Updated the website with PDFs for the 6 transistor top and bottom layer.
Top is mirrored!
Bottom is not mirrored!
A new group buy is ok with me, but I will not be making any changes to the layout this time arround.
Feel free to download the gerbers for the group buy.
All the best
\Jens
Top is mirrored!
Bottom is not mirrored!
A new group buy is ok with me, but I will not be making any changes to the layout this time arround.
Feel free to download the gerbers for the group buy.
All the best
\Jens
Gerber files for leach amp 10 transistors
Hi jens, kindly upload your gerber files for leach amp 10 transistors?
thanks...
Hi jens, kindly upload your gerber files for leach amp 10 transistors?
thanks...
I don't think that Jens has offered to mediate a new GB.
He has put a lot of time and effort into previous GB's and it is perhaps someone else's turn to take the lead on this.
It's not really difficult. You could do this yourself. You just start a Wiki and invite interested members to add their names.
Andy
He has put a lot of time and effort into previous GB's and it is perhaps someone else's turn to take the lead on this.
It's not really difficult. You could do this yourself. You just start a Wiki and invite interested members to add their names.
Andy
Andy,
I updated the schmatics of the 6 and 10 transistor version
Feel free to use these during a GB....
I have moved all parts to the same page as my original multipage schmatics was not user friendly.
\Jens
I updated the schmatics of the 6 and 10 transistor version
Feel free to use these during a GB....
I have moved all parts to the same page as my original multipage schmatics was not user friendly.
\Jens
Here is a picture update of my project....
\Jens
An externally hosted image should be here but it was not working when we last tested it.
\Jens
Can anyone help with the following.
I am having a real hard time even finding output transistors MJL4302 and MJL4281 down under this side of the world.
The best our local supplier can do is MJL1302 and MJL3281 at wait for $12.00 US a piece!!! $144.00US + delivery AHHHH
I only need 6 x each MJL4302 and MJL4281. IF any of you guys building multipal amps with bucket loads of these lying around could spare a few? or can someone put me into contact with a reasonably priced supplier.
Cheers
Ian
p.s. my better half is getting sick of me searching the www each night for the bits i need.
I am having a real hard time even finding output transistors MJL4302 and MJL4281 down under this side of the world.
The best our local supplier can do is MJL1302 and MJL3281 at wait for $12.00 US a piece!!! $144.00US + delivery AHHHH
I only need 6 x each MJL4302 and MJL4281. IF any of you guys building multipal amps with bucket loads of these lying around could spare a few? or can someone put me into contact with a reasonably priced supplier.
Cheers
Ian
p.s. my better half is getting sick of me searching the www each night for the bits i need.
Head Office - Melbourne
Arrow Electronics Australia Pty Ltd.
9 -10 Bastow Place,
PO Box 623,
Mulgrave, Victoria 3170, Australia
Tel: (61) 3 9574 9300
Fax: (61) 3 9561 2148
They carry the On Semi product line where these transistors originate.
Sorry, got no pricing/availability for you.
You could try to get samples thru:
http://www.onsemi.com/PowerSolutions/home.do
If you sign up, they will sample you 25 each for the cost of shipping.
In the US that's $11.00 by FedEx overnight.
http://www.em.avnet.com/
also carries stock, check them for pricing.
Shipping down under could be costly from avnet tho.
Regards
Arrow Electronics Australia Pty Ltd.
9 -10 Bastow Place,
PO Box 623,
Mulgrave, Victoria 3170, Australia
Tel: (61) 3 9574 9300
Fax: (61) 3 9561 2148
They carry the On Semi product line where these transistors originate.
Sorry, got no pricing/availability for you.
You could try to get samples thru:
http://www.onsemi.com/PowerSolutions/home.do
If you sign up, they will sample you 25 each for the cost of shipping.
In the US that's $11.00 by FedEx overnight.
http://www.em.avnet.com/
also carries stock, check them for pricing.
Shipping down under could be costly from avnet tho.
Regards
Hi Slowhands,
that's a nice compact layout. Well done.
Suits a 100mm high chassis and sink, saving 25mm compared to the 3pair version.
I realise that you have squeezed a lot into half the board area, but could you manage to fit 3pair onto the same size PCB?
Any chance of posting an Eagle or PDF or PCB Express?
that's a nice compact layout. Well done.
Suits a 100mm high chassis and sink, saving 25mm compared to the 3pair version.
I realise that you have squeezed a lot into half the board area, but could you manage to fit 3pair onto the same size PCB?
Any chance of posting an Eagle or PDF or PCB Express?
Hi,
I just ran some numbers on a 2pair MJL4281.
Tc=34degC 35Vac 600VA (for 2channel) transformer +-45mF/channel.
