10VA and 128Vac will provide a maximum continuous output current of 78mAac.
When this feeds a capacitor input filter and is rectified to give a DC output voltage ~=1.4*Vac then the maximum continuous DC current will be ~39mA.
I would advise that a 10VA 128Vac cannot reliably supply one channel of the Leach clone front end.
I think a 60-0-60 would just about do for a +-75Vdc front end supply.
For +-80Vdc at least 65-0-65 will be needed.
Last edited:
Tad, These are optional as per Dr. Leach original design... they will kill oscillation (if any) when the current limiter is active (from memory)
\\\Jens
Doubt it.
Pitch is correlated to the length of the cap, just think how tall a 1uF Wima 160Vac MKS2 type in 5mm would become.
A compromise would be to use something like a BC components/Vishay MKT 368.
Has it's leads already pointing to the center at the base of the cap, easy to fold back to a 5mm pitch.
Pitch is correlated to the length of the cap, just think how tall a 1uF Wima 160Vac MKS2 type in 5mm would become.
A compromise would be to use something like a BC components/Vishay MKT 368.
Has it's leads already pointing to the center at the base of the cap, easy to fold back to a 5mm pitch.
Luke,
The c5,c7,c9,c16,c19 positions all need the 1 uf 5 mm cap. I am looking over the schematic again and all of these positions have a resistor between the power and cap. I have yet to do the math -- as always -- but it would appear the voltage is being reduced enough to use the smaller voltage rated caps. With the frontend close to 90 volts I was just trying to be a little --whole lot of little - on the safe side. The 100 volt items I have on order from Mouser will work here.
I had been contemplating standing an axial on end.
BTW I had some .15uf 250 volt polypro caps to use for the .1 uf locations. Do you think this will make a major difference? I can still add the exact items to my Mouser order for the TR5 fuse holders. The mjl4281 On Semi's will be here this week from Avnet. Just 3.50 each.
I have most of one board stuffed and soldered. I now have to orient the leads on the Sanyo or Toshiba transitors for the VAS. Doing this and making it look nice and neat will be a chore.
This is my first build using my new Hakko soldering station and the Cardas quad eutectic solder. Makes for fairly easy assembly.
Anyone have any experience with the polycarbonate caps? A place in Florida has a good price on the .1 uf size.
Must now decide on dual mono or two separate chassis.
The entire build is coming in under 500.00 not including the free aluminum.
Thanks for the help Tad
The c5,c7,c9,c16,c19 positions all need the 1 uf 5 mm cap. I am looking over the schematic again and all of these positions have a resistor between the power and cap. I have yet to do the math -- as always -- but it would appear the voltage is being reduced enough to use the smaller voltage rated caps. With the frontend close to 90 volts I was just trying to be a little --whole lot of little - on the safe side. The 100 volt items I have on order from Mouser will work here.
I had been contemplating standing an axial on end.
BTW I had some .15uf 250 volt polypro caps to use for the .1 uf locations. Do you think this will make a major difference? I can still add the exact items to my Mouser order for the TR5 fuse holders. The mjl4281 On Semi's will be here this week from Avnet. Just 3.50 each.
I have most of one board stuffed and soldered. I now have to orient the leads on the Sanyo or Toshiba transitors for the VAS. Doing this and making it look nice and neat will be a chore.
This is my first build using my new Hakko soldering station and the Cardas quad eutectic solder. Makes for fairly easy assembly.
Anyone have any experience with the polycarbonate caps? A place in Florida has a good price on the .1 uf size.
Must now decide on dual mono or two separate chassis.
The entire build is coming in under 500.00 not including the free aluminum.
Thanks for the help Tad
Hi Tad,
C7 and C19 are across 2 X 20V zeners, they keep the voltage at 20V, 63 volt wimas with 5mm pitch is fine here.
C16 just shunts ac to ground very low capacitance here.
C9 is an input cap, this will see no more than 2 -4 V ac probably for full output.
C19 is the Vce across a Bc546, again 63V wima MKS2 will do it.
The 0.15uF polyprops will do fine in other locations, more is better here.
Polycarbonate are less desireable than Polyprops and polystyrene in audio but better than Polyester from memory. I think any good MKS or MKP will be fine in any location.
