• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

ST-70 Tube Amplifier DIY Kit

W

wdecho

Active Member

I bought one. Antek actually offered a different one first, which I bought and recently built with toroidal iron.


I ran into some problems with the first one which is documented on diyaudio.com in the tube section but finally got it to run after purchasing a better PS transformer. Still could not make it perform like a 35 watt per channel ST-70. Right now I have it putting out around 25 watts one channel at clipping. As far as I am concerned the boards with this kit are nothing but trouble. I could not get it to run with the supplied driver board. PS board is OK. Might have to do with the inadequate PS transformer included with the kit when I tried to use it. No documentation or voltages on the schematic. I have my doubts the designer ever built the amplifier with the supplied PS transformer. I bought a larger PS transformer and the VTA 12AU7 driver board and finally after weeks made a successful PP EL-34 amplifier. I would not recommend the toroidal kit. Sounds good with good numbers now during testing but it was a struggle for a hobbyist like myself to get it to run.

The new kit appears to be using the same iron as the VTA Dynaco ST-70. I hope so. If so it will be a good deal.
 
Last edited:
Register to hide this ad
I should have made the heading, Antek ST-70 Tube Amplifier DIY Kit.

Perhaps a moderator could change it for me and add Antek.
 
I also noticed the new Antek ST-70 clone with non-toroidal transformers. Please let us know how this one goes.

Hope you are right, but I will be surprised if the OT’s are the same U.S. wound magnetics as VTA uses at this pricepoint, and since Antek is an importer of Chinese goods. I have used many of their power transformers and had no issues if used within specs.
 
I have used Antek transformers for years myself, more than a dozen builds, without any problems until the toroidal kit. Even after upgrading to the AS-3TD360 - 360V Tube Amp Transformer I am having trouble biasing the output tubes to 40ma apiece like the tubes4hifi St-70 with the 12AU7 driver board. Could be a problem in my build but my thoughts is the PS transformer not up to the task of running 4 EL34 tubes at 40ma. There is a lot of discussion of the build at diyaudio.com if anyone interested. Right now it is biased at 30ma per output tube and is stable making around 25watts at clipping. Hey it is stable, running, testing good with my limited hobbyist equipment. Three scopes analog ones and one Tektronics digital with FFT along with one of those Chinese $100 when bought signal generators using the Arta distortion software program. Sounds good too with a very clean sound.

Unless I am mistaken tubes4hifi does not wind transformers but rather buys them from a source. My hope is Antek is buying from the same source with these EI transformers in the new kit offering. Repeating myself but unless I am wrong Antek does not wind transformers either. Their toroidals are made in China. They are the distributors, middle man. In the description they are calling the iron the same as the tubes4hifi ones. Time will tell. I hope I can make a decent 35 watt per channel PP El-34 amplifier with this kit. Not that I need 35watts. I have Klipsch LaScala's and only need a few watts. My experience with these speakers is 5 watts plenty with a tube amp. Most of tube builds are SE making around 5 watts average. I actually prefer the SE sound. Just building for the sake of building.
 
I've used various Toroidal power transformers without issue in amplifier and preamp builds without issue. I built a VTA ST-70 kit and I do like there driver board. I've thought about trying Toroidal OPT's in an amp build, but like everything else you get what you pay for. There does seem to be better made Toroidal OPT's than what Antek produces. I've seen some very nice amp builds using them including a 300B SET amp where these other brands of more expensive Toroidals were used.
 
wdecho said:
I have used Antek transformers for years myself, more than a dozen builds, without any problems until the toroidal kit. Even after upgrading to the AS-3TD360 - 360V Tube Amp Transformer I am having trouble biasing the output tubes to 40ma apiece like the tubes4hifi St-70 with the 12AU7 driver board. Could be a problem in my build but my thoughts is the PS transformer not up to the task of running 4 EL34 tubes at 40ma.
Click to expand...
These are fixed bias? Presuming so, is it a matter of bias voltage available rather than what seems like an implication of transformer current capability?
 
Myself and others on the other forum did some calculations of the transformer and we decided it was not up to the task. If enough current, not enough voltage. If enough voltage not enough current to run 4 EL-34 tubes at Dyanco specifications. This is using the transformer data sheet. The transformer is doing all it is supposed to according to the spec sheet. If I had been more knowledgeable on reading a PS transformer data sheet I could done the calculations myself. Nothing wrong with the transformer just not up to the task. Could be wrong.
 
I had a conversation with an employee at Antek about this transformer and below is his response. At his figures the voltage is 361.3 volts. The voltage on the plates of a Dynaco build is 410Vs.


