Andypairo said:
The power supply is 50V for the 100W unit. 63V caps should be fine. That is the value I have always used for the large filter caps and I have never had a cap failure. The spec for DC on caps is for rated voltage or DC + ripple. The 63V caps will be at a maximum of 80% of rated voltage. That leaves plenty of room for ripple.
AndrewT said:
I have never had a stability problem using a cap on the inverting input. Perhaps I have been lucky.
There is no problem splitting that resistor and placing the cap at the junction as you suggest. I will do that.
The only issue I have with the non-inverting configuration (in this implementation) is the inability to go below unity gain. My simulation shows better results with with the inverting configuration.
Keep in mind, I have not built and tried this. It should work, but without building it we can't know for sure. If you do build it with the non-inverting configuration, remember to turn the LED in the LDR around for proper polarity.
I gladly accept any input that makes something better. Especially if it is as cheap as 2 resistors.
Hi Steve,
I was talking about C17 and C18, that have to sustain about 90V with the specified rails.
Anyway I checked better and the 100V figure is reached with substantially higher AC voltage (41V) so 100V is fine for these caps.
C15 and C16 can be 63V without problems.
Ciao
Andrea
I was talking about C17 and C18, that have to sustain about 90V with the specified rails.
Anyway I checked better and the 100V figure is reached with substantially higher AC voltage (41V) so 100V is fine for these caps.
C15 and C16 can be 63V without problems.
Ciao
Andrea
Andypairo said:
I misunderstood which caps you were referring to. These caps will have a fairly ripple, but they should still have less than 70V on them in the50W amps. At just under 70% of their rating they will be fine.
I specified 160V caps for the 100W unit. These will run close to 100V and will be will within their ratings. I don't like to cut things too close.
I am starting to get feedback from people building the 50W amp. Here are some of the things brought to my attention.
There are two holes at one end of R10. The closer spacing is so C22 can be left out. If R10 is installed in these two holes, then no jumper is needed across C22. If you have already installed R10 in the standard space holes, then you will need to add a jumper across the mounting holes for C22.
There is a space on the board for trim pot R26. Use this instead of R60 shown on earlier schematics. Before the amp is powered up, set the resistance as measured across R5 to 35 to 40 ohms. After power up, adjust R26 for minimum DC at the base junction of Q7 and Q10. You can also measure this voltage at the end of R11 closest to Q5. If you add the limiting resistor R20 (there are no mounting holes for R20) then you still set the resistance the same way, but you cannot over adjust the trimmer into a potentially damaging value. Adjust R26 slowly and carefully.
R28 (the large one near R19) has been changed to a 5W resistor as it should have been all along.
I am attaching the updated schematic with some changes. R26 and R27 have been changed to single turn pots as that is all I seem to be able to find that fits the board space allotted for them. R27 has been changed to 10K because that style pot is not available in the 5K value I had originally used.
Other changes on the schematic are some resistor values. These values changed as a result of moving the location of R26. The resistors that were changed are R5, R12 and R14. R1 was changed from 220 ohms to 1K to improve the balance between Q1 and Q2.
R20 and R25 have been added as "safety" resistors. There is no place for them on this board, but they can be added by cutting a trace in the case of R20. R25 can be tacked across the diode string where the wires come back to the board. No trace needs to be cut for this addition.
I have added R29, R30, R31 and R32 to the schematic for output stage DC offset adjustment. To add these, you will need to cut two traces on top of the board and add them (carefully) on the bottom of the board. I feel this is a worthwhile change to make. If you choose not to make this change, then the offset will be adjusted by R26 and measured at the output with no load attached. If you do add these resistors, then you first set DC as described above with R26, then set the DC offset of the output using one of the added trim pots. Both pots should be Set for minimum resistance. If you have positive DC on the output, you adjust R30. For negative DC you adjust R29. Only one pot will need to be adjusted.
