I don't think the error was malicious either.
The tube types are a minor issue- with things like this sometimes its all about what tube you can actually get to work. So apparently my comments apply to either.
I think there are enough positive comments from people who've built the Mellow OTL to have a reasonable expectation of a successful project so let's see how things go.
Picking up on a couple of your comments though,
Picking up on a couple of your comments though,
- Tim Mellow comments in his original article about the possibility of a balanced input arrangement and it's something I will likely use as I have a couple of source components with balanced outputs and I plan to explore a balanced attenuator.
- I am likely to dispense with the input caps too as there will be no DC from my sources.
- I am considering seperate power supplies for the different stages, possibly using some Maida regulated supplies I have avaialble.
Thanks, I think the same, using HT3 and HT4 minus-potential points from either left or right side PSU (for ALL) is possible, while this applies for the potential shift only, the heating power itself would come from common transformers.
But the taps to HT3 and HT4 can be either left or right side then. Okay. Thanks for confirming.
Sure, I'll give it a cold-start without tubes first.
I found the Swap Meet listing for the ceramic sockets I mentioned a few posts back - I've no idea whether the guy has any still available but he said he still had twenty in 2021 - these sockets are far superior to the usual sockets you can find on ebay and the like.
https://www.diyaudio.com/community/threads/tube-socket-for-6c33c-plk7-collect-eu-locate-new.281396/
https://www.diyaudio.com/community/threads/tube-socket-for-6c33c-plk7-collect-eu-locate-new.281396/
I don't doubt that it works. I just pointed some of those things out as I've been building OTLs now for over 45 years and I've tried some of the things I see in that schematic. I'm not knocking it, but if you want to try some things, some of them are easy to do.
Not sure I know of any source that makes DC!
Yes, I agree, there is always room to try things out and I think my looking into using the sub-min tubes is a prime example.
As for DC, I've seen plenty of source components out there with some small, and sometimes not so small, level of DC on there outputs, which is why Tim Mellow uses the input caps and references them in his article.
As for DC, I've seen plenty of source components out there with some small, and sometimes not so small, level of DC on there outputs, which is why Tim Mellow uses the input caps and references them in his article.
Hi, does IT mean that Mellow's power supply toroid 2x120V 625VA can feed two channels?Hi gentlemen,
I'm finally putting together all the stuff.. not an easy task while working and having a girlfriend too
Anyway, I have some spare time now and I thought I finish my little OTL of Tim Mellow.
Can you please help clarifying some basic things ?
Question1:
In the original article, there's 1 PSU shown, feeding both Left and Right side circuits, right ?
I think schematics is for mono circuit (of course) but the PSU was meant for driving both sides (stereo).
T1 is specified for 15VA ->
for stereo when feeding 2 input tubes (2x 0.37A) and 4 driver tubes (4x 0.22A) altogether, heater current draw is 0.74 + 0.88 = 1.62A (~10.2VA) so we're still good
T2 is specified for 225VA ->
for stereo when feeding 4x 6C33C tubes (4x 3.6A /12Vseries connected in each/), heater current draw is 14.4A -> ~172 VA so this is again ok
T3 is specified for 625VA
for stereo I assume. One side only needs half of this approximately (or even less).
Question2:
I might be in trouble with my transformers. Or even not - can you elaborate the 'issue' if there's any, please ?
To whatever reason, I ordered 4 transformers for this project 3-4 years ago:
T1 (25VA - more than needed)
T2 (250VA - more than needed)
2x T3 (400VA each - less than original but I have 2 of these then, this way more capacity again)
The idea was to have heaters for both sides on common secondaries as in the original article (I assume), but for the anode voltages let's use 2 separate transformers and full circuits behind them.
So far nice, but now comes the trick and my question
Thinking in stereo integrated constellation:
- all 4 driver tubes' heater potential needs to be pushed down to -450V (HT3).
and I have 2 HT3-s, one for Left and one for Right side and one common heater transformer secondary.
- 2 (of 4) 6S33S' heater potential needs to be pushed down to -150V (HT4).
and I have 2 HT4-s, again for Left and Right side - and one common heater transformer secondary.
Which HT3 and HT4 points shall I use for these common fed heaters ? Left or right ?
Can I do it at all without screwing up something ? (E.g. using Left side's HT3 for all affected heatings and Right side's HT4 for all affected heatings).
Made a little drawing for quicker understanding.. are the blue arrows (the potential shifts) okay so, as an example ?
(I have 2 T3-s).
So in theory only one 2x120V 312,5VA Is neccessary for mono amp. I am not sure about IT because in audioXpress article Is mentioned only one Channel but there Are two Channels in chassis images.
