dwhitf,
Actually, I prefer Edcor, because of the relatively poor HF performance of EI core transformers versus toroids. Lots of noise on the AC mains. An EI transformer allegedly can filter some of that, whereas the toroid lets more of it though. However, the Antek transformers I have bought for a SS amp project were quiet and worked as advertised. Carefully look at the Antek catalog since many of the high voltage listings also have multiple windings not listed in the overall part number/title.
I do like to overspecify when possible. In other words, if the amp needs 500VA, pick 600-750VA or so.
The Antek AN-64115 is rated 600VA, has two 2.6A secondary windings, PLUS a 6V 5A winding and a 12V 5A winding, for $69
If that seems marginal, then the AN104115 certainly works for $100. You get 2 x 6.2A 114V windings (voltage goes to 120 as current goes to 0A), one each 6V and 12V 8 amp windings!
Stuart
PS I do want to add that you will still need another low voltage winding set, both in 6.3V and 12V becuase of the biasing needs to keep the filament-cathode voltage relationships within limits., but that transformer can be much smaller. Look at the $18 AN-1212 for instance.
Last edited:
If anyone has the time do these look like the correct choices from Antek
AN-0206
AN-3212
Question on the next one it is spec'd at 600VA will this be ok, the schematic calls for 625VA?
AN-64115
EDIT: Looks like I was beat to it!
AN-0206
AN-3212
Question on the next one it is spec'd at 600VA will this be ok, the schematic calls for 625VA?
AN-64115
EDIT: Looks like I was beat to it!
Last edited:
Usually such a control is for the voltage/driver tubes. It might work for the overall gestalt, if so, matched power tubes are a good place to start. I find though that if the control is over to one side, you get enough distortion from the input stage as to make the exercise pointless.
Stuart, would this be strong enough to run both the 120V and the 6.3V
EDCOR Electronics Corporation. XPWR098-120
and how about this for the 12V
EDCOR Electronics Corporation. XPWR027-120
EDCOR Electronics Corporation. XPWR098-120
and how about this for the 12V
EDCOR Electronics Corporation. XPWR027-120
You need much higher current capacity for the 12.6V supplies for the 6C33C heaters. Each 6C33C draws around 3.5A at 12.6V, so you need two 12.6V supplies that can comfortably give at least 7A each.
I am using an Edcor XPWR098-120 in a different OTL I also built (design by Hans Beijner). (In fact, I rather think the XPWR098-120 is the legacy of one they made for me as a special order.) In the HB amplifier, I'm using a Hammond 185G24 for the 6C33C heaters; it has two 12V, 7.3A secondaries, and it gets VERY hot. The XPWR098-120 gets pretty warm too.
For the Tim Mellow amp. I switched to using toroidal transformers. They don't get anywhere near as hot. Also, I have the impression the toroid I'm using for the HT supplies is much better able to supply the required current without "sagging," as compared with the Edcor XPWR098-120.
I haven't tried the Antec toroids; they certainly seem to be significantly cheaper than the Hammonds I have used, so they could be a good choice, I guess.
Chris
Hey yan6,
The 120V transformer is too small. You need at least 2.67A, I'd go for a 3A pair of windings (or 240VCT) per channel. Call Edcor, see if they will build you what you need. They do customs, then add them to their product line. A couple of friends said they were quite affordable. BTW, a line of chokes is on the way, available now if you call.
The 12V transformer is too small also. It can provide 1 amp of current. Each output tube needs up to 3.6 amps at 12.6 volts. so each transformer winding must deliver 7.2 amps. I'd choose an 8 amp winding, minimum, and that's two windings needed.
For the 6.3V windings, the V1 power supply can actually be 6.3 or 12.6 volts depending on how you connect the tube. 1/2 amps 12.6/6.3 volts is plenty in either case if you're supplying both channels from one PT. For the EF86 tubes, the 6.3V 200mA per tube heater means 0.4A per winding, so 1 amp is okay here also.
Stuart
Beat me to it!