If you don't have access to a distortion meter, then set the output bias at a level you are comfortable with. 75mA should give good results. If you can measure distortion, then adjust bias for minimum distortion at some set output power (this will of course require a load on the output). I would recommend setting the bias for minimum distortion at a power level that reflects actual listening levels, say 10W, instead of maximum unclipped output.
After the Bias is set, let the unit idle for 30 minutes with no load, then check and adjust R26 and R27 in that order. Add back the load and reset the bias if using a distortion metre.
The setup for the 100W amp is the same.
There is a small feed through hole near Q6. This should be filled with solder.
Some diodes and caps have multiple mounting holes to accommodate different lead spacings. Don't get excited if you have unfilled holes after assembly. These extra holes may be filled with solder or left empty.
The diodes on the diode board that mounts to the heat sink should be installed as flat to the board as possible to avoid possible damage when the board is mounted. Use a fiber or nylon spacer under the board at the mounting hole. This pacer should be close to the diode height on the board. Use an insulator for a TO220 transistor under the board to make sure the diode leads can't short to the heat sink. I prefer one of the soft silicone types for this. If you use mica, remember to use a small amount of heat sink compound to insure good thermal tracking,
I have decided to recommend adding hear sinks to the pass transistors Q17 and Q18. They will be running within their ratings, but a small bolt on or clip on heat sink will run them at a more conservative temperature.
I'm sure I missed something. I will post it when I or one of you notices what it might be. I have tried to match the schematic to the boards I am selling. Hopefully I have it right this time.
There are two holes at one end of R10. The closer spacing is so C22 can be left out. If R10 is installed in these two holes, then no jumper is needed across C22. If you have already installed R10 in the standard space holes, then you will need to add a jumper across the mounting holes for C22.
There is a space on the board for trim pot R26. Use this instead of R60 shown on earlier schematics. Before the amp is powered up, set the resistance as measured across R5 to 35 to 40 ohms. After power up, adjust R26 for minimum DC at the base junction of Q7 and Q10. You can also measure this voltage at the end of R11 closest to Q5. If you add the limiting resistor R20 (there are no mounting holes for R20) then you still set the resistance the same way, but you cannot over adjust the trimmer into a potentially damaging value. Adjust R26 slowly and carefully.
R28 (the large one near R19) has been changed to a 5W resistor as it should have been all along.
I am attaching the updated schematic with some changes. R26 and R27 have been changed to single turn pots as that is all I seem to be able to find that fits the board space allotted for them. R27 has been changed to 10K because that style pot is not available in the 5K value I had originally used.
Other changes on the schematic are some resistor values. These values changed as a result of moving the location of R26. The resistors that were changed are R5, R12 and R14. R1 was changed from 220 ohms to 1K to improve the balance between Q1 and Q2.
R20 and R25 have been added as "safety" resistors. There is no place for them on this board, but they can be added by cutting a trace in the case of R20. R25 can be tacked across the diode string where the wires come back to the board. No trace needs to be cut for this addition.
I have added R29, R30, R31 and R32 to the schematic for output stage DC offset adjustment. To add these, you will need to cut two traces on top of the board and add them (carefully) on the bottom of the board. I feel this is a worthwhile change to make. If you choose not to make this change, then the offset will be adjusted by R26 and measured at the output with no load attached. If you do add these resistors, then you first set DC as described above with R26, then set the DC offset of the output using one of the added trim pots. Both pots should be Set for minimum resistance. If you have positive DC on the output, you adjust R30. For negative DC you adjust R29. Only one pot will need to be adjusted.
If you don't have access to a distortion meter, then set the output bias at a level you are comfortable with. 75mA should give good results. If you can measure distortion, then adjust bias for minimum distortion at some set output power (this will of course require a load on the output). I would recommend setting the bias for minimum distortion at a power level that reflects actual listening levels, say 10W, instead of maximum unclipped output.
