This is great news. I built the 25W choke loaded source follower SIT amp a few months ago using THF-51S, Hammond 195T chokes and Modushop 5U/400 case, as I had these to hand. Without question, it is the most transparent and musical sounding amp I've ever heard in my audio system, driven by Aksa Lender preamp. However, a bit more power would be nice to have on occasions, so this new design has much appeal to me. I used perf board in the 25W amp with success, but the prospect of PCB's for the 45W version is enticing. Should a GB eventuate, please count me in.
Ben, many thanks for sharing your knowledge and efforts in developing these projects. I have learnt a great deal about the wonders of SIT's from yourself and many others following these conversations. As you've noted previously, there's no going back!
Ben, many thanks for sharing your knowledge and efforts in developing these projects. I have learnt a great deal about the wonders of SIT's from yourself and many others following these conversations. As you've noted previously, there's no going back!
R13, 220R across speaker output, 3W should do.
I have used wire wound. I don't know if you can hear any difference. Speakers are wire wound, although you can hear them.
I have also been looking at the input capacitor. It is quite conservative at 10uF, and 10uF can quite big if you want polypropylene. After some investigation 4uF should be enough when combined with the 10k Ohm gate resistor. That will allow a wider choice of capacitors that can fit onto the board.
I have used wire wound. I don't know if you can hear any difference. Speakers are wire wound, although you can hear them.
I have also been looking at the input capacitor. It is quite conservative at 10uF, and 10uF can quite big if you want polypropylene. After some investigation 4uF should be enough when combined with the 10k Ohm gate resistor. That will allow a wider choice of capacitors that can fit onto the board.
Ben Mah,
Thanks for the recommendation of the BIG GATOR - have one arriving tomorrow.
Earlier in the thread you mentioned the power requirements of this amplifier was 200 watts.
An academic question - my Singing Bush amplifiers require 160 watts.
Any idea why your circuit would require more power? My curiosity gets the best of me since I cannot figure these things out for myself.
Take care,
Thanks for the recommendation of the BIG GATOR - have one arriving tomorrow.
Earlier in the thread you mentioned the power requirements of this amplifier was 200 watts.
An academic question - my Singing Bush amplifiers require 160 watts.
Any idea why your circuit would require more power? My curiosity gets the best of me since I cannot figure these things out for myself.
Take care,
Last edited:
The power draw is approximately 200W, that is, 60 to 65V and 3.0A. As for the power supply I would provide at least double that. Your Singing Bush is probably running the same voltage but at a lower current, perhaps 2.5A.
I am running the amp at 3.0A because I found that it works well at 3.0A, and gives higher power output. I have measured 50W into 8R with my amps before the onset of clipping.
When I had the amp running as common source, the BAF2015 circuit which the Singing Bush is based on, I tried 3.0A but it was not a good operating point. Zen Mod subsequently determined that a lower current was better for distortion levels. However, the maximum power output is probably around 35W into 8R.
I am running the amp at 3.0A because I found that it works well at 3.0A, and gives higher power output. I have measured 50W into 8R with my amps before the onset of clipping.
When I had the amp running as common source, the BAF2015 circuit which the Singing Bush is based on, I tried 3.0A but it was not a good operating point. Zen Mod subsequently determined that a lower current was better for distortion levels. However, the maximum power output is probably around 35W into 8R.
I received the amplifier boards yesterday. I went with the cheapest shipping option and it was air mailed. I ordered the boards on October 31 and the boards arrived 10 days later.
This was the first time ordering boards so I received a new-customer discount. The cost for five each of two different boards was $28.02 total, shipped to my door. Shipping was probably only a few days slower than by courier, but without the hassle of dealing with extra courier customs clearing costs.
With the boards in hand, I can see that the footprint I chose for the 1/4W resistors are a bare minimum in size. I will tweak that and move a few things around to give the components a bit more room. It's not absolutely necessary as the components should fit, but on the tight side.
I am going to order some parts from Mouser and hope to have a trial run sometime next week. If it goes well, I will order another test set.
Meanwhile I have ordered a test set of the power supply boards. I won't be stuffing a board for testing since I don't need another power supply. The 80V high micro-Farad capacitors aren't cheap. I will test them by just checking for correct electrical continuity.
This was the first time ordering boards so I received a new-customer discount. The cost for five each of two different boards was $28.02 total, shipped to my door. Shipping was probably only a few days slower than by courier, but without the hassle of dealing with extra courier customs clearing costs.
