Terry,
That connectionon the on the terminal block should be a soldered connection. Should one lead of a resistor come loose the current draw will shoot up on the other OP devices. Also, fromr a look at your photos I believe you will need more heat sinking on the drivers as well. That was the reason for the board re-design. Other than that it looks great!
Mark
That looks correct Larry. The reason you didn't get any reading on ohms is because your meter does not have sufficient dc violtage present across the probes to bias the transistor's junctions on to get a reading. You will generally find that on diode test there is a higher voltage precent acrioss your probes. This is common with DMM's. With older analog meters like a Simpson 260 there is sufficient voltage in ohms to bias a junction for a reading. Your reading is 520 ohms which is right about what it should be in a good silicon junction.
Here is a link to a TO-3 basing diagram.......
TO-3 basing diagram
Mark
That connectionon the on the terminal block should be a soldered connection. Should one lead of a resistor come loose the current draw will shoot up on the other OP devices. Also, fromr a look at your photos I believe you will need more heat sinking on the drivers as well. That was the reason for the board re-design. Other than that it looks great!
Mark
That looks correct Larry. The reason you didn't get any reading on ohms is because your meter does not have sufficient dc violtage present across the probes to bias the transistor's junctions on to get a reading. You will generally find that on diode test there is a higher voltage precent acrioss your probes. This is common with DMM's. With older analog meters like a Simpson 260 there is sufficient voltage in ohms to bias a junction for a reading. Your reading is 520 ohms which is right about what it should be in a good silicon junction.
Here is a link to a TO-3 basing diagram.......
TO-3 basing diagram
Mark
OK, I have another question about the fan I have.
I tried it with two 150 ohm resistors in series and it still puts out too much air and is too noisy. I tried using one side of the secondaries (30VAC) but the fan barely turns. with the secondaries in series and using the 60VAC from the two outside legs it seems about right.
Here are the specs for the fan.
If I pull the power for the fan off of the outer legs ahead of the bridge, will I still have enough vA for the amp? The fan is rated at 0.21A
Thanks, Terry
I tried it with two 150 ohm resistors in series and it still puts out too much air and is too noisy. I tried using one side of the secondaries (30VAC) but the fan barely turns. with the secondaries in series and using the 60VAC from the two outside legs it seems about right.
Here are the specs for the fan.
If I pull the power for the fan off of the outer legs ahead of the bridge, will I still have enough vA for the amp? The fan is rated at 0.21A
Thanks, Terry
Mark A. Gulbrandsen said:
Hi Mark,
Thanks for the reply. The first thing I did this evening was to solder the emitter resistors to the terminal blocks as well as the feedback wires.
I suppose I could mount the drivers to the main sinks as well. Shouldn't be too much trouble to run a few more wires.
Blessings, Terry
The fan is rated for 120V so if you connect to 60V the draw will double but dont think you could have any problem of too much draw from the secondaries.
I dont have in mind what is the total VA from your trafo or how many output devices you have or even if your running class A all the way or type of filtering on the PSU or...
I dont have in mind what is the total VA from your trafo or how many output devices you have or even if your running class A all the way or type of filtering on the PSU or...
The problem is that its a bit clumsy mounting the three transistors from the board down under the board on the sink.... but it can certainly be done.
For my fans I use two 12 volt DC Pabst fans and I run them on about 5 to 6 volts or just slightly less than half speed, but also just fast enough speed to keep the sinks at 50C. I've also thought about a temperature servo loop for this as well but won't have time to mess with that for about a month. You cannot hear my fans unless your ear is right at the heatsinks...... this feat has amazed some friends on occasion that claimed the fans would be audible. ... keep in mind that I'm using two fans here and my sinks are a bit more spread out(flat backs,but your sink should also work very well. At any rate I use two simple LM337 adjustable voltage regulators fed from some extra windings on the trannys... one regulator for each fan... a cheap and reliable solution although the fans were about 5 bucks more expensive than the same AC version. You might try a simple AC wall type dimmer to vary the speed.... and there are also types of simple dimmers available that are designed to vary the speed of a motor.... but keep in mind that any of these usually use a triac whose firing phase angle us changed to adjust the output. Its possible that the fan might not like the resilting AC waveform.... but one of the type thats designed for motors woulod be safe to use.... or just aquire a small 5 amp surplus varic as it would also do the trick and would be quite inexpensive. I can get you the name end number of a place in town here that has some surplus small varics if you need it.
P.S. When this project is all done I will also need to hook that temperature servo on my air conditioning system to keep the room at a decent temp
.