Hi,
A 2*120V 625VA transformer can supply c. 2.6A. At 25W into 8 ohms each channel will require 0.625A, thus a stereo amplifier will require1.25A. So the 625VA transformer can supply almost twice as much current as needed.
At 25W into 8 ohms it will deliver a voltage of c. 14V. Suppose it can also do that into 4 ohms. That is equal to c. 49W. 49W into 4 ohms will require 1.75A - in stereo times 2 = 3.44A. Now, the Mellow OTL is not recommended for 4 ohms, and I doubt it'll be able to supply 14V into 4 ohms (have not tested it and I am on holiday at the moment with no access to my lab, but I can do it sometime next week if anybody is interested).
The bottom line is that a 625VA transformer should be more than enough for a stereo amp and consequently a 312,5VA should be fine for a monoamp.
I hope this helps!
Good luck with the build!
A 2*120V 625VA transformer can supply c. 2.6A. At 25W into 8 ohms each channel will require 0.625A, thus a stereo amplifier will require1.25A. So the 625VA transformer can supply almost twice as much current as needed.
At 25W into 8 ohms it will deliver a voltage of c. 14V. Suppose it can also do that into 4 ohms. That is equal to c. 49W. 49W into 4 ohms will require 1.75A - in stereo times 2 = 3.44A. Now, the Mellow OTL is not recommended for 4 ohms, and I doubt it'll be able to supply 14V into 4 ohms (have not tested it and I am on holiday at the moment with no access to my lab, but I can do it sometime next week if anybody is interested).
The bottom line is that a 625VA transformer should be more than enough for a stereo amp and consequently a 312,5VA should be fine for a monoamp.
I hope this helps!
Good luck with the build!
Hi
Its been quite some time since i have been on here. Way too much work, now slowing down, so hopefully be able to get back to my OTL amp.
just been looking through posts and noticed that on page 37 "ochabot" posted a schematic of the ARES Audio v4. This amp uses 4 Output tubes per channel, as does my amp. (mine is the 8X built with 4 seperate PCB's, rather than just one big one) Looking at the way the 6c33 Heaters are wired, in v4 schematic it is possable to run the amp at 1/2 power by switching off the heaters to two of the tubes on each channel ???. This seems very strange, if they were switched off while it was on with the HT still present it would proberably go BANG!
Anyway, I then noticed the way the filaments are wired. My amp uses the same transformer for the heaters it has 4 X 12v 9 amp windings and there is 8 tubes. So each winding has to power two sets of heaters. In the v4 schematic they are using the same winding to power the heaters on both the high and low sides !
since the center taps of the heaters are connected to ground on the high side and -150v on the low, isn't this pulling the -150v to ground through the heaters?
back when i was testing I used 1 winding per tube (as i only had 4), but I always planned to have each winding power the two parallel tubes. Am I missing something ? or is this just another schematic mistake by ARES.
Its been quite some time since i have been on here. Way too much work, now slowing down, so hopefully be able to get back to my OTL amp.
just been looking through posts and noticed that on page 37 "ochabot" posted a schematic of the ARES Audio v4. This amp uses 4 Output tubes per channel, as does my amp. (mine is the 8X built with 4 seperate PCB's, rather than just one big one) Looking at the way the 6c33 Heaters are wired, in v4 schematic it is possable to run the amp at 1/2 power by switching off the heaters to two of the tubes on each channel ???. This seems very strange, if they were switched off while it was on with the HT still present it would proberably go BANG!
Anyway, I then noticed the way the filaments are wired. My amp uses the same transformer for the heaters it has 4 X 12v 9 amp windings and there is 8 tubes. So each winding has to power two sets of heaters. In the v4 schematic they are using the same winding to power the heaters on both the high and low sides !
since the center taps of the heaters are connected to ground on the high side and -150v on the low, isn't this pulling the -150v to ground through the heaters?
back when i was testing I used 1 winding per tube (as i only had 4), but I always planned to have each winding power the two parallel tubes. Am I missing something ? or is this just another schematic mistake by ARES.
Hi, I'm at the beginning of the design of the hw part of the otl amplifier (pcb, components, chassis). And since I have little space in the chassis, I have to use a common power supply (+-150V) for both channels. My question is, will this connection reduce crosstalk between channels?
The output voltage ripple should be reduced by 1Vp-p compared to four 1mF parallel capacitors - according to the simulation.
Do you see any other advantage except of the ripple ?
The output voltage ripple should be reduced by 1Vp-p compared to four 1mF parallel capacitors - according to the simulation.
Do you see any other advantage except of the ripple ?