If you go with V- to the source and ground the drain, and if it is done correctly, you do not have +30V at ground. Ground is ground, and there would be -30V in the circuit relative to ground.
The rack mounting convention makes sense. Large footprint flat amps are very common in the DIY community, for clear reasons. I think I'm hooked on the vertical design for at least a few more projects. It's also a great format to show off the centerpiece parts, such as the SITs. Personally, I love exposing things like tubes, transformers, and can transistors on the exterior--it's a chance to show off the special sauce. I know it's polarizing to some (or at the very least they tend to be more vocal about it XD).
My roommate bought the Marantz and his Bose 901s as his "audiophile starter kit". No turntables or vinyl... We are strictly a R2R household 😆
I've amassed the DXF files for the flat pattern cuts (aluminum and hardboard) and the full STEP file CAD assembly of the amp. The hardware is mostly specialty product but it's all metric standard dimensions. M3 for panel attachments and PCB hardware. M4 in some limited situational cases. M6 for the chassis structural connections. and M8 for the wood-to-chassis fixation. All the holes that need tapped are precut via laser, and manually tapped by me after the fact. I used Send-Cut-Send for all the metalwork, laser cuttings, and anodizing. The wood box was hand built by me.
Things I would have done differently:
- More lenient location and size tolerances on all manually drilled holes
- Better thermal paste coverage on SIT sink-to-front plate interface
- Fix small error on PSU board (forgot to connect a spade connector to GND)
- Quieter fans (splurge for noctua 60mm)
- PMW control for fans instead of voltage control
- Change SIT mounting screws and drain attachment (went through a few iterations and I'm confident what I finished with will suffice, but it doesn't fully live up to my standards for quality high-current electrical connections).
- Some holes got drilled off-center here and there and had to be widened/fudged. Not really noticeable but it bothers me nonetheless.
What I intended to say was that if I attached the DAC's pin 1 to the same leg as the load resistor there would be 30+ volts going to the DAC's pin 1. Which would be really dumb. Still kicking myself for not seeing the obvious. I should be used to that by now ...
Made good headway in the conversion of the amplifier number one yesterday.
It always takes longer to convert than to build.
Plus I got confused about how to handle the raw supply with ZM's CAP BANK boards and spent much more time with those that I should have. Would have been simple with raw boards but I did not want to turn the capacitors around. Turns out to be just fine. ZM recommends wiring the two boards together in a way that is not what the dilletante would expect but is unquestionably correct once one see the logic of it.
What do you think Ben Mah? If I use the transformer to convert the DAC's balanced output to single ended would you expect there will be a need for a connection between the DAC's pin one and the ground of TDV51?
I am always confused about how to use transformers for this.
I know when I tried a different transfomer with the SINGING BUSH I got a distorted signal wihtout connecting DAC pin one to ground but I did not get any additional volume whcih confused me. It was a 1:1 transformer - the CINEMAG in the round can.
Thsi time I will be using a CMOQ2 - a bigger core but unshielded; still 1:1.
Thanks and take care,
My understanding of pin1 is that it is connected to the shield of the cable, and generally the shield should be connected to chassis ground (not amplifier circuit ground). So if you want to connect pin 1 at the amplifier, it should be connected directly to the amplifier chassis. You can also try no connection, so that the shield is connected to the DAC chassis only. But usually the best is for the shield to be grounded at both ends. In a short cable run, there probably would not be much difference if any.
In your case of distortion with your DAC connected without pin 1 connected, I don't know what the issue was. You can look into your DAC to see where pin 1 is connected inside.
If your transformer is 1:1, there is no voltage amplification so there should be no increase in volume if you compare balanced output to balanced output converted to single ended with a transformer.
In your case of distortion with your DAC connected without pin 1 connected, I don't know what the issue was. You can look into your DAC to see where pin 1 is connected inside.
