which Andrew are you asking/quoting?lgreen said:
I never said anything that could be construed to fit your interpretation.
I said stop being a skin flint and go and buy resistors in bulk and gradually build up a stock of all 6decades of E24 values.
Re: zobel
What I called R2 is actually R28 on the board. The 8 is lost in the poor silkscreen because it falls on the outline of R1.
You can place the Zobel directly on the speaker output as some manufacturers do. I have never had a problem with it's location on the board. This amp is very stable.
lgreen said:
What I called R2 is actually R28 on the board. The 8 is lost in the poor silkscreen because it falls on the outline of R1.
You can place the Zobel directly on the speaker output as some manufacturers do. I have never had a problem with it's location on the board. This amp is very stable.
great
Thanks Steve.
I received the 33 ohm resistors from PH104, thank you very much. Nice Dale's too. Wow.
With everyone's help this is coming along.
Except that I noticed when I put the resistors in that I had a diode installed backwards!!! You try and be careful putting everthing in exactly as it should be, checking off the components, measuring resistances, reading the schematic, BOM and stuffit pic, and this happens. I had triple checked the boards too.
Anyway, time to drill some holes in the sink. I mounted the transistors via one leg only to mark the drill holes. After drilling I will remove the transistors, screw them into the heatsink, and then resolder. They should and turn out to be in exactly the same position as they are now (yeah right!). This way the transistor package will not face a lot of stress.
It would have been better to wait to install the big caps as the last step but hey live and learn.
Also, I did not see anyone mention the LED resistor. According to my calcs, a 5.6K 2 watter is what you need there (about 10 mA current draw on the LED gives about 1 W dissipation @ 50V). And this is what is installed. I know LEDs use about 20 mA but they are sufficiently bright with 10.
As you see I decided to AC couple the feedback. I snuck in a big 220 uF 160V cap connected to R10 that really didn't fit. But I had them in the junk box so they are propped up on the board.
Thanks Steve.
I received the 33 ohm resistors from PH104, thank you very much. Nice Dale's too. Wow.
With everyone's help this is coming along.
An externally hosted image should be here but it was not working when we last tested it.
Except that I noticed when I put the resistors in that I had a diode installed backwards!!! You try and be careful putting everthing in exactly as it should be, checking off the components, measuring resistances, reading the schematic, BOM and stuffit pic, and this happens. I had triple checked the boards too.
Anyway, time to drill some holes in the sink. I mounted the transistors via one leg only to mark the drill holes. After drilling I will remove the transistors, screw them into the heatsink, and then resolder. They should and turn out to be in exactly the same position as they are now (yeah right!). This way the transistor package will not face a lot of stress.
It would have been better to wait to install the big caps as the last step but hey live and learn.
Also, I did not see anyone mention the LED resistor. According to my calcs, a 5.6K 2 watter is what you need there (about 10 mA current draw on the LED gives about 1 W dissipation @ 50V). And this is what is installed. I know LEDs use about 20 mA but they are sufficiently bright with 10.
As you see I decided to AC couple the feedback. I snuck in a big 220 uF 160V cap connected to R10 that really didn't fit. But I had them in the junk box so they are propped up on the board.
Re: great
I haven't said much about the resistor for the diode for a reason. It appears you have figured it out. The resistor used there will be determined by the supply voltage, current requirements of the LED and the desired brightness of the LED. No practical value can be listed that will work in all situations.
lgreen said:
I haven't said much about the resistor for the diode for a reason. It appears you have figured it out. The resistor used there will be determined by the supply voltage, current requirements of the LED and the desired brightness of the LED. No practical value can be listed that will work in all situations.
Re: Re: great
Ha! This is something that I have experience with. LEDs will usually look fine in the range of from 5-20 mA. So this value and rating should be ok for most situations. The problem is the high supply voltage (50V in the typical case), means you have to dissipate a lot of power in the resistor.
Steve Dunlap said:
Ha! This is something that I have experience with. LEDs will usually look fine in the range of from 5-20 mA. So this value and rating should be ok for most situations. The problem is the high supply voltage (50V in the typical case), means you have to dissipate a lot of power in the resistor.
