AndrewT said:
Sorry my bad C7=C12=470uF!
Regarding time constants, I think that the amp should be able to go as low as possible. I have lowered the low cut off freq to well below 1Hz (simulated) in order to have DC protection, while still play at 0dB at very low freq. What you decide is up to you, but as I will have a huge powersupply with lots of caps in it I think I can justify the choise of BW in the amp.
\Jens
Hi Pinco,
as far as I know lowering the PSU voltage will not degrade the sound quality. Others with more experience may be able to guide us on a minimum PSU voltage for the Leach clone.
A lower PSU voltage does have the advantage that it lowers the thermal stress on the output stage which is then able to deliver more current into difficult loads.
Could this be the reason that some amps sound less good when the manufacturer has upgraded the supply voltage to give more power?
Bypass caps:- I would go with 100nF and PP rather than 1uF PE,MKT etc.
Jens,
Your input RC time constant (RCtc) is about 10seconds. This is about 6 octaves lower than your PSU and NFB RCtcs. I assume you do not agree that the input RC should relate to the other RCtcs to prevent motorboating!
as far as I know lowering the PSU voltage will not degrade the sound quality. Others with more experience may be able to guide us on a minimum PSU voltage for the Leach clone.
A lower PSU voltage does have the advantage that it lowers the thermal stress on the output stage which is then able to deliver more current into difficult loads.
Could this be the reason that some amps sound less good when the manufacturer has upgraded the supply voltage to give more power?
Bypass caps:- I would go with 100nF and PP rather than 1uF PE,MKT etc.
Jens,
Your input RC time constant (RCtc) is about 10seconds. This is about 6 octaves lower than your PSU and NFB RCtcs. I assume you do not agree that the input RC should relate to the other RCtcs to prevent motorboating!
pinco said:
The factory is in the former Eastern Germany area, that is why.
They're not bad.
You need 40 volts on the rails for the Zeners.
For accurate voltage regulation the Zeners need some current, and R3/R22 need to deliver current for T2-T5.
I'd follow the advice and stay above 50 volts
Attachments
Here is a great transformer for this Leach amp. At 700VA, only one is needed for those trying to save some money on the hardware. 700va is plenty for two channels, and would have stupendous voltage regulation if used one for each channel or monoblock, which, at $50 each, is a great deal for 700 VA. I'm told they are also bifilar wound, which is good for balanced windings.
700VA toroidal transformer
700VA toroidal transformer
i am using an EI traffo posted here:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=16465&pagenumber=5
http://www.diyaudio.com/forums/showthread.php?s=&threadid=16465&pagenumber=5
Hi Guys
Sorry - I messed up a calculation earlier on and hence posted incorrect values for basic VI Limiting in the leach clone (as explained by Jacco).
Here is a spreadsheet I created for myself - it might save somebody else from making a hash of things.
In my situation described earlier R26 / 34 should be 5.8k.
Thanks Andrew for alerting me to this problem
Sorry - I messed up a calculation earlier on and hence posted incorrect values for basic VI Limiting in the leach clone (as explained by Jacco).
Here is a spreadsheet I created for myself - it might save somebody else from making a hash of things.
In my situation described earlier R26 / 34 should be 5.8k.
Thanks Andrew for alerting me to this problem
Hi Byrd,
Can I send you the modified Bensen spreadsheet?
It does not agree with your predictions.
Your protection locus is outside the 100mS SOAR when driving negative current and allows a full load current of about 56A into 0.9 ohms load. This is well above the 100mS SOAR.
However your numbers are close.
Try R26=13k and add R68=30k brings the locus just inside the 35degC SOAR.
These sugestions are still subject to trimming R28 to ensure actual short circuit current can be confiirmed =7.5A
Can I send you the modified Bensen spreadsheet?
It does not agree with your predictions.
Your protection locus is outside the 100mS SOAR when driving negative current and allows a full load current of about 56A into 0.9 ohms load. This is well above the 100mS SOAR.
However your numbers are close.
Try R26=13k and add R68=30k brings the locus just inside the 35degC SOAR.
These sugestions are still subject to trimming R28 to ensure actual short circuit current can be confiirmed =7.5A
OK - Just finished putting the boards together. Still quite a job - especially if you dont have a nice temperature controlled iron for the ground plane bits, but a HELL of alot less work and fiddly bits than the original Leach layout
Soundwise - Interesting as I was expecting it to be the same. I feel the highs (acoustic guitar arena) are slightly mellower but somehow more detailed and airy. It still seems to maintain its basic character though. Anybody compared the new amp to the original. Could this mainly be attributed to the higher bandwith transistors?
Anyhow - Thanks Jens for creating this PCB layout - It really was a pleasure to work with.
Soundwise - Interesting as I was expecting it to be the same. I feel the highs (acoustic guitar arena) are slightly mellower but somehow more detailed and airy. It still seems to maintain its basic character though. Anybody compared the new amp to the original. Could this mainly be attributed to the higher bandwith transistors?