Last edited:
no I had the pinout right it just did not work and it appears it took down the neon blub. I checked the pin out 3 times before I put the amp on with the 6SL7 in it and once after and it was correct. I have used the 6SL7 several times and I am familiar with its pin ot so I know it was correct the bottom line is it does not work in that arrangement. As you previously suggested it would need a different front end to work correctly. Thx
Does anyone have a good source for nice looking meters? Also would it be possible to combine the two volume into one pot/attenuator. I think I recall seeing pots with matched decks, but how about a stepper attenuators like that?
Unless you really want to have meters for the sake of appearance, you could leave them out put low-value resistors there instead, which allows you to measure the voltage drop, and hence determine the current flow. I am using 0.1 ohm resistors, with a pair of tip jacks for each. 200mA corresponds to 20mV across the resistor. You only need to measure the current occasionally, after the initial setting up.
Regarding the volume controls, I would say it depends on whether you plan to use a preamp with its own controls or not. The OTL works by itself just fine, if your audio source is a CD player or equivalent. In which case it is convenient to be able to control the volume levels of the two channels independently.
Chris
Ya I think my plan is to isolate the power supply in a separate box and have a input selector switch and single volume control on the OTL box. I wouldnt mind meters if I could find something nice.
There are some decent looking meters here, along with other nice parts and chassis. Look towards the bottom of the 'switch' page.
I have used and listened to P&G faders, ladder and series stepped attenuators, transformer volume controls, etc. Never heard anything sound as good as LDR attenuators. Look into it. "Lightspeed" thread and others.
Stuart
Parallel Outut Tube PSPICE SIMULATION
I "built" this amp in PSPICE using parallel output tubes and I'm getting about 70 watts into 8 ohms. Does anyone have a schematic of an actual working circuit?
I "built" this amp in PSPICE using parallel output tubes and I'm getting about 70 watts into 8 ohms. Does anyone have a schematic of an actual working circuit?
An externally hosted image should be here but it was not working when we last tested it.
Frequency Response
Frequency response plot:
Frequency response plot:
An externally hosted image should be here but it was not working when we last tested it.
Output on verge of clipping
Here's a trace of the output at the onset of visible distortion:
Here's a trace of the output at the onset of visible distortion:
An externally hosted image should be here but it was not working when we last tested it.
Build?
I'd really like to build this with two output tubes. Has anyone successfully built this with parallel output tubes and if so do you have a schematic?
My main concerns are current balance in the output tubes and the dissipation of the EF86 driver.
I'd really like to build this with two output tubes. Has anyone successfully built this with parallel output tubes and if so do you have a schematic?
My main concerns are current balance in the output tubes and the dissipation of the EF86 driver.
New circuit
Added cathode and grid resistors. Output impedance is now 0.13ohms.
Added cathode and grid resistors. Output impedance is now 0.13ohms.
An externally hosted image should be here but it was not working when we last tested it.
New Response Curve
The response is down -0.4dB at 3Hz and -0.6dB at 100kHz.
The response is down -0.4dB at 3Hz and -0.6dB at 100kHz.
An externally hosted image should be here but it was not working when we last tested it.
If you are running 4 6C33s, 70 watts sounds entirely plausible if the amp is not class A. I would increase the current-limiting resistors in the output section to 2 ohms though. They should be rated at 10 watts.
The amp is class A totem pole PP output. But it can still output 70 watts!
I actually did as you advised but I made the resistors 5 ohms. After running multiple simulations I found that 5 ohms will reduce the power out by about 10% and raise the output impedance from 0.12 to 0.15 ohms (approx.) which is no big deal.
I'll actually probably reduce them to 2 ohms (your suggestion) and raise the voltage on the rails to +/-160 (I will have no choice with the trannys I'm using) which should push the output over 70 watts.
I also added C5 and C6 to control the bandwidth of the amp and tone down a slight high frequency resonance above the audio spectrum.
Here's the latest schematic:
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
FREQ Response
Here's the response curve for the simulation:
Here's the response curve for the simulation:
An externally hosted image should be here but it was not working when we last tested it.