Re: Re: Re: great
Each stands clear of the PCB, one on top as normal and one mounted on the uncut leads of the first but on the bottom side.
use two resistors in parallel.lgreen said:
Each stands clear of the PCB, one on top as normal and one mounted on the uncut leads of the first but on the bottom side.
Re: Re: Re: Re: great
I left space for a 5W wire wound, so a large resistor should be no problem.
AndrewT said:
I left space for a 5W wire wound, so a large resistor should be no problem.
Krill Startup Sequence
Is there a post somewhere about what voltages to watch when starting up the Krill for the first time?
I will monitor the suppy rails of course but what else should be looked at as I crank up (slowly) the variac?
And what are the initial settings of the 2 pots and what should they be adjusted to?
I think I will power up this weekend or early next week.
Is there a post somewhere about what voltages to watch when starting up the Krill for the first time?
I will monitor the suppy rails of course but what else should be looked at as I crank up (slowly) the variac?
And what are the initial settings of the 2 pots and what should they be adjusted to?
I think I will power up this weekend or early next week.
Re: Krill Startup Sequence
That has been posted before, but it will be quicker to repeat it than to find it.
Before power up:
Set R26 to give a reading of about 36 ohms across R5.
Set R27 to minimum.
Monitor current draw while powering up on the variac. It should be very low.
Check the regulated supplies. The voltages should be about the same value on the + and -. When your input voltage goes high enough to turn on the zener, check to see that the regulated voltages are correct.
Set the offset to a acceptably low value with R26. Set bias with R27. I would recommend a bias of 35ma or greater for each output. If you have a distortion analyzer, adjust R27 for minimum distortion at 1kHz at rated power into 8 ohms.
Let the unit idle for 30 minutes and check and readjust the bias and offset if needed. The offset should be set with no load on the output.
Listen and enjoy.
That should give you a starting point.
lgreen said:
That has been posted before, but it will be quicker to repeat it than to find it.
Before power up:
Set R26 to give a reading of about 36 ohms across R5.
Set R27 to minimum.
Monitor current draw while powering up on the variac. It should be very low.
Check the regulated supplies. The voltages should be about the same value on the + and -. When your input voltage goes high enough to turn on the zener, check to see that the regulated voltages are correct.
Set the offset to a acceptably low value with R26. Set bias with R27. I would recommend a bias of 35ma or greater for each output. If you have a distortion analyzer, adjust R27 for minimum distortion at 1kHz at rated power into 8 ohms.
Let the unit idle for 30 minutes and check and readjust the bias and offset if needed. The offset should be set with no load on the output.
Listen and enjoy.
That should give you a starting point.
startup
OK, I started it up, slowly increased the variac and everything was looking good and I was feeling lucky.
All power supplies measure ok, + and - 90 VDC unreg, and about + and - 55 reg and about + and - 50 larg ps.
The problem is the adjustments of DC offset and bias.
I cannot reduce the dc offset below 218 mV, as you turn the pot from low to high it just goes up.
I cannot set the bias to 7.9 mV across the 0.22 ohm emitter resistor. As I turn R27, bias only increases to about 2.8 mV.
My diode board is screwed onto the heat sink in the middle above the central transistors (hottest part), the glass diodes are pressed to touch a mica insulator which is on the heat sink, and thermal paste is used.
Other than this, works great. What do you think the problem is?
OK, I started it up, slowly increased the variac and everything was looking good and I was feeling lucky.
All power supplies measure ok, + and - 90 VDC unreg, and about + and - 55 reg and about + and - 50 larg ps.
The problem is the adjustments of DC offset and bias.
I cannot reduce the dc offset below 218 mV, as you turn the pot from low to high it just goes up.
I cannot set the bias to 7.9 mV across the 0.22 ohm emitter resistor. As I turn R27, bias only increases to about 2.8 mV.
My diode board is screwed onto the heat sink in the middle above the central transistors (hottest part), the glass diodes are pressed to touch a mica insulator which is on the heat sink, and thermal paste is used.
Other than this, works great. What do you think the problem is?
Re: startup
The first problem is a bit puzzling. As you turn the pot up (increase the resistance), the offset should go lower in value and should eventually become negative. Make sure the values of R5 and R14 are correct.
The second problem has been encountered by two other people, that I know of. The solution was to use a small heat sink on all the TO220 transistors except Q8 and Q11. Also make sure the bias pot is the correct value.