Mark
For my fans I use two 12 volt DC Pabst fans and I run them on about 5 to 6 volts or just slightly less than half speed, but also just fast enough speed to keep the sinks at 50C. I've also thought about a temperature servo loop for this as well but won't have time to mess with that for about a month. You cannot hear my fans unless your ear is right at the heatsinks...... this feat has amazed some friends on occasion that claimed the fans would be audible. ... keep in mind that I'm using two fans here and my sinks are a bit more spread out(flat backs,but your sink should also work very well. At any rate I use two simple LM337 adjustable voltage regulators fed from some extra windings on the trannys... one regulator for each fan... a cheap and reliable solution although the fans were about 5 bucks more expensive than the same AC version. You might try a simple AC wall type dimmer to vary the speed.... and there are also types of simple dimmers available that are designed to vary the speed of a motor.... but keep in mind that any of these usually use a triac whose firing phase angle us changed to adjust the output. Its possible that the fan might not like the resilting AC waveform.... but one of the type thats designed for motors woulod be safe to use.... or just aquire a small 5 amp surplus varic as it would also do the trick and would be quite inexpensive. I can get you the name end number of a place in town here that has some surplus small varics if you need it.
P.S. When this project is all done I will also need to hook that temperature servo on my air conditioning system to keep the room at a decent temp
.Mark
apassgear said:
Hummm, I think you are distorting the law of physics Sir. If you halve the Voltage you halve the Power, hence the Current draw will be half. If the internal resistance of the AC motor coils remains constant, no amount of proclomation will double the current draw... sorry...
Terry, the current draw of the FAN across the 60VAC will less than 25VA.
Regards
Anthony
not to split hairs...
...but, with a resistive load, if you halve the voltage, it will halve the current and thereby quarter the power...
In any case the circuit Terry shows with the thermal switches is kind of cool...initially the thermal switches are open, the fan fed through the 2 resistors, as the temp rises, each switch in turn closes, shorting out the respective resistor making the fan spin faster.
As to the temp at which the switches should close, that is something you probably need to determine with a little trial and error. Finding the correct value of the resistors will also need some experimentation, the temps and resistor sizes are going to interact.
There are 12v fans for computers that have thermistors built in that can spin from 500-3000rpm based on the temp, Arctic cooling used to sell some quite inexpensive models, 80mm I think, and even at full speed they were pretty quiet, moved about 30cfm IIRC.
Stuart
...but, with a resistive load, if you halve the voltage, it will halve the current and thereby quarter the power...
In any case the circuit Terry shows with the thermal switches is kind of cool...initially the thermal switches are open, the fan fed through the 2 resistors, as the temp rises, each switch in turn closes, shorting out the respective resistor making the fan spin faster.
As to the temp at which the switches should close, that is something you probably need to determine with a little trial and error. Finding the correct value of the resistors will also need some experimentation, the temps and resistor sizes are going to interact.
There are 12v fans for computers that have thermistors built in that can spin from 500-3000rpm based on the temp, Arctic cooling used to sell some quite inexpensive models, 80mm I think, and even at full speed they were pretty quiet, moved about 30cfm IIRC.
Stuart
Terry,
the Fan on the picture you posted is a Sunon, and a mighty hefty one for audio purposes, at 107 CFM.
I am not going to boring dB stuff again, but on nominal voltage at 3050 rev its a noisy one.
The blades of the Sunon are shaped for displacement, not low noise.
The reason Krell used a Papst on their amplifiers at the time was because those were both the longest lasting and the least noisy.
Current Papst models go as low as 18 dB at full rev, do not need derating at all for audio purpose heatsink tunnels.
I doubt that you can get this one derated to an acceptable noise level and still push enough air.
(the blade pitch is optimised for the nominal rpm, at very low rev the rotational efficiency will skydrop)
As its an AC model i would keep experimenting with a metal encased high power resistor in series on 115vac wallsocket juice.
At 18 watts/115vac the fans resistance is around 735 Ohms.
Putting a 735 Ohms resistor is series with the fan will drop the voltage over the fan to half nominal value, current will be half too.
=> power of the fan will be 1/4th= 4.5 watts
The power of the fan is proportional to the square of the airspeed, if efficiency is kept constant.
4 times more power=> double airspeed
Courtesy of a French guy.
Same with the "Zoebel". A Zobel is some sort of mink skunk in Russia, the network is also called a Boucherot after the inventor, the Frenchman Paul Boucherot (lived and died before WW2).
Some people would change the fuse without pulling the cable, in case it would be safer to have the fuseholder behind the switch.
the Fan on the picture you posted is a Sunon, and a mighty hefty one for audio purposes, at 107 CFM.