After the Bias is set, let the unit idle for 30 minutes with no load, then check and adjust R26 and R27 in that order. Add back the load and reset the bias if using a distortion metre.
The setup for the 100W amp is the same.
There is a small feed through hole near Q6. This should be filled with solder.
Some diodes and caps have multiple mounting holes to accommodate different lead spacings. Don't get excited if you have unfilled holes after assembly. These extra holes may be filled with solder or left empty.
The diodes on the diode board that mounts to the heat sink should be installed as flat to the board as possible to avoid possible damage when the board is mounted. Use a fiber or nylon spacer under the board at the mounting hole. This pacer should be close to the diode height on the board. Use an insulator for a TO220 transistor under the board to make sure the diode leads can't short to the heat sink. I prefer one of the soft silicone types for this. If you use mica, remember to use a small amount of heat sink compound to insure good thermal tracking,
I have decided to recommend adding hear sinks to the pass transistors Q17 and Q18. They will be running within their ratings, but a small bolt on or clip on heat sink will run them at a more conservative temperature.
I'm sure I missed something. I will post it when I or one of you notices what it might be. I have tried to match the schematic to the boards I am selling. Hopefully I have it right this time.
Here is the BOM for the 50W amp schematic in the previous post. I have included the Mouser and Digi-Key part numbers as well as the number of each part in one channel. I have also included the schematic designations for all parts.
Hi Steve, Thanks for the updates!! 😀 😀
Now I guess that I'll need to go back to work on this thing! 🙄
Now I guess that I'll need to go back to work on this thing! 🙄
Parts just arrived...thank you Steve Dunlap and also thanks
to Thomas Jennings,
Carlos
to Thomas Jennings,
Carlos
An externally hosted image should be here but it was not working when we last tested it.
A good friend, a local one, will assemble to me Thomas
Blood circulation is turning worse.... so he will build as i cannot be sitted....you see... now i have a protoboard (breadboard) to assemble layed down into the bed.
Sorry for that..... a friend will do and we gonna produce pictures.
I will be helping him but not using my hands and body... will be intelectual support only.
regards,
Carlos
Blood circulation is turning worse.... so he will build as i cannot be sitted....you see... now i have a protoboard (breadboard) to assemble layed down into the bed.
Sorry for that..... a friend will do and we gonna produce pictures.
I will be helping him but not using my hands and body... will be intelectual support only.
regards,
Carlos
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.
Take care Carlos, I hope your condition improves soon! You are staying upbeat & optimistic which is a great attitude.
Good health to you
Good health to you
relief from circulation problems
Hi Carlos, I found the above on the net ... might be helpful ... what do you think ...
🙂
destroyer X said:
Hi Carlos, I found the above on the net ... might be helpful ... what do you think ...
🙂
Here is another small correction to the component layout map. I had D9 and D10 swapped around.
I'm trying to keep Steve honest by doing a point to point continuity check on the PWB and the schematic. (Man - my eyes aren't what the used to be
).
I'm working on a newer version of the layout that uses a scanned image of the PWB. On top of that I'm adding the components and also coloring in the voltage supply runs. This should help when it comes to cutting tracks and adding components. Hope to have something to show in a day or two.
I'm trying to keep Steve honest by doing a point to point continuity check on the PWB and the schematic. (Man - my eyes aren't what the used to be
).I'm working on a newer version of the layout that uses a scanned image of the PWB. On top of that I'm adding the components and also coloring in the voltage supply runs. This should help when it comes to cutting tracks and adding components. Hope to have something to show in a day or two.
Attachments
Hi,
since I'm having trouble finding 330uF-160V that fit in the board space I'm thinking about using a different approach.
I have a toroid trafo with double 35+35 and 50+50Vac.
Separately rectifying the 50Vac gives me about 70Vdc, enough to feed the 56V regulator.
This approach can be used without cutting traces (just a unusual placement for 2 diodes of the bridge) and allows me to use some very good Sikorel caps I have in stock.