120W into 8r @ 60deg phase angle, one channel driven.
210W into 4r @ 53deg phase angle,
250W into 3r @ 45deg phase angle.
safe into 2ohms resistive for testing.
Looks like this would drive any 8ohm speaker and should cope with 4 to 8ohm that have become popular.
Should work with most 4ohm in a domestic environment.
I just ran some numbers on a 2pair MJL4281.
Tc=34degC 35Vac 600VA (for 2channel) transformer +-45mF/channel.
120W into 8r @ 60deg phase angle, one channel driven.
210W into 4r @ 53deg phase angle,
250W into 3r @ 45deg phase angle.
safe into 2ohms resistive for testing.
Looks like this would drive any 8ohm speaker and should cope with 4 to 8ohm that have become popular.
Should work with most 4ohm in a domestic environment.
Id like boards also, but would like to see the schmatics and the PCB traces first.
I would also generate footprints with larger pads to ease soldering.
\Jens
I would also generate footprints with larger pads to ease soldering.
\Jens
The board is very tight, but could fit in a third output pair if the drivers were brought on card and heatsinked there. I chose not to do so for several reasons: size, cost, simplicity of layout, thermal density and confidence that the existing pair is sufficient.AndrewT said:
My main goal in this project was incorporate the best plastic packaged output devices available soldered directly to the board, hopefully leading to a more repeatable implementation than a hand wired output section using older TO3 metal packages. I also wanted to mount everything on a vertical flat surface, with the output devices spread out as much as possible for better heat spreading. I analyzed the safe operating area of the new MJL43XX output devices and believe that two pair is quite sufficient ---at the design targets of voltage (56v) and load (4 ohm minimum). In fact their SOA is better than the originals', and their extended beta is superb, making life easier for the driver. All other components in the design can handle much higher voltage. All caps are 100v, and transistors run at 1/2 their rated voltage or less. But the real subtle issue is risk of second breakdown of the outputs if we take voltage significantly higher.
To be audible, the power would have to double, and that means increasing voltage by about 41%. You can't do that reliably in the non-cascode output configuration with the current state of the art bipolar devices (no problem for MOSFETs, but that is another subject). So I prefer this conservative, reliable design to one that takes risks for marginal return. We know we have some reserve for high line voltage, reactive loads, and high ambient temperature and that's a good thing.
I really think if you want significantly more power, don't mess with a small increment -- go for the Leach Super Amp and quadruple power. You can conservatively quadruple power using that approach, if you are willing to pay for quadrupling power, heatsinking and of output device quantity and cost. You get what you pay for, but you do have to pay.
As for publishing the design, I probably won't broadcast it because I don't want it copied by commercial interests. I will share it with registered members of this community freely. I've thought about a Group Buy if there is interest. Since that would be non-profit, I would want it to be very low overhead for me and safe and reliable for everyone. I would want full accountability for both buyer and seller. I do have big concerns with international shipping, based on experience. These are issues I would want to have a good grip on before going into that. Also, it would be a good idea to encourage our Company Store to carry the MJL4381A and MJL4302A output devices, so anyone could buy them conveniently (I buy from a distributor nearby for about $2.50, so I am shocked at the prices elsewhere for these).
In the meantime, there is nothing at all wrong with the Leach Amp original board sold for $12.50 each by Dr. Leach, other than the packaging and wiring complexity. If someone is impatient, that is an easy way to get instant gratification.
Hi Slowhands,
I agree with everything you said except one:
Look at the 20V & 5V vce vs DC current gain. With both the NPN and PNP the lower voltage 1302 & 3281 has about twice the current, or better, before the graph falls significantly. However the absolute gain is better with the newer pair, but they can't maintain even the same gain to those high currents.
In just about every other parameter the newer higher voltage BJTs do better.
A group buy or private Emails to DIYaudio members will not exclude lurkers out to steal your layout. They will get it and reproduce it at will, if they can see a profit for themselves.
The power predictions in my later post fully confirm the obvious ability of 2pair that you had foreseen.
I agree with everything you said except one:
MJL4281/4302 is not seen as good as MJL3281/1302.
Look at the 20V & 5V vce vs DC current gain. With both the NPN and PNP the lower voltage 1302 & 3281 has about twice the current, or better, before the graph falls significantly. However the absolute gain is better with the newer pair, but they can't maintain even the same gain to those high currents.
In just about every other parameter the newer higher voltage BJTs do better.
A group buy or private Emails to DIYaudio members will not exclude lurkers out to steal your layout. They will get it and reproduce it at will, if they can see a profit for themselves.
The power predictions in my later post fully confirm the obvious ability of 2pair that you had foreseen.
- Status
- Not open for further replies.