I am not going to boring dB stuff again, but on nominal voltage at 3050 rev its a noisy one.
The blades of the Sunon are shaped for displacement, not low noise.
The reason Krell used a Papst on their amplifiers at the time was because those were both the longest lasting and the least noisy.
Current Papst models go as low as 18 dB at full rev, do not need derating at all for audio purpose heatsink tunnels.
I doubt that you can get this one derated to an acceptable noise level and still push enough air.
(the blade pitch is optimised for the nominal rpm, at very low rev the rotational efficiency will skydrop)
As its an AC model i would keep experimenting with a metal encased high power resistor in series on 115vac wallsocket juice.
At 18 watts/115vac the fans resistance is around 735 Ohms.
Putting a 735 Ohms resistor is series with the fan will drop the voltage over the fan to half nominal value, current will be half too.
=> power of the fan will be 1/4th= 4.5 watts
The power of the fan is proportional to the square of the airspeed, if efficiency is kept constant.
4 times more power=> double airspeed
Courtesy of a French guy.
Same with the "Zoebel". A Zobel is some sort of mink skunk in Russia, the network is also called a Boucherot after the inventor, the Frenchman Paul Boucherot (lived and died before WW2).
Some people would change the fuse without pulling the cable, in case it would be safer to have the fuseholder behind the switch.
Thanks again everyone.
Looks like maybe I should buy a different fan. I have to agree with jacco. This fan seems exceptionly noisy even at fairly low cfm. I don't want to risk overheating just to get the noise level acceptable.
So the Zobel thingy is a good idea?
More questions.
1) Is it OK to run a seperate rail feed directly to each set of collectors on the output transistors? Should these feeds have their own fuses?
2) I asked before but didn't get an answer. Is the single ground point next to the input sufficient for grounding the driver board or am I missing something?
Thanks again, Terry
Looks like maybe I should buy a different fan. I have to agree with jacco. This fan seems exceptionly noisy even at fairly low cfm. I don't want to risk overheating just to get the noise level acceptable.
So the Zobel thingy is a good idea?
More questions.
1) Is it OK to run a seperate rail feed directly to each set of collectors on the output transistors? Should these feeds have their own fuses?
2) I asked before but didn't get an answer. Is the single ground point next to the input sufficient for grounding the driver board or am I missing something?
Thanks again, Terry
zobel network
is a good idea, it can't hurt and may help.
The single ground wire taken from the driver board and run to a takeoff from the star ground seems to be the normal solution. The input ground is taken from the board itself and run to the socket, in parallel with the signal input, as a twisted pair, or a shielded cable.
I collected the power feeds for each output transistor together to make one large cable run from each of my heatsinks to feed from the caps, seemed like a more reliable connection. YMMV. Of course I also have 32 output transistors, you probably don't have quite the same concern about a huge rats nest of wires...
Rail fuses are quite common, but I have no idea if they react quickly enough to truly be effective in protecting speakers or amp, but I don't think they can hurt...it may be possible to achieve the same level of 'protection' using a single fuse on the primary side of the transformer.
Stuart
is a good idea, it can't hurt and may help.
The single ground wire taken from the driver board and run to a takeoff from the star ground seems to be the normal solution. The input ground is taken from the board itself and run to the socket, in parallel with the signal input, as a twisted pair, or a shielded cable.
I collected the power feeds for each output transistor together to make one large cable run from each of my heatsinks to feed from the caps, seemed like a more reliable connection. YMMV. Of course I also have 32 output transistors, you probably don't have quite the same concern about a huge rats nest of wires...
Rail fuses are quite common, but I have no idea if they react quickly enough to truly be effective in protecting speakers or amp, but I don't think they can hurt...it may be possible to achieve the same level of 'protection' using a single fuse on the primary side of the transformer.
Stuart
jacco vermeulen said:
AC motor speed is defined by the cycles (frequency), not the voltage, so if you succeed to make it run slower with a resistor, it's because the motor is not running proper and skipping cycles due to lack of power.
You can for some types of AC motors run them half speed by putting a diode in series with the motor, thus running it on half a cycle.
Magura
AC motors
I don't think all AC motors are using the freq of the mains for their specific speed. I have some examples that presumably rectify and internally use DC for their drive, they operate at a broad range of speeds without cogging or vibration of an discernible sort, and can spin stably at any speed within their normal range, governed by the applied voltage...
Stuart
I don't think all AC motors are using the freq of the mains for their specific speed. I have some examples that presumably rectify and internally use DC for their drive, they operate at a broad range of speeds without cogging or vibration of an discernible sort, and can spin stably at any speed within their normal range, governed by the applied voltage...
Stuart