This is not the maximum power rating of the transformer for AS-3TD360. The data shown are test measurements used to determine the transformer’s effective output impedance (Z). The output voltage under load can be estimated using:
Vout = Vno-load − I × Z

where:

  • Vout = output voltage under load
  • Vno-load = output voltage with no load
  • I = load current
  • Z = effective output impedance of the transformer
  • Z = ΔV / ΔI, calculated from the change in output voltage divided by the change in load current. This impedance value can be used to estimate how much the output voltage will drop as the load current increases.

  • For example, using the data from the test sheet:

    ΔV = 378 V − 368 V = 10 V
    ΔI = 0.18 A
    Therefore, Z = ΔV / ΔI = 10 V / 0.18 A ≈ 55.5 Ω

    If the load current increases to 0.3 A, we can estimate the output voltage as:

    Vout = 378 V − (0.3 A × 55.5 Ω) ≈ 361.3 V

    This is an estimate based on the assumption that the voltage drop is approximately linear with load current. It should not be interpreted as the transformer's maximum current or power rating.
 
The voltage on the plates of the EL34's in my build is 466Vs with a current of 30ma per tube and if my math is correct I am running a 25watt tube at 13.98 watts. About 1/2 or 50% of max. I typically run my SE EL-34 tubes at 20 watts, roughly 80%. Hey not a bad build, just not what I expected when the kit is called a ST-70. I was expecting to get 35 watts per channel. If it had been named a 25watt PP El-34 kit I would not be complaining one bit.

I had never used toroidal OPT's in a build before and was skeptical. I was always told never to do so but I am more than happy with the results using Antek's PP toroidal OPT's. Still skeptical of toroidal SE OPT's. They claim making gap now where the core will not saturate. Until I try one I cannot say. In a PP transformer the current is always changing between the two plates so no saturation. Not so in a SE.

Anyone here tried toroidal OPT's in a SE build?
 
Fixed bias = negative supply to the grid, adjustable or not is another matter. The other option is cathode bias. If its using one of those VTA driver boards with the trim pot per-tube, its fixed bias.

Too-negative a grid supply, or too-low plate voltage make for low current. Can't do too much about the plate supply but its not real difficult to tweak the bias supply to be less-negative and bump the current. The lower voltage will obviously make less power but having it under-biased increases the distortion a fair bit and thats probably more noticeable than the drop in output power.

if the transformer isn't capable of delivering enough current though, winding up the tubes will suck down the voltage further. Its going to do that when playing anyway though, so a wimpy power transformer is really not a good thing in an amplifier.
 
gadget73 said:
Fixed bias = negative supply to the grid, adjustable or not is another matter. The other option is cathode bias. If its using one of those VTA driver boards with the trim pot per-tube, its fixed bias.

Too-negative a grid supply, or too-low plate voltage make for low current. Can't do too much about the plate supply but its not real difficult to tweak the bias supply to be less-negative and bump the current. The lower voltage will obviously make less power but having it under-biased increases the distortion a fair bit and thats probably more noticeable than the drop in output power.

if the transformer isn't capable of delivering enough current though, winding up the tubes will suck down the voltage further. Its going to do that when playing anyway though, so a wimpy power transformer is really not a good thing in an amplifier.
Click to expand...
Thanks for reply. I was under the impression cathode bias is what one calls fixed bias. All my other builds have been cathode biased. I changed the bias by changing the cathode resistor.

The negative supply is working fine and my voltage is 466 volts on the plates of the tubes biased at 30ma each. It is when I increase the bias towards 40ma voltage crashes. My thoughts, it has to be the transformer not capable of driving the output tubes at 40ma. Works fine at 30ma with really good s/n ratio and distortion with noise numbers that are really good on Arta. I cannot hear or find any problem with scope or Arta distortion software with the amplifier when run at 30ma per tube with the equipment I have. A few days ago I noticed some noise on the top of the positive sine wave starting at7 watts and lowered the voltage to where it is supposed to be at on the VTA driver board, 375V's, and took care of that problem. Below 7 watts the noise was not there. The voltage before was somewhat high. I did not think it would be a problem. I never would have heard it with my LaScala speakers but wanted to fix it anyway. With over 30 working amplifiers I am just playing.

The late Roger Modjeski ran one of his EL84 amplifiers at 700Vs on the plates with something like 15ma on the tubes just from memory. A 12 watt tube. A higher voltage is where you get the power.
 
Last edited:
Something weird is going on in this build with your AS-3TD360. Let’s try ti figure it out.
wdecho said:
At his figures the voltage is 361.3 volts. The voltage on the plates of a Dynaco build is 410Vs.
Click to expand...
Under what conditions do you get 410 volts on the plates? The Antek specified 361.3 volts (at 300 ma) should yield close to 500 Vdc (with SS rectifiers) to the OT center tap. You say you are pulling only 4x30 ma for EL 34s plus ~25 ma for everything else at idle (no input signal). That is only 145 ma! Your Power trannie should laugh at this. What HV are you feeding the OT CT. It should be fairly close to EL34 466vdc anode volts, unless there are huge losses in the OT.