If your transformer is 1:1, there is no voltage amplification so there should be no increase in volume if you compare balanced output to balanced output converted to single ended with a transformer.
I have not opened the DAC so I am not sure what it is doing with pin 1.
All of this voltage output stuff has me confused which you are quite used to wiht me!
Single ended output voltage is specified as 2.5 volts and balanced 5 volts so I thought i would see an increase of three dBs.
My amplifiers have no chassis per se - just a DYNACO 400 heatsink laying atop a birch plywood "plate" to form a tunnel for the fan to blow through.
Of course, I will know soon when I give it a try but always like to find out what NOT to do.
Thanks, Ben Mah
All of this voltage output stuff has me confused which you are quite used to wiht me!
Single ended output voltage is specified as 2.5 volts and balanced 5 volts so I thought i would see an increase of three dBs.
My amplifiers have no chassis per se - just a DYNACO 400 heatsink laying atop a birch plywood "plate" to form a tunnel for the fan to blow through.
Of course, I will know soon when I give it a try but always like to find out what NOT to do.
Thanks, Ben Mah
I don't know why your first attempt with the transformer did not work out. But you can try again and see what you get this time.
Here is someone's (from another forum) transformer XLR to RCA: Balanced XLR to Single Ended RCA
Here is someone's (from another forum) transformer XLR to RCA: Balanced XLR to Single Ended RCA
Thanks, Ben Mah,
I had seen that.
I am wondering if there is something unusual about the ground with the SINGING BUSH?
Do I remember something about it being a derived ground? Not that I know what that is. I remember ZM say9ing something about the ground arrangement.
As you said there is no substitute for trying and hearing what happens.
Take care,
I had seen that.
I am wondering if there is something unusual about the ground with the SINGING BUSH?
Do I remember something about it being a derived ground? Not that I know what that is. I remember ZM say9ing something about the ground arrangement.
As you said there is no substitute for trying and hearing what happens.
Take care,
In making my conversion to TDV from SINGING BUSH I have to change the power transformer.
I follow ZM's advice of wiring secondarties in series so using ANTEK transformers I have a choice of 30 v0lts (2 x 15) or 24 voltsd (2 x 12) - 24 is much too low and 30 using a CLC filter is too high.
I then tried placing a resistor after the rectifiers (Power Supply Designer II) and I lose about five volts and gain some heat from the resistors. I will use 4 -12 watts 3R OHMITE Brown Devils. They should be able to take the abuse.
Placing the resistor after the rectifier is more effective than placing it after the choke.
I figured 41 volts B- was too much. 36 volts seems more reasonable.
I follow ZM's advice of wiring secondarties in series so using ANTEK transformers I have a choice of 30 v0lts (2 x 15) or 24 voltsd (2 x 12) - 24 is much too low and 30 using a CLC filter is too high.
I then tried placing a resistor after the rectifiers (Power Supply Designer II) and I lose about five volts and gain some heat from the resistors. I will use 4 -12 watts 3R OHMITE Brown Devils. They should be able to take the abuse.
Placing the resistor after the rectifier is more effective than placing it after the choke.
I figured 41 volts B- was too much. 36 volts seems more reasonable.
I have Antek 30VAC power transformers in my Tokin 2SK180 Hammond 193V follower monoblocks. The CLC power supply with Hammond 159ZJ choke outputs about 38VDC under load. I run Iq at around 2.5A. Sounds great, so 30VAC power transformer is not a probem.
Ben, that is about 35V x 2.5A, or 88W on one device. Probably okay for the big Tokins. I would be wary of pushing the 2087C that high.
Rick, you should try to match the voltage and the bias current. For example, at 2A bias current, the peak power delivered to an 8 ohm load is 32W and the voltage swing needed is 16V peak to peak. So, anything beyond 20V will not be used effectively. At 2.8A bias and 8 ohm load, you will be fine with 25-30V.
Of course, higher voltage won’t hurt and maybe there are better operating points out there, but it means bigger hardware all around.
