kind request for help:
Before I find myself running around in circles I suggest that a BOM is made. This should include housing types and footprint information for ALL parts and give me a better chance to create the PCB artwork that will fit most of our needs.
I have in the past just made the layout to my own requirements, but as this was supposed to be a group effort I would like to see some volunteers to help me with this work. This will also mean that one or two test builders are needed to verify and debug the PCB design. I have plenty of other projects to attend to and therefore kindly request some help.
I don’t want to spend my own money prototyping (again) for a group buy so let’s see some hands!
Thank you
\Jens
Before I find myself running around in circles I suggest that a BOM is made. This should include housing types and footprint information for ALL parts and give me a better chance to create the PCB artwork that will fit most of our needs.
I have in the past just made the layout to my own requirements, but as this was supposed to be a group effort I would like to see some volunteers to help me with this work. This will also mean that one or two test builders are needed to verify and debug the PCB design. I have plenty of other projects to attend to and therefore kindly request some help.
I don’t want to spend my own money prototyping (again) for a group buy so let’s see some hands!
Thank you
\Jens
Hi,
Well put Jens.
Doing the PCB layout is more than enough workload. You contribute plenty to this forum.
Well put Jens.
Doing the PCB layout is more than enough workload. You contribute plenty to this forum.
hello jens,
just currious, why did u mount T9+T9A then T13+T13A to a heatsink? i think these are metal cans or do you see other options?
roland
just currious, why did u mount T9+T9A then T13+T13A to a heatsink? i think these are metal cans or do you see other options?
roland
Hi,
can someone confirm the quiescent currents through the VAS, predriver and driver?
Then we can see the power dissipation of each.
I would prefer to see less than 25% of power rating used at quiescent current for each active device and it's cascode partner.
What was Leach's opinion on this?
can someone confirm the quiescent currents through the VAS, predriver and driver?
Then we can see the power dissipation of each.
I would prefer to see less than 25% of power rating used at quiescent current for each active device and it's cascode partner.
What was Leach's opinion on this?
rolandong said:
I just added the room for a sheet of alu to ensure enough cooling if someone should run the amp with high bias or use it for a small PA.
The extra cooling will not be needed under normal load conditions.
\Jens
with the super leach amp, this will be the least of my concerns, these transistors are stacked on top of each other and will see smaller voltages at rails of +/-85 than with the lo-tim amp version with rails of +/-65 volts.😀
SOAR - VI protection
Hi,
Jens,
is there space to add a pair of caps (upto 100nF) in parallel to r28 & r30? to allow provision of a time delay for short term peak currents before triggering the protection.
They would be optional as with the whole protection scheme.
Hi,
Jens,
is there space to add a pair of caps (upto 100nF) in parallel to r28 & r30? to allow provision of a time delay for short term peak currents before triggering the protection.
They would be optional as with the whole protection scheme.
Hi Tony,
The reason for the Rolandong asking about sinks and my prompting for Iq is the possibility of heating the transistors and then needing a sink or bigger sink as an afterthought.
Keep in mind that 42V * 3mA = 25% of 500mW (some To92) rating.
If my suggestion for a To92 power limit relative to rating, then we need to account for Iq. Then use it to determine the need/size for sinks for each of VAS and predriver and definately driver.
Has anyone a view, regarding performance and reliability, of how close to maximum power rating we should take our active devices when drawing quiescent current?
The reason for the Rolandong asking about sinks and my prompting for Iq is the possibility of heating the transistors and then needing a sink or bigger sink as an afterthought.
Keep in mind that 42V * 3mA = 25% of 500mW (some To92) rating.
If my suggestion for a To92 power limit relative to rating, then we need to account for Iq. Then use it to determine the need/size for sinks for each of VAS and predriver and definately driver.
Has anyone a view, regarding performance and reliability, of how close to maximum power rating we should take our active devices when drawing quiescent current?
rolandong said:
hi,
these TO-39 transistors were the best leach could find at the time he designed the super leach amps. nowadays, they are almost obsolete. in the 80's i used japanese 2sc510/2sa510 or 2sc484/2sa484 transistors as substitues.
today, MJE340/MJE350 are used in a TO77 package, thus a flat heatsink will do
there are othere japanese transistors 2SC4793/2SA1873 2SD669/2SB649 that can be used.
at biased refference of 40volts and tail resistors of 10k,with a tail current of 4mA, 2mA current therefore flows accross cascoded front end to92's, therefore quiscient dissipation at 42volts of 84mW which is actually 16.8% of Pdiss rating.
jens changed this to 12K in the lo-tim, you can change this to 15k if you like. keep in mind that increasing this value decreases the currents at VAS collectors lowering their quiscient dissipation commensurately.
Hello Tony,
Far from being obsolete, a two minute google search led to 10's of thousands of 2n5416 and 2n3439 devices. One on-line source offered them for $.18 each in small quantities up to several thousand. Here in sunny northern California, one can buy them at most retail or surplus electronics stores any day of the week. Northern Philippines, well that may be somewhat different. Nothing wrong with substitutions, but the death of parts to build the amp as described by the Professor is premature.
MJE's run about $.43 at mouser and digikey.
Regards
Far from being obsolete, a two minute google search led to 10's of thousands of 2n5416 and 2n3439 devices. One on-line source offered them for $.18 each in small quantities up to several thousand. Here in sunny northern California, one can buy them at most retail or surplus electronics stores any day of the week. Northern Philippines, well that may be somewhat different. Nothing wrong with substitutions, but the death of parts to build the amp as described by the Professor is premature.
MJE's run about $.43 at mouser and digikey.
Regards
Ace,
When I build my first Leach amp I searched for the original parts but was unable to find them here.
So I changed to parts that a easy to get (at least here)
I later found them here, but they are more expensive than the MJE340/350 types.
\Jens
When I build my first Leach amp I searched for the original parts but was unable to find them here.
So I changed to parts that a easy to get (at least here)
I later found them here, but they are more expensive than the MJE340/350 types.
\Jens
hi ace,
i stand corrected, but those are not generally availble anywhere in manila except, thru importation. and i am afraid if ever they are found, they are either remaked or fakes.
better to use japanese parts, higher Ft's and easier heatsinking are reasons to change.
i stand corrected, but those are not generally availble anywhere in manila except, thru importation. and i am afraid if ever they are found, they are either remaked or fakes.
better to use japanese parts, higher Ft's and easier heatsinking are reasons to change.
- Status
- Not open for further replies.