Hi Everyone,
Hope the builds are going well. Vunce has already finished his. Hope you are having fun with FET "rolling," Vunce.
Time for an update to the design and PCB. Those who have them, please let me know the changes you'd like to see. I have a list going.
Changes to PCB:
1. Bigger pads where appropriate.
2. LED power indicator on the PCB.
3. Smaller overall PCB size.
4. More holes at different pitches for input/output caps.
5. Get rid of TO-92 PCB footprint--this package can be put into the TO-220 slot just as easily.
6. Get rid of jumpers in power supply. I have a better idea here to adjust PS voltage.
(Could use some feedback on these next ones)
7. Add option for a volume pot at the input. Is this a good idea or leave it off the PCB?
8. Separate PS PCB--I'm torn on this one. On the one hand, keeping it on the same PCB makes everything neat and tidy in the chassis. On the other, it makes the PCB larger and folks like to use their own supplies anyway. Thoughts?
Design updates:
Over the next week or so I want to test a few things:
1. Mu-follower output. This is pretty easy, just take the output from the source pin of the second CCS FET instead of from below the trimpot. I prefer the CCS out instead of the mu-follower out on the 10Y tube preamp, but worth a go here. It could be made jumper-selectable on the PCB.
2. Add output buffer. A simple FET follower like the LuminAria would do and it would relieve the gain FET from driving variable loads. The performance is already quite good, so maybe this is not needed, but some FETs do better than others in terms of bass.
3. Use only one FET in the CCS. This will save one DN2535 FET. This is very much doable, probably with no effect on the sound and performance.
4. Modify the CCS by using the Aleph CCS and ghosting portion of the output to the CCS. This is similar to #1 and I intend to test it out.
Are you having fun? I know I am. Thanks for all the encouragement so far. Time to add some of these updates and make a better PCB.
Hope the builds are going well. Vunce has already finished his. Hope you are having fun with FET "rolling," Vunce.
Time for an update to the design and PCB. Those who have them, please let me know the changes you'd like to see. I have a list going.
Changes to PCB:
1. Bigger pads where appropriate.
2. LED power indicator on the PCB.
3. Smaller overall PCB size.
4. More holes at different pitches for input/output caps.
5. Get rid of TO-92 PCB footprint--this package can be put into the TO-220 slot just as easily.
6. Get rid of jumpers in power supply. I have a better idea here to adjust PS voltage.
(Could use some feedback on these next ones)
7. Add option for a volume pot at the input. Is this a good idea or leave it off the PCB?
8. Separate PS PCB--I'm torn on this one. On the one hand, keeping it on the same PCB makes everything neat and tidy in the chassis. On the other, it makes the PCB larger and folks like to use their own supplies anyway. Thoughts?
Design updates:
Over the next week or so I want to test a few things:
1. Mu-follower output. This is pretty easy, just take the output from the source pin of the second CCS FET instead of from below the trimpot. I prefer the CCS out instead of the mu-follower out on the 10Y tube preamp, but worth a go here. It could be made jumper-selectable on the PCB.
2. Add output buffer. A simple FET follower like the LuminAria would do and it would relieve the gain FET from driving variable loads. The performance is already quite good, so maybe this is not needed, but some FETs do better than others in terms of bass.
3. Use only one FET in the CCS. This will save one DN2535 FET. This is very much doable, probably with no effect on the sound and performance.
4. Modify the CCS by using the Aleph CCS and ghosting portion of the output to the CCS. This is similar to #1 and I intend to test it out.
Are you having fun? I know I am. Thanks for all the encouragement so far. Time to add some of these updates and make a better PCB.
When thinking through what a power amplifier with this topology would look like, I came across the Arch Nemesis and this schematic. Look familiar?
I also experimented with high voltage operating points for the FETs. In fact, I hooked up perfboards to the 10Y preamp (without the tubes, obviously). So, operating points near 200-250V and 10-20 mA, similar to the 10Y. I was wondering if there is anything special happening at higher voltages. Not really, the sound was about the same. But it was a good experiment because it satisfied my curiosity about whether higher voltage in itself is better. I have heard this as a virtue of tubes, but now we have FETs that can operate at similar voltages and currents and the sound isn't better. So, no, higher voltage by itself doesn't sound good, at least not in this topology.
Hey Ra,
Yup definitely having fun with your SCG! Actually, just installed MKP1839 bypass caps in C102/202. No Fet rolling yet.
I agree with your comments 1-6 for revised pcb, I have a few additional suggestions and comments to add for SCG v2 pcb:
-LED indicators for each channel psu
-Onboard 10R NTC//22nF cap GLB output to chassis GND
-Transformer snubbers at AC input a’la Quasimodo
-Gate resistors closer to FETs, or add SMD pads as option for builders.