It sounds like your mounting of the diode board is correct. The only other suggestion at this point is the standard "check parts value and soldering".
If every thing is correct, try changing R5 to 100 ohms.
lgreen said:
The first problem is a bit puzzling. As you turn the pot up (increase the resistance), the offset should go lower in value and should eventually become negative. Make sure the values of R5 and R14 are correct.
The second problem has been encountered by two other people, that I know of. The solution was to use a small heat sink on all the TO220 transistors except Q8 and Q11. Also make sure the bias pot is the correct value.
It sounds like your mounting of the diode board is correct. The only other suggestion at this point is the standard "check parts value and soldering".
If every thing is correct, try changing R5 to 100 ohms.
Re: Re: startup
How am I supposed to do that? Those TO-220's are virtually touching. I cannot even get a screw through the ones that are near the big caps. Actually I can only get a screw through like 3 of them. The staggering of the group prevents wide ones from fitting.
Are there clip-ons or anything that can slip on? The clip ons I have seen will be too wide and move the nearby ones.
edit- how about
this is #1 or
this is #2
Steve Dunlap said:
How am I supposed to do that? Those TO-220's are virtually touching. I cannot even get a screw through the ones that are near the big caps. Actually I can only get a screw through like 3 of them. The staggering of the group prevents wide ones from fitting.
Are there clip-ons or anything that can slip on? The clip ons I have seen will be too wide and move the nearby ones.
edit- how about
this is #1 or
this is #2
Re: Re: Re: startup
Try these from Digi-Key.
HS279-ND
If they will not fit, flatten them. Remember Q8 and Q11 do not need heat sinks. The one near the large cap can be bent over slightly if the leads were left long enough. You can also remove it from the board to attach the heat sink.
lgreen said:
Try these from Digi-Key.
HS279-ND
If they will not fit, flatten them. Remember Q8 and Q11 do not need heat sinks. The one near the large cap can be bent over slightly if the leads were left long enough. You can also remove it from the board to attach the heat sink.
heatsinks
Thanks Steve,
those heatsinks need a screw to hold them in and for the life of me I cannot get screws through most of the to-220's unless I remove them and put on the heatsink and then resolder them in. This requires dissassembly of a few things first.
How about some thermal glue, that would make things easier? I've never used it before but it seems like the way to go now with the transistors already in there.
Thanks Steve,
those heatsinks need a screw to hold them in and for the life of me I cannot get screws through most of the to-220's unless I remove them and put on the heatsink and then resolder them in. This requires dissassembly of a few things first.
How about some thermal glue, that would make things easier? I've never used it before but it seems like the way to go now with the transistors already in there.
I ran into the same problem. I just used some strips of brass stock at first then pieces of 1/2" aluminum angle when I went to +/- 70VDC on those transistors. There are pics in posts 150 and 151.
With a little gentle bending of the transistors, you can get the heat sinks mounted with 4-40 machine screws. I used a pair of hemostats to hold the nuts while I screwed (ok c2cthomas -- take your best shot!).
With a little gentle bending of the transistors, you can get the heat sinks mounted with 4-40 machine screws. I used a pair of hemostats to hold the nuts while I screwed (ok c2cthomas -- take your best shot!).
tape
I think I'll try some thermal tape temporarily to see if I can get it working. I'd need super short 4-40 screws and that would take some time to locate. I'll try and kludge in some small heatsinks; I've got a bunch that are slightly bigger than the ones Steve posted and some aluminum pieces and angles and so forth.
FYI, the + and - 56V supplies do not exactly match. Do you think this was due to my forgetting to heatsink the regulating (pass) transistors? And could it be the reason I cannot adjust offset?
I think I'll try some thermal tape temporarily to see if I can get it working. I'd need super short 4-40 screws and that would take some time to locate. I'll try and kludge in some small heatsinks; I've got a bunch that are slightly bigger than the ones Steve posted and some aluminum pieces and angles and so forth.
FYI, the + and - 56V supplies do not exactly match. Do you think this was due to my forgetting to heatsink the regulating (pass) transistors? And could it be the reason I cannot adjust offset?
what's "exactly"? How different are those voltages? And what are the voltages at the bases of the pass transistors?