I think the case would work fine as a complete amp.You can get a couple of inches spacing from the power boards to the amp boards mounted on the heatsinks, which is a decent amount.. and those heatsinks would be a shame to waste..
Last edited:
OK Mark,
I'll wait for the boards and see if I can fit them.
Sonic signatures of diode bridges.
Because I had a couple of diode bridges left over (36MB20A) from the power supply side recently I decided to put them in my trusted Exposure 18 mono-blocks, replacing the GBPC2504 units. On the face of it my reasoning behind this was simply that replacing old with new had to be a good idea but really I just wanted to hear if there was a difference. There surely was!
The Exposures give a warm yet detailed analogue sound with good mid band presence. Perhaps I hoped for a little more detail with the 36MB20A replacement but after a week of listening I wasn’t getting it. Upper mid-band and treble were being accentuated and there was a loss of mid-band presence. Piano notes had a brittle quality as of wooden block being struck and the bass had lost some of its weight. I had made a big mistake but to my delight the GBPC units are still being made—in China now not Malaysia—so I have just put a couple of fresh pairs back in the amps and they sound great again.
I was going to use the 36MB for the Nelson design but I have ordered some of the GBPC instead now. Little things do make a big difference.
Regards,
Chris
I'll wait for the boards and see if I can fit them.
Sonic signatures of diode bridges.
Because I had a couple of diode bridges left over (36MB20A) from the power supply side recently I decided to put them in my trusted Exposure 18 mono-blocks, replacing the GBPC2504 units. On the face of it my reasoning behind this was simply that replacing old with new had to be a good idea but really I just wanted to hear if there was a difference. There surely was!
The Exposures give a warm yet detailed analogue sound with good mid band presence. Perhaps I hoped for a little more detail with the 36MB20A replacement but after a week of listening I wasn’t getting it. Upper mid-band and treble were being accentuated and there was a loss of mid-band presence. Piano notes had a brittle quality as of wooden block being struck and the bass had lost some of its weight. I had made a big mistake but to my delight the GBPC units are still being made—in China now not Malaysia—so I have just put a couple of fresh pairs back in the amps and they sound great again.
I was going to use the 36MB for the Nelson design but I have ordered some of the GBPC instead now. Little things do make a big difference.
Regards,
Chris
Further testing of power supply.
Along the lines of some contributors here I have connected 4 X 12V Halogens across the output of one of the two power supplies, which will supply the finished amp. As each of these Halogens is 35W I am running this power supply at about 140W.
I’ve noticed the four 3W, 1ohm resistors are getting too hot to touch after a few minutes (the Schottkys haven’t had time to warm up).
So is this what I should expect and in fact does connecting 140W of Halogens across the power supply realistically reflect what the finished amp will draw. At present each of these power supplies will supply three pairs of Mosfets, not the six as outlined in the original BA-2 design but who knows I might add some extra at a later date.
Regards,
Chris
Along the lines of some contributors here I have connected 4 X 12V Halogens across the output of one of the two power supplies, which will supply the finished amp. As each of these Halogens is 35W I am running this power supply at about 140W.
I’ve noticed the four 3W, 1ohm resistors are getting too hot to touch after a few minutes (the Schottkys haven’t had time to warm up).
So is this what I should expect and in fact does connecting 140W of Halogens across the power supply realistically reflect what the finished amp will draw. At present each of these power supplies will supply three pairs of Mosfets, not the six as outlined in the original BA-2 design but who knows I might add some extra at a later date.
Regards,
Chris
No !
attaching a PTC to the output as a load gives no useful indication of what your PSU is capable of doing.
The filament in an incandescent light bulb is a PTC.
The cold resistance is very different from a hot resistance and very different again from a warm resistance.
You must use a load that has measurable and predictable resistance over the temperature range that the test will force it to have. If you start with 10.37 ohms cold and you know it is 10.47 ohms during the test then it is a useful load.
attaching a PTC to the output as a load gives no useful indication of what your PSU is capable of doing.