Steve do you see any drawbacks? (maybe R23-R24 need to be lowered?)
Ciao
Andrea
since I'm having trouble finding 330uF-160V that fit in the board space I'm thinking about using a different approach.
I have a toroid trafo with double 35+35 and 50+50Vac.
Separately rectifying the 50Vac gives me about 70Vdc, enough to feed the 56V regulator.
This approach can be used without cutting traces (just a unusual placement for 2 diodes of the bridge) and allows me to use some very good Sikorel caps I have in stock.
Steve do you see any drawbacks? (maybe R23-R24 need to be lowered?)
Ciao
Andrea
Andypairo said:
I see no drawbacks at all. R23 and R24 can be left out.
I'm still updating the BOM for the 100W amp. For the 50W I have confirmed that the parts fit the board. I will do the same for the 100W, changing values as needed.
I have made another small change to the 50W amp schematic. I will make changes to the rest and post them latter.
This change will only effect those of you using a preamp with a capacitor coupled output. I have no idea how common those still are. I haven't had a cap in a signal path in my system for over 25 years.
If you have a preamp with a cap in the output, do not add R20. You never had to add R20, it was just a safety feature in case R26 was over adjusted. Change R26 to 100 ohms.
Nothing else changes.
I am attaching an integrated 50W amp schematic. Some of you may be interested in an integrated amp. C2 needs to be in this input to prevent the DC offset from changing with the input level adjustments. The 4.7uF is a somewhat arbatrary. Use the highest value you can afford in a good film cap.
This change will only effect those of you using a preamp with a capacitor coupled output. I have no idea how common those still are. I haven't had a cap in a signal path in my system for over 25 years.
If you have a preamp with a cap in the output, do not add R20. You never had to add R20, it was just a safety feature in case R26 was over adjusted. Change R26 to 100 ohms.
Nothing else changes.
I am attaching an integrated 50W amp schematic. Some of you may be interested in an integrated amp. C2 needs to be in this input to prevent the DC offset from changing with the input level adjustments. The 4.7uF is a somewhat arbatrary. Use the highest value you can afford in a good film cap.
I have made another small change to the 50W amp schematic. I will make changes to the rest and post them latter.
This change will only effect those of you using a preamp with a capacitor coupled output. I have no idea how common those still are. I haven't had a cap in a signal path in my system for over 25 years.
If you have a preamp with a cap in the output, do not add R20. You never had to add R20, it was just a safety feature in case R26 was over adjusted. Change R26 to 100 ohms.
Nothing else changes.
I am attaching an integrated 50W amp schematic. Some of you may be interested in an integrated amp. C2 needs to be in this input to prevent the DC offset from changing with the input level adjustments. The 4.7uF is a somewhat arbatrary. Use the highest value you can afford in a good film cap.
This change will only effect those of you using a preamp with a capacitor coupled output. I have no idea how common those still are. I haven't had a cap in a signal path in my system for over 25 years.
If you have a preamp with a cap in the output, do not add R20. You never had to add R20, it was just a safety feature in case R26 was over adjusted. Change R26 to 100 ohms.
Nothing else changes.
I am attaching an integrated 50W amp schematic. Some of you may be interested in an integrated amp. C2 needs to be in this input to prevent the DC offset from changing with the input level adjustments. The 4.7uF is a somewhat arbatrary. Use the highest value you can afford in a good film cap.
Steve Dunlap said:
Ehm... R23 and R24 are the resistors that set the current in the zeners, I don't think they can be left out. But since the dropout has been halved (from 35V to less than 20V) I thought that their value should be lowered as well.
Am I wrong?
Ciao
Andrea
Andypairo said:
My mistake. I was thinking R21 and R22.
I select R23 and R24 to pass the test current for the zeners. That varies from device to device, but from 1 to 3 mA is a good choice. Any value resistor from 4.7K to 15K will work.
- Status
- Not open for further replies.