Anyhow - Thanks Jens for creating this PCB layout - It really was a pleasure to work with.
Mikett said:
Think it has been long enough for others to give their opinion.
(T1), D1,D2,C3,C4 and R3 form a basic voltage regulator. Performance of regulator IC's has gone up considerably, a 317/337 would give more accurate voltage on the base of T1.
But the signal route passes between T1 and R2, you'd still have the effects of powersupply voltage variation on R2.
The ultimate gain would be the elimination of the zener noise.
If opting for better voltage regulation, the obvious choice would be for complete front end regulation, for which the boards are laid out for. The secondary choice could then be to exchange the zener diodes for an integrated regulator, (or discrete).
A crazy hat idea would be to exchange the zeners for an array of LEDs, with white or blue types the required number would be in the 10 to 12 range. A LED lacks the noise of a zener diode and is much more accurate.
Red low efficiency LEDs have gone out of fashion, big batches are offered frequently on the web for a few bucks. For a 40vdc setting the number of red LEDs required would be like 24.
Big deal if you've bought batches of 500 LEDs for a few bucks, as i did.
Nice bonus of using LEDs is that they also act as operation monitors.
I dislike zener diodes, they make noise and are often the cause of failure. Shame you can't get 40 volts drop LEDs.
We are already exchanging the 20 watt halogen ceiling lights in our home by high efficiency LED versions. LED lighting is the way of the future.
JFTR:
upgrading/modding factory equipment was the fashion in the late 1980s/1990s.
A lot of small audio companies made good money exchanging and adding capacitors, swapping dip8 opamps and installing voltage regulation boards.
The LED string is not a bad idea at all and can be used to advantage for cosmetic effects. Gets me thinking here. I could just use a prototyping board that has traces that would string leds easily.
As for Zeners, I have had two break down on me beore despite being overrated several times over. The worse thing though is the inaccuracy of the voltages, which produces offset issues.
As for Zeners, I have had two break down on me beore despite being overrated several times over. The worse thing though is the inaccuracy of the voltages, which produces offset issues.
My original leach uses the same zeners - it has been running for about 2 years now. It is on I would say at least 80% of the time.
As far as mismatched voltages go - you can match them - have a look at leach's site. Even without matching zeners / any of the transistors on my forst prototype board offset was only @ 60mv. This is now down to 8mV after doing a bit of matching.
OK now for my question. Terry mentioned earlier that it was difficult to desolder components. Damn he is 100% correct. Even with a solder sucker it is almost impossible to get out. I had to feed compnent legs through the holes while heating them to get the solder out.
Is there a better way of doing this - I have already pulled one of the pads off the board.
Nice haircut BTW Terry
<---- You can tell I have been trained to notice these sorts of things
As far as mismatched voltages go - you can match them - have a look at leach's site. Even without matching zeners / any of the transistors on my forst prototype board offset was only @ 60mv. This is now down to 8mV after doing a bit of matching.
OK now for my question. Terry mentioned earlier that it was difficult to desolder components. Damn he is 100% correct. Even with a solder sucker it is almost impossible to get out. I had to feed compnent legs through the holes while heating them to get the solder out.
Is there a better way of doing this - I have already pulled one of the pads off the board.
Nice haircut BTW Terry
<---- You can tell I have been trained to notice these sorts of thingsHere is a thread you guys may find helpful:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=35821&highlight=
http://www.diyaudio.com/forums/showthread.php?s=&threadid=35821&highlight=
Byrd said:My original leach uses the same zeners - it has been running for about 2 years now. It is on I would say at least 80% of the time.
As far as mismatched voltages go - you can match them - have a look at leach's site. Even without matching zeners / any of the transistors on my forst prototype board offset was only @ 60mv. This is now down to 8mV after doing a bit of matching.
OK now for my question. Terry mentioned earlier that it was difficult to desolder components. Damn he is 100% correct. Even with a solder sucker it is almost impossible to get out. I had to feed compnent legs through the holes while heating them to get the solder out.
Is there a better way of doing this - I have already pulled one of the pads off the board.
Nice haircut BTW Terry
Hi Byrd,
Thanks for noticing the haircut.
It's probably not the best way, but I ended up using a drill bit of the right size and just drilling out the solder after I got as much as I could with the solder sucker. Like I said, it helped me be more careful when selecting parts.
Blessings, Terry
Thanks - will give this a go this weekend if I need to - hopefully not again.
Never even knew there such things as de-soldering stations - but wow mega bucks here so will certainly attempt the solder wick suggestion first.
Maybe I'll even try the womans vacuum cleaner This hobby is getting more & more dangerous all the time might as well go all out.
Never even knew there such things as de-soldering stations - but wow mega bucks here so will certainly attempt the solder wick suggestion first.
Maybe I'll even try the womans vacuum cleaner
This hobby is getting more & more dangerous all the time might as well go all out.- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.