The filament in an incandescent light bulb is a PTC.
The cold resistance is very different from a hot resistance and very different again from a warm resistance.
You must use a load that has measurable and predictable resistance over the temperature range that the test will force it to have. If you start with 10.37 ohms cold and you know it is 10.47 ohms during the test then it is a useful load.
Resistors dissipating any significant portion of their power rating will get too hot to touch. The real question is how much power are they dissipating? Assuming that you have 4 1r in parallel, netting 0r25, multiply the voltage across them by 4 to determine the current, then multiply that by the voltage again to determine the total dissipation. If this is less than 6W you are OK - as long as the resistors have some room between them for air circulation.
As Andrew said, with a light bulb you really have no idea what the PSU is delivering. A known resistance makes it easier because you can calculate your power from a single measurement. However, with the current determined above and measured voltage across your bulbs you can determine its current operating point.
As Andrew said, with a light bulb you really have no idea what the PSU is delivering. A known resistance makes it easier because you can calculate your power from a single measurement. However, with the current determined above and measured voltage across your bulbs you can determine its current operating point.
...from the BA2 manual
...from the manual
Most people uses those power resistor with heat sink.
I believed whatever Power Ratings the manufacturer stamp on their lamps. Because they have the manufacturing processes and expensive equipment to test it. If they say it will draw 35W, I stick to it. Doesn't matter what it's molecular composition or whatever, the fact is, it draws it's rated power or very close to it. You can verify the power draw by measuring as suggested on previous post. It is a very good load because it's a resitive load, dissipate by converting heat to light and is cheap. An ideal but very costly alternative resitive load is a laboratory Rheostat, if you want to be precise on your measurements.
...from the manual
We are testing the psu for MAXIMUM bias current draw which is 3 amps per voltage rail. Connecting two 12V 35W lamp in series for each rail is a good way to do it, it will give you a good 'indication' of what your psu are capable of. You can always reduce the bias if its becomes too hot for your h/sink.
Most people uses those power resistor with heat sink.
Hi Backbones,
I'll be using one of these . Resistor,metal cased,power,25W,5%,0.22R, and will be mounted on a heatsink or on a ground plane on top of my capacitor bank.
Arcol | Passives | Resistors | High Power | Aluminium Housed, Wirewound - 5 to 600W |HS25 R22 J
Btw your current draw is looking good!!
Regards,
ReytNZ
I'll be using one of these . Resistor,metal cased,power,25W,5%,0.22R, and will be mounted on a heatsink or on a ground plane on top of my capacitor bank.
Arcol | Passives | Resistors | High Power | Aluminium Housed, Wirewound - 5 to 600W |HS25 R22 J
Btw your current draw is looking good!!
Regards,
ReytNZ
You get the same filtering effect. Watch the derating factor if you don't mount the 25W resistor to a proper heat sink, though. For example, a Vishay Dale RH025 must be derated by at least 50% if you don't mount it. Mounting it to a reasonably thick chassis is enough, but you still must derate it slightly.
With 4 x 3W resistors, you spend a dollar instead of 6 and don't need to drill and tap mounting holes, etc. I'm cheap and lazy, you know which way I'd go.
With 4 x 3W resistors, you spend a dollar instead of 6 and don't need to drill and tap mounting holes, etc. I'm cheap and lazy, you know which way I'd go.
Derating refers to power handling. For example see http://www.mouser.com/catalog/specsheets/rhnh.pdf A 25W resistor would be good for unmounted operation.
Heat sinking
I´ve found a couple of nice heat sinks for a reasonable price but I´m not sure they will suffice. What is your take on a pair of these, are they sufficient or do I probably need larger ones?
12*4*1.56 in
kind regards
/Forsman
I´ve found a couple of nice heat sinks for a reasonable price but I´m not sure they will suffice. What is your take on a pair of these, are they sufficient or do I probably need larger ones?
12*4*1.56 in
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
kind regards
/Forsman