C7 and C19 are across 2 X 20V zeners, they keep the voltage at 20V, 63 volt wimas with 5mm pitch is fine here.
C16 just shunts ac to ground very low capacitance here.
C9 is an input cap, this will see no more than 2 -4 V ac probably for full output.
C19 is the Vce across a Bc546, again 63V wima MKS2 will do it.
The 0.15uF polyprops will do fine in other locations, more is better here.
Polycarbonate are less desireable than Polyprops and polystyrene in audio but better than Polyester from memory. I think any good MKS or MKP will be fine in any location.
If you are changing transistors here, you'd better be prepared to alter the value of the lag compensation caps. The capacitance of the VAS transistors will likely be different, and require different value lag caps to compensate the amp.
Pooge,
Thanks for this valuable information. This is way out of my realm of knowledge. I have done a preliminary check on the capacitance of the BC546/556 which Jens has on the BOM and the devices I WAS going to substitute for the VAS are 1-1.5 pf different.
I do not know if a change this small would make huge difference in the stability of the overall amp.
Dr. Leach, after rereading this area on the website, describes the lag caps as devices which control the Miller effect.
If this change will make or break my construction I may well leave everything as is exactly per the BOM. I was just looking for the best possible transitors for the VAS like what was described in the discussion here at DIY audio on this subject.
Aksa and others recommended using devices with the least amount of capacitance which I was trying to do. The BC items have very low cob themselves.
Tad
Thanks for this valuable information. This is way out of my realm of knowledge. I have done a preliminary check on the capacitance of the BC546/556 which Jens has on the BOM and the devices I WAS going to substitute for the VAS are 1-1.5 pf different.
I do not know if a change this small would make huge difference in the stability of the overall amp.
Dr. Leach, after rereading this area on the website, describes the lag caps as devices which control the Miller effect.
If this change will make or break my construction I may well leave everything as is exactly per the BOM. I was just looking for the best possible transitors for the VAS like what was described in the discussion here at DIY audio on this subject.
Aksa and others recommended using devices with the least amount of capacitance which I was trying to do. The BC items have very low cob themselves.
Tad
It is not a simple matter of looking at the COB of transistors, because the Miller capacitance is also a function of the circuit parameters. In other words, the COB of the transistor is multiplied by circuit voltage.
The lag comp cap has to be determined somewhat by trial and error. One way is to guesstimate the cap, check the square wave performance, and lower (or raise) the cap value until the square wave looks good. Some argue that a perfect square wave performance is not audibly ideal, and that you should lower the cap value until there is some evidence of overshoot.
The best thing to do would be to start with a known working BOM, and modify from there after you get it working. Less variables for headaches that way. I am modifying the transistors in an old Leach amp before building a new one so I can get it right before building a new one.
The lag comp cap has to be determined somewhat by trial and error. One way is to guesstimate the cap, check the square wave performance, and lower (or raise) the cap value until the square wave looks good. Some argue that a perfect square wave performance is not audibly ideal, and that you should lower the cap value until there is some evidence of overshoot.
The best thing to do would be to start with a known working BOM, and modify from there after you get it working. Less variables for headaches that way. I am modifying the transistors in an old Leach amp before building a new one so I can get it right before building a new one.
pooge,
There is no on line calculator that may be of help here?
We know what works and we know the parameters of the knew device. I have very--a lot of very-- limited expertise in this area so maybe someone else has substituted faster more refined devices in the Leach VAS with success. Jens has done an outstanding job to date and he knows way more than I on this subject.
I think I will leave this alone for now. The BCxx devices will be more than adequate at this time. So much to learn and at my age. I should know better than to tweak something that is not broken.
Thanks for the help pooge. Let me know how your experiment turns out.
Tad
There is no on line calculator that may be of help here?
We know what works and we know the parameters of the knew device. I have very--a lot of very-- limited expertise in this area so maybe someone else has substituted faster more refined devices in the Leach VAS with success. Jens has done an outstanding job to date and he knows way more than I on this subject.
I think I will leave this alone for now. The BCxx devices will be more than adequate at this time. So much to learn and at my age. I should know better than to tweak something that is not broken.
Thanks for the help pooge. Let me know how your experiment turns out.
Tad
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.