With the boards in hand, I can see that the footprint I chose for the 1/4W resistors are a bare minimum in size. I will tweak that and move a few things around to give the components a bit more room. It's not absolutely necessary as the components should fit, but on the tight side.
I am going to order some parts from Mouser and hope to have a trial run sometime next week. If it goes well, I will order another test set.
Meanwhile I have ordered a test set of the power supply boards. I won't be stuffing a board for testing since I don't need another power supply. The 80V high micro-Farad capacitors aren't cheap. I will test them by just checking for correct electrical continuity.
Attachments
I can see you heatsink is one piece so it must be a 300mm (not 400 mm)?
I dream every night about 400 mm heatsinks
so chassis is a 5U 300mm not 400mm as mentioned at top?Apart from that a nice build!
I have not read the full thread so how much power do you burn in those THF-51 and how hot does the heatsinks get?
You're right, heatsinks are 5U 300mm; my mistake. Thanks re build. To fit it into the Modushop case, I made a two level chassis with AC wiring, Meanwell SMPS's and chokes on lower level with bias and input boards on upper level. Re power, I'm running the THF-51S's at the operating points suggested by Ben for 2SK180's. That is, 35V and 2.5A respectively so dissipation is around 87.5W per channel. Heatsinks don't exceeded about 60-63 degrees (measured by infrared thermometer), even on a warm day. The SIT's measure less, around 52-55 degrees. I haven't run power or distortion measurements as yet so I don't know precisely how much power its making, but it sounds great in my system. I've tried it at 38V and 3A and apart from producing more power and running hotter, it doesn't sound much different.I can see you heatsink is one piece so it must be a 300mm (not 400 mm)?
I dream every night about 400 mm heatsinks
Apart from that a nice build!
I have not read the full thread so how much power do you burn in those THF-51 and how hot does the heatsinks get?
There is something not right about those temperatures. I assume they are in Celsius. The heatsink measured hotter than the SIT does not make sense. I run my choke loaded 2SK180 at that dissipation (88W) with one SIT on a 300mmx210mmx40mm Modu heatsink and the heatsink does not get very hot. I have not measured the temperature but the fins only felt warm and I could keep my hand on them indefinitely, so l would say that the heatsink temperature was much lower than 50 degrees C. The 2SK180 was definitely much hotter.
The infrared thermometer reading is very suspect.
The infrared thermometer reading is very suspect.
OK, yes I also wondered about that but concluded that heatsink was correct that area is rather large and should be easy to "hit" with the gun and the measuring direct at SIT was done from an angle maybe so reading was not that precise.
But nice to get other readings. The plan is to increase bias until heat sink gets around 65 C or so?
Could be interesting to watch distortion at a given output level and see how much difference there is between high and very high bias.
But nice to get other readings. The plan is to increase bias until heat sink gets around 65 C or so?
Could be interesting to watch distortion at a given output level and see how much difference there is between high and very high bias.
65 C at the heatsink fins is too hot. 55 C is generally the accepted maximum, although 60 C may be doable. Nelson's temperature determination:
"Posted by Nelson Pass on 01-22-2003 06:18 AM:
Well let's see if your temperature calibration nomenclature
matches mine:
Blimey hot is 10 seconds hands on = 45 deg C.
Crikey hot is 5 seconds = 50 deg C.
Bloody hot is 2 seconds = 55 deg C.
X*?@! is 60 deg C.
All of these are within an acceptable range, although
X*?@! is resrved for the more mature constructor."
"Posted by Nelson Pass on 01-22-2003 06:18 AM:
Well let's see if your temperature calibration nomenclature
matches mine:
Blimey hot is 10 seconds hands on = 45 deg C.
Crikey hot is 5 seconds = 50 deg C.
Bloody hot is 2 seconds = 55 deg C.
X*?@! is 60 deg C.
All of these are within an acceptable range, although
X*?@! is resrved for the more mature constructor."
Temps are degrees Celsius. I have been wondering for some time about the accuracy of my infrared thermometer. It is quite old and seen a lot of action. Your comments tend to confirm my suspicions that it is not reporting accurate temperatures. I can hold my hand on the hottest part of the heatsinks for more than 10 seconds, so based on the temperature calibration nomenclature above, the actual temp is probably not exceeding about 50-55 degrees. This would be consistent with your observations at the same dissipation on the same sized heatsink. I think I need a new thermometer. Thanks for your comments.