Rick, you should try to match the voltage and the bias current. For example, at 2A bias current, the peak power delivered to an 8 ohm load is 32W and the voltage swing needed is 16V peak to peak. So, anything beyond 20V will not be used effectively. At 2.8A bias and 8 ohm load, you will be fine with 25-30V.
Of course, higher voltage won’t hurt and maybe there are better operating points out there, but it means bigger hardware all around.
Yes - I am using the THF51.
Being the eternal dilletante I have tried to stay as close to what others, most notably Ben Mah, have done,
I have yet to connect the supply to the circuit but from the sims it looks like 36 volts under load around 2.5 amps.
I do know what the unloaded supply actaully does and have worked backewards with the simulator.
So for those who know so much more than I do - If I used 41 volts B- what Iq would I shoot for? I figure it would be less? Would it be dangerous? I think I have plenty of heatsink and it is fan cooled.
(My line voltage is 122 volts and the ANTEK transformers are rated for 115 so I am seeing 33.3 ACV)
I would love to dispense with the input resistors if I could.
Willing toi try anything.
Being the eternal dilletante I have tried to stay as close to what others, most notably Ben Mah, have done,
I have yet to connect the supply to the circuit but from the sims it looks like 36 volts under load around 2.5 amps.
I do know what the unloaded supply actaully does and have worked backewards with the simulator.
So for those who know so much more than I do - If I used 41 volts B- what Iq would I shoot for? I figure it would be less? Would it be dangerous? I think I have plenty of heatsink and it is fan cooled.
(My line voltage is 122 volts and the ANTEK transformers are rated for 115 so I am seeing 33.3 ACV)
I would love to dispense with the input resistors if I could.
Willing toi try anything.
I have built choke loaded 2SK180 monoblocks using Antek AS-3216, 300VA with 2x16V in series and get 40V at 1.5A with a CLC arrangement. I use bias supplies so I run the 2SK180s at about 40V, 1.5A. Currently, I'm working on 2SK182ES amps with 60V power supplies, source biased, with Vds of about 55V, 1.6A.
So, I think that a THF-51 would not be that much different. I find that higher voltages and lower current works well with these big Tokins in choke loaded follower mode.
My build is here:https://www.diyaudio.com/community/threads/mo-chokes-building-2sj28-2sk180-monoblocks.370479/
So, I think that a THF-51 would not be that much different. I find that higher voltages and lower current works well with these big Tokins in choke loaded follower mode.
My build is here:https://www.diyaudio.com/community/threads/mo-chokes-building-2sj28-2sk180-monoblocks.370479/
The THF-51S is a 400W device. With adequate heat dissipation, 2.5A or even a bit more should not be a problem.
The 193V has 1.0 Ohm DCR so that will drop the voltage by 2.5V. If you need a source resistor, that will drop the voltage some more. The power dissipation of the SIT is Vds x Iq. So even if the loaded power supply voltage is 41V (it may be lower) and Iq is 2.5V, the power dissipation would be (41V-2.5V) x 2.5A = 96W. Usually dissipation should be kept to less than 1/4 to 1/3 of specified maximum dissipation, which is usually specified at 25 degrees C. So with adequate heatsink/cooling, 2.5A is safe.
Once you have it up and running , you can do some distortion measurements and vary the Iq to find an Iq that you prefer.
For initial Vgs and choice of source resistor in conjunction with the choke for source bias voltage, it would be helpful if you know the Vgs of the Tokins when they were in the Singing Bush. If the Vgs is not known, then you will have to guess and choose a resistor value that is on the high side to be safe. Perhaps something in the 1 to 2 Ohm range, and at least 10W if 1 Ohm and 20W if 2 Ohm. Monitor the Iq by measuring the voltage drop across the 193V. With a DCR of 1 Ohm, voltage drop equals current in Amperes. Turn off immediately if current is too high. First powerup with a DBT to check for shorts.