Volume pot onboard is a bit tricky, that would take away some freedom of board placement in chassis.
My opinion is DIY’ers have many more LV psu options in their toolbox compared to HV psu’s. In the case of SCG having the HV psu section on the same board is a benefit and should stay. If a builder wants to use their own supplies, add pads for direct DC power input bypassing the onboard supply.
The SSR delay probably can be removed, It’s delay is not long enough to get past the slow dc ramp up then decline at power up. Also, its not immediately disengaged at power down because the filter caps store enough energy to keep the SSR powered for several seconds.
As long as you practice “preamp on first and off last” the dc at power up will never be bothersome.
Yup definitely having fun with your SCG! Actually, just installed MKP1839 bypass caps in C102/202. No Fet rolling yet.
I agree with your comments 1-6 for revised pcb, I have a few additional suggestions and comments to add for SCG v2 pcb:
-LED indicators for each channel psu
-Onboard 10R NTC//22nF cap GLB output to chassis GND
-Transformer snubbers at AC input a’la Quasimodo
-Gate resistors closer to FETs, or add SMD pads as option for builders.
Volume pot onboard is a bit tricky, that would take away some freedom of board placement in chassis.
My opinion is DIY’ers have many more LV psu options in their toolbox compared to HV psu’s. In the case of SCG having the HV psu section on the same board is a benefit and should stay. If a builder wants to use their own supplies, add pads for direct DC power input bypassing the onboard supply.
The SSR delay probably can be removed, It’s delay is not long enough to get past the slow dc ramp up then decline at power up. Also, its not immediately disengaged at power down because the filter caps store enough energy to keep the SSR powered for several seconds.
As long as you practice “preamp on first and off last” the dc at power up will never be bothersome.
Attachments
ra7,
I have most of the board stuffed. I’m just waiting for some parts to finish my build. I’m going to put it in a Pesante chassis with baseplate which has 10mm x 10mm mounting holes, so I will need to drill some holes for mounting. If the board mounting holes were laid out to fit that would be great.
I have most of the board stuffed. I’m just waiting for some parts to finish my build. I’m going to put it in a Pesante chassis with baseplate which has 10mm x 10mm mounting holes, so I will need to drill some holes for mounting. If the board mounting holes were laid out to fit that would be great.
Just a progress update. Majority of the components are populated. Using STF5NK100Z at the power supply regulation, 10M45s for the CCS since these are left over parts from my project years ago. Gain FET STP30N10F7 which are kindly provided by ra7 are installed at this moment.I am using my favorite Nippon Chemicon KMH series for the Psu.
Toroidal transformer is going to be delivered soon I hope, while I am still waiting for the lower P channel FET, which is still vacant on the board. Beautiful board, and I look forward to firing it up.
Toroidal transformer is going to be delivered soon I hope, while I am still waiting for the lower P channel FET, which is still vacant on the board. Beautiful board, and I look forward to firing it up.
Attachments
The rectifier diodes are MURF1060, bought a few dozens some time ago, same goes for the bypass caps.... Trying to use parts I have on hand, they have been collecting dust.Great progress CT!!
Nice to have the ccs and reg mosfets already in your stash. By chance, we are both using the same bypass cap in C102/202. What rectification diodes are you using?
Hope that transformer gets delivered ASAP!!
Observation:
The SCG pairs wonderfully with low gain, follower and buffer type amplifiers. It’s not a good match with amplifiers that have their own built-in voltage gain of say 20dB and up. Volume control becomes very difficult with only 1/8th twist of the dial and music is too loud.
I have different preamps I can swap for these situations, but I wonder if the SCG gain can be made adjustable? But I suspect this will mess up the “Schade” effect.
Just thinkin’ out loud…..
The SCG pairs wonderfully with low gain, follower and buffer type amplifiers. It’s not a good match with amplifiers that have their own built-in voltage gain of say 20dB and up. Volume control becomes very difficult with only 1/8th twist of the dial and music is too loud.
I have different preamps I can swap for these situations, but I wonder if the SCG gain can be made adjustable? But I suspect this will mess up the “Schade” effect.
Just thinkin’ out loud…..
commstech,
Can you post a picture on how you mount the MURF1060 to the pcb ?Thanks.
The gain can be adjusted by changing the ratio of the schade feedback resistors , but yes i wonder if that could actually cause things to mess up or it may just work fine.Observation:
The SCG pairs wonderfully with low gain, follower and buffer type amplifiers. It’s not a good match with amplifiers that have their own built-in voltage gain of say 20dB and up. Volume control becomes very difficult with only 1/8th twist of the dial and music is too loud.
I have different preamps I can swap for these situations, but I wonder if the SCG gain can be made adjustable? But I suspect this will mess up the “Schade” effect.
Just thinkin’ out loud…..