenting with different finishes on pine but have not found the secret - yet
but I know its out there. just thinking that someone here has a tip
-
fwiw;
are you using #1 clear pine, or #2 with knots? that alone makes a difference.
Pine has very porous/open uneven grain. being an artist, I experimented with my oil paints, cut it with linseed until it "became stain".
then I brush it on, wait a few minutes, wipe down. repeat for color depth and hue. let it dry for DAYS. once you can run your finger over it hard and not pick up any, or very little stain, I knock down the hi-spots with 400 wet/dry - dry. Wipe it down with a clean rag, slightly damp.
then it's on to whatever clear-coat for OIL based stain you want. Usually 5 coats with light sanding after about the 3rd coat.
you can never get "uniform" from non-uniform wood, what you can get is a good finish that accentuates the wood you have.
-
its clear pine. The knotty parts occurred at regular intervals in the board and I was able to cut clear boards between them. The Lumber Shop sold it as Sugar Pine.
-
My cable is hard wired to the upper chassis and has a male octal plug that connects to a standard octal socket on the lower chassis. No chance of being bitten when the cable is unplugged.
On another note... Looks like you have the upper chassis mounted to the cab with a couple wood blocks on each end of the chassis. Is that right? If so, will you have access to those screws once the chassis is full of stuff?
Would this cable work for me? It has 3 STP inside a convenient common jacket. Each STP has a foil shield and a bare ground wire. Am running two 6v heaters, one HV, pair of signal from Push pull output
-
Absolutely.