The 193V has 1.0 Ohm DCR so that will drop the voltage by 2.5V. If you need a source resistor, that will drop the voltage some more. The power dissipation of the SIT is Vds x Iq. So even if the loaded power supply voltage is 41V (it may be lower) and Iq is 2.5V, the power dissipation would be (41V-2.5V) x 2.5A = 96W. Usually dissipation should be kept to less than 1/4 to 1/3 of specified maximum dissipation, which is usually specified at 25 degrees C. So with adequate heatsink/cooling, 2.5A is safe.
Once you have it up and running , you can do some distortion measurements and vary the Iq to find an Iq that you prefer.
For initial Vgs and choice of source resistor in conjunction with the choke for source bias voltage, it would be helpful if you know the Vgs of the Tokins when they were in the Singing Bush. If the Vgs is not known, then you will have to guess and choose a resistor value that is on the high side to be safe. Perhaps something in the 1 to 2 Ohm range, and at least 10W if 1 Ohm and 20W if 2 Ohm. Monitor the Iq by measuring the voltage drop across the 193V. With a DCR of 1 Ohm, voltage drop equals current in Amperes. Turn off immediately if current is too high. First powerup with a DBT to check for shorts.
These choke-loaded amps with the bias provided on the source pin are relatively safe because the bias does not run away. And yet, I have a naughty 2087 that likes to run like a horse given the chance. So, yeah, bring them up one channel at a time, monitor the Iq and keep your hand on the off switch. I like to use a Variac.
Waves of gratitude to all of you.
I am using the LUNDAHL LL2733 which has a slightly lower rated resistance than the 193V.
With a measured 10.1R resistor in series with the LL2733 - it seems like it is very close to the rated 0.85R. A little bit higher than that - using those red meters used by ZM.
In the SINGING BUSH Vgs was in the range as recommended by ZM - I am not sure exactly but is was between 35 and 36 volts for all four amplifiers..
I planned to begin with 1R - maybe I will add 0.5R in series to begin. Or would 2R be an even better place to start?
I am using the LUNDAHL LL2733 which has a slightly lower rated resistance than the 193V.
With a measured 10.1R resistor in series with the LL2733 - it seems like it is very close to the rated 0.85R. A little bit higher than that - using those red meters used by ZM.
In the SINGING BUSH Vgs was in the range as recommended by ZM - I am not sure exactly but is was between 35 and 36 volts for all four amplifiers..
I planned to begin with 1R - maybe I will add 0.5R in series to begin. Or would 2R be an even better place to start?
35V and 36V would be the Vds. The Vgs probably would be in the range of 2V to 4V, perhaps even lower or higher, for that Vds and Iq of about 2A.
As a start, if Vgs required is 3V for 2A, then R = 3V/2A = 1.5R. For choke DCR = 0.85R, additional R = 1.5R - 0.85R = 0.65R.
So 1R is probably a good start. That assumes a higher Vgs, which is erring on the side of safety. But monitor the current and shut down if it goes too high.
As a start, if Vgs required is 3V for 2A, then R = 3V/2A = 1.5R. For choke DCR = 0.85R, additional R = 1.5R - 0.85R = 0.65R.
So 1R is probably a good start. That assumes a higher Vgs, which is erring on the side of safety. But monitor the current and shut down if it goes too high.
I am always getting mixed up on those two measurements. Maybe I will get it right the day before I expire.
AS ALWAYS, thanks to you Ben Mah for your patience and kindness. You and ra7 are DIY audio saints.
Well, now to do the final connections and see what happens.
When it is running I will follow Ben Mah's instructions for measuring distortion. First I will listen to it.
AS ALWAYS, thanks to you Ben Mah for your patience and kindness. You and ra7 are DIY audio saints.
Well, now to do the final connections and see what happens.
When it is running I will follow Ben Mah's instructions for measuring distortion. First I will listen to it.