Could you provide some voltage measurements of your power supply chain starting with mains AC voltage, with tubes and without, please. That might spark some iseas.

Also, have you checked for high frequency oscilations? You may also check your PS capacitors for leakage. Aby of them getting warm?
 
Last edited:
^^^
Yeah, real quick plug into PSUD suggests that trafo with caps and choke like ST-70, albeit SS FW rectifier, should be 475-ish volts with 175mA load.

But, it doesn't look like it has a choke and caps are probably different.

If you @wdecho could provide at least the PS component values and arrangement up to first node (b+ tap off point) i can plug them in quick.

Also, how are you measuring the current?
 
Last edited:
This is what a guy on the other forum had to say.

Antek AS-3TD360 datasheet shows the B+ rated at 180 ma. before rectification,so probably around 140 ma. or so after.

I was trying to run it at 160ma. Makes sense to me but to be honest a lot of this over my head. Simple ohms law is about as much as I can comprehend. I do not believe it is in the power supply section. Very simple supply that I have built many times with multiple 100uf caps on the driver board to clean up the signal.

Very possible high frequency oscillation and I have thought of that but I do not know the cause. My first thought would be the toroidal output transformers causing the problem but there is no simple solution to solve it if that is the problem. It does act like that may be the problem because I can adjust the tubes slowly in steps to 40ma until the last tube. When I get close to 40ma with it the current suddenly jumps to a volt or more and the other tubes bias wanders all over the place too.

Many thanks for responding. I understand it is some very smart talented people on this forum.
 
On Antek's website in the kits listed there, some of the information specifically mentions potential HF oscillation with toroidal outputs and lists several things that might help. If that's actually it, of course.
 
After giving it some thought the problem could very well be oscillation. Tomorrow I am going to add some more resistance to the feedback circuit. Too much signal going back to the input could very well make the amplifier oscillate. The feedback circuit is on the VTA driver board and is set up for signal taken off the 16ohm tap on the Dynaco OPT's. These toroidal OPT's do not have a 16ohm tap and the signal is now being taken off the 8ohm tap. Spantou has given me something else to consider other than the PS transformer. Now remember this is not the transformer that comes with the kit. This is a 300VA transformer I bought after trying to make the supplied driver board work with the 200VA transformer. The one that comes with the kit is a 200VA one. All kind of issues happened with that build try. Gave up and ordered the VTA driver board and a larger transformer.

I am of the opinion it is an oscillation problem the way it acts. If the PS transformer was the problem the voltages would drop and the transformer would get hot rather than bias current being unstable when pushed. I am adding some more resistance in the morning decreasing the amount of feedback and post the result. Not too hard to do.
 
wdecho said:
This is what a guy on the other forum had to say.

Antek AS-3TD360 datasheet shows the B+ rated at 180 ma. before rectification,so probably around 140 ma. or so after.
Click to expand...
“The guy on the other site” is perhaps wrong. As explained by the Antek fellow you communicated with (post #9) 180 ma represents the increase in current (from 0 ma) that would result in a drop of 10 volts from 378 to 368 Vac. It is NOT the rated current capacity of the PT. Your AS-3TD360 is rated for at least 400 ma before rectification at 360 Vac, and possibly 2x400 ma.

Bottom line; your PT should have no trouble pushing your amp to full power perhaps limited by the OTs compared to a classic ST70. That was why I felt that something weird is going on.

Could I ask why you are targeting a B+ well above what Dynaco specified for the ST70. The classic ST70 had a B+ after the choke of only 415 Vdc (430 Vdc before)
 
Last edited:
Is there a schematic you could show, or if proprietary could you give a detailed description of you power supply.
 
wdecho said:
The late Roger Modjeski ran one of his EL84 amplifiers at 700Vs on the plates with something like 15ma on the tubes just from memory. A 12 watt tube. A higher voltage is where you get the power.
Click to expand...
Thats one way, watts are simply volts times amps, an EL84 can only deliver so much current so voltage is the only thing left to work with. Current flow in a tube depends quite a bit on cathode and plate size, and by extension heater power and an EL84 is not exactly a hefty tube.

Running an EL84 at 700 volts seems pretty insane to me though. There is an amount you can push the ratings but at some point physics will start to take over. The elements are spaced awful darn close inside, and those pins are tight on the socket.
 
You must log in or register to reply here.
Back
Top Bottom