I would be interested in hearing about the performance
of your 2nd schematic. Ordinarily one would ground the
cross point between the JFET source resistors, but nobody
is saying yours wouldn't work.
If you have very well matched devices, then grounding or floating makes very little difference, as the X point is virtually at zero potential. In practice, not everyone has perfectly matched devices, so grounding is a more robust approach. But that removes the cross feedback and makes it two F5s driven out of phase.
I tried both, but grounded the X-point in the end just to put my mind at ease, as there was no audible difference to my ears.
Patrick
I tried both, but grounded the X-point in the end just to put my mind at ease, as there was no audible difference to my ears.
Patrick
Last edited:
PS
Absolute DC is an issue as in AX, so recommend loading both outputs to Gnd. I used 22R MP930.
Drift is better with Toshiba MOSFETs than Fairchild.
Trim each (L-R) half the amplifier for DC with the X point grounded first.
You can install a DPDT switch to ON/OFF the ground connection at the X, so that you may close them during the warm-up phase.
Patrick
Absolute DC is an issue as in AX, so recommend loading both outputs to Gnd. I used 22R MP930.
Drift is better with Toshiba MOSFETs than Fairchild.
Trim each (L-R) half the amplifier for DC with the X point grounded first.
You can install a DPDT switch to ON/OFF the ground connection at the X, so that you may close them during the warm-up phase.
Patrick
As with all things repetetive, the more I understand the process, the easier the process becomes.
I started sanding and polishing by hand but soon realized a small palm sander and a car buffer turned out a mirror finish in quick order. The faceplate was finished once and I discovered that a few scratches remained so I did most of the process again. The faceplate measures 19" x 8". It took about an hour to get it to a shiney surface.
Attachments
Could you comment on whether there should be (or could be) another reference to ground around the feedback loop. This was not omitted on the Aleph JX, though it is a slightly different arrangement from the original.[/QUOTE]
Thanh
I think if you read the post of Mr. Pass and EUVL you get answer for your question .
By the way are you planing to use multiple power mosfet or just one pair on each board .
Or you will design a PC board to the X both half together ?
I use separate PC board like Smyslow .
I'm very happy the Master himself gave answer to our question , also Patrick answer is very valuable !
Now I will go on and each the PC boards .
Greets
Thanh
I think if you read the post of Mr. Pass and EUVL you get answer for your question .
By the way are you planing to use multiple power mosfet or just one pair on each board .
Or you will design a PC board to the X both half together ?
I use separate PC board like Smyslow .
I'm very happy the Master himself gave answer to our question , also Patrick answer is very valuable !
Now I will go on and each the PC boards .
Greets
Last edited:
Hi,
You MUST NOT use a dual rectifier dual polarity PSU with a centre tapped transformer.
Centre tapped must use a single rectifier dual polarity supply
Dual secondary can use either a single rectifier or a dual rectifier.
Last edited:
Hey westend,
Thanks for the reply. The palm sander must have been dirty work. I hand wet sanded which was dust free. But like I said, it was along process. I was very pleased with the results though. I mayr try your technique this time. What's the highest grit you used on the palm sander?
Steve
Thermal switch?
I suppose that would be the famous thermistors, TH1/TH2 in original schematic
Tho some have mentioned they may not be effective with laterals(Toshiba)
Regarding Xed balanced F5
I suppose it will see half the load impedance, as usual with balanced amps
Which may half its classA output, bias depending
Or am I missing something
I suppose that would be the famous thermistors, TH1/TH2 in original schematic
Tho some have mentioned they may not be effective with laterals(Toshiba)
Regarding Xed balanced F5
I suppose it will see half the load impedance, as usual with balanced amps
Which may half its classA output, bias depending
Or am I missing something
Last edited:
Thermal Switch
Hey Tinitus,
No, not thermistors. I'm talking about a thermal switch that is wired into the primary side of the transformer that goes open if the temp gets above a certain level -- I think 75deg C in this case. It's for protection in case the amp starts overheating.
Thanks,
Steve
Hey Tinitus,
No, not thermistors. I'm talking about a thermal switch that is wired into the primary side of the transformer that goes open if the temp gets above a certain level -- I think 75deg C in this case. It's for protection in case the amp starts overheating.
Thanks,
Steve
I was considering mounting an actual thermal switch to each heatsink that would open above 70C. If either one opened it would disconnect power from the amplifier. I had not worked out the exact details yet as to whether it would be directly wired, or if it would trigger a latching relay. I prefer to keep AC voltage away from my fingers, I also prefer not to need a small secondary power source for protection duties.
I chose 70c ish because my heatsinks typically run in the mid 50's. My logic was that the only condition the heatsink would get that hot is if airflow was impeded or too much power was being dissapated indicating a fault.
I don't know if they are necessary. Ideally they will never be called to action. In my case I have the heatsinks on the inside and a slow, quiet fan moving air over them. If the fan was to stop for some reason, or if the amp was set on shag carpet instead of a hard surface - I could see it getting too hot.
I would say it is not necessary.
I would say it is not necessary.
Last edited:
Firstly on matching JFETs for the balanced version. If you were to use different source resistor values for the JFETs as I did, then you should match the Id of the JFET with the corresponding source resistor in place (i.e. not at Idss).
Zobel at the output would not work, as it only loads the amp at HF and not DC, which is what we want in this case.
On thermal drift, you can improve drift stability by coupling all JFETs together to a common thermal ground (a piece of aluminium or even better a heatsink). You may of course use a pair of my Dual JFET heatsinks (see GB forum), but that might limit your PCB layout freedom.
http://www.diyaudio.com/forums/group-buys/135359-toshiba-dual-jfet-heatsink.html
As to thermal switch, I do use them on my reference amp, and I have them mounted with thermal grease on the main heatsinks.
Since smyslow has apparently built the X-version with floating X, maybe he can also share his experiences.
Patrick
Zobel at the output would not work, as it only loads the amp at HF and not DC, which is what we want in this case.
On thermal drift, you can improve drift stability by coupling all JFETs together to a common thermal ground (a piece of aluminium or even better a heatsink). You may of course use a pair of my Dual JFET heatsinks (see GB forum), but that might limit your PCB layout freedom.
http://www.diyaudio.com/forums/group-buys/135359-toshiba-dual-jfet-heatsink.html
As to thermal switch, I do use them on my reference amp, and I have them mounted with thermal grease on the main heatsinks.
Since smyslow has apparently built the X-version with floating X, maybe he can also share his experiences.
Patrick
Last edited:
Thank You Patrick
I was ready to order the parts then I sow your post , so I canceled the order .
I will match the JETs Idss and I will use the same value source resistor for all 4 place ! Do I use 10R or less is better?
What is the purpose you use different feedback resistor value ?
I see Smyslow use 4x100R .
Yes I realized the Zobel wouldn't work in here .
I just look up some AX schematic . So I will buy Caddock MP930 resistors .Probably with a small TO220 heat sink .
Thank you one more time .
Greets
Jeno
You can cobble together the NP Power Supply on the F5 service manual webpage, then incorporate Troy Huebner's ideas from Nat Semi Application Note AN-1849 "An Audio Amplifier Power Supply Design". Although I don't like bridge rectifiers!
http://www.national.com/an/AN/AN-1849.pdf
Anyone heard from Troy recently?