| pollypocket |
I am looking for some opinions....
I recently acquired a commercial Leach amp: The Leach Amp manufactured by LSR&D back in the 80's. It is based on the Leach Low TIM amp design. The specs are fairly impressive, and the power supply is massive. I have had the opportunity to listen to a Leach Double Barrelled amp in a fairly high end system and was impressed with the sound. Listening to this amp for a short period left me disappointed. The sound was veiled and undetailed. The sound stage was almost non existent.
I am driving a pair of Martin Logan Aerius electrostats, but it shouldn't be a problem for the amp. My source is a McCormack CDP and a Curcio Daniel pre amp. My system amp is a Radford ESA225 tube amp (25 wpc Class A) and the basis of my comparison.
So, are the Leach and Radford in two different classes and should I move on in my quest for better sound? Could the Leach be out of spec (it seems to have had a hard life before coming to me!).
Comments and suggestions welcome.
thanks,
Paul |
|
|
| djk |
That amp is at least 25 years old. It's junk, just send it to me.
A clapped-out one (like yours) went on eBay last month for $442, and a Heath-Kit Leach in the same kind of shape went last week for $572. A newer Threshold S200 only brought $595.
Replace the two 22µF power supply bypass caps per channel, the two 100µF caps in the front end per channel, and the 470µF cap in the feedback loop for each channel before you bad-mouth it. Then consider an up-grade of all the 0.1µF mylars to polypropylene. |
|
|
| pollypocket |
DJK
Thanks for your response. I wasn't necessarily bad mouthing the Leach, just stating the facts. My Radford is at least as old as the Leach and sounds excellent, albeit having the tubes upgraded recently.
I will replace the recommended caps and have another listen.
Paul |
|
|
| djk |
All the listed original caps were made by Callins. They sounded better than almost anything available, but they die after a short life, 7~10 years. If you like 'warm' bass leave out the 22µF caps.
The LSR&D 101 is one of my favorite amps. The latest version is even better, but is only available as DIY. |
|
|
| pollypocket |
djk;
I wasn't aware that LSR&D were still in business. Do they have a website?
thanks,
Paul |
|
|
| djk |
LSR&D isn't, that's why I mentioned DIY.
WM Leach still sells the boards for the v4.5, there was a group buy here at DIY Audio for a new version with plastic ouputs, it looks well thought out. There was also a group buy on a Krell design that is more similar than different.
http://users.ece.gatech.edu/~mleach/lowtim/ |
|
|
| Gold_xyz |
Only a question, on many amplifier I see a resistor that separates
Ground signal from Ground Chassis.
(as the leach) which are the advantages?
tnx ! :D |
|
|
| AndrewT |
Hi Gold,| quote: | | see a resistor that separates Ground signal from Ground Chassis. (as the leach) |
no, it doesn't.
Leach has a signal ground connection from PCB to audio ground and correctly runs a power ground connection from the PCB to audio ground.
These two routes could be quite long, so he has added a short route, with lower and consistent impedance, on the PCB between signal ground and power ground.
There is a mistake in the PSU schematic where he shows the same symbol for the safety earth and the PSU 0volt. This implies these should be connected. Don't.
Keep the safety earth separate from the audio ground.
Run a wire from PSU 0v to audio ground.
Run a wire from safety earth to a disconnecting network and then a wire from the other side of the disconnecting network to audio ground. |
|
|
| sreten |
Hi,
Hmmm.......
As this is a DIY site consider converting your speakers to bi-wiring
and then bi-amping them with both amplifiers, one of them with
adjustable gain.
just a thought ........
:)/sreten. |
|
|
| Gold_xyz |
| quote: | Originally posted by AndrewT
Hi Gold, no, it doesn't.
Leach has a signal ground connection from PCB to audio ground and correctly runs a power ground connection from the PCB to audio ground.
These two routes could be quite long, so he has added a short route, with lower and consistent impedance, on the PCB between signal ground and power ground.
There is a mistake in the PSU schematic where he shows the same symbol for the safety earth and the PSU 0volt. This implies these should be connected. Don't.
Keep the safety earth separate from the audio ground.
Run a wire from PSU 0v to audio ground.
Run a wire from safety earth to a disconnecting network and then a wire from the other side of the disconnecting network to audio ground. |
Hi AndrewT
Yes the question was the resistor between ground signal and
power ground (or output ground) R51 82Ohms
http://users.ece.gatech.edu/~mleach/lowtim/
Schematics link http://img166.imageshack.us/img166/8651/ampleachfq5.gif
bye. :D
|
|
|
| AndrewT |
Hi,
yes, we are both referring to the same resistor.
IT DOES NOT CONNECT CHASSIS TO SIGNAL GROUND.
Read my post again. |
|
|
| acenovelty |
Howdy folks,
This review from May 2000 may give some insight to pollypocket's observations.
http://www.hometheaterhifi.com/volu...s-i-5-2000.html
In particular:
"I found that the Aerius i's to be a very picky speaker where they are placed. Although proper calibration is a necessity for good sound with any loudspeaker, the MLs require three times the effort just to accomplish bearable sound. Because of the dipole nature of the electrostatic panel, the environment and positioning must be well prepared, for them to perform optimally. To this effect, Martin Logan produces a 23-page owner's manual with about half the content devoted to speaker positioning and environment. The manual includes recommendations for floor and wall types, acoustical wall treatment, distance from side and rear walls, distance from each other, distance from the listening position, and on and on."
Electrostatic panel with cone driver for low frequencies
ESL Panel Size: 40" H x 9"W
Cone Size: 8"
Crossover: 450 Hz
Lateral Dispersion: 300
Nominal Impedance: 4 Ohms - Min 1.7 Ohms at 20 kHz
Sensitivity: 89 dB/2.83 volts/m
A"veiled and undetailed" sound description give many of us flights of fancy bordering on the whimsical.
Since it sounds so bad driving those window panes, you really should unload this piece of junk for the price of shipping to member djk. He will give it a loving home.
Regards |
|
|
| Bob2 |
Gold_xyz
This is what Leach states about R51 under the power supply section
The circuit board has two ground leads, both of which connect to the central power supply ground. One lead grounds the signal reference points for the diff amp input stage. The other grounds the power supply decoupling capacitors and provides a ground reference for the protection circuit. R51 connects the two ground leads together on the circuit board. This resistor is small enough to look like a signal short circuit between the two grounds but large enough to force the currents in the two grounds to flow to central ground through the separate wires. This helps to prevent hum induced by power supply ripple currents in the ground system.
Bob |
|
|
| AndrewT |
Hi,
thanks Bob for confirming Leach's own words.
Note that chassis NEVER gets a mention in the grounding description.
I use my own names for the references to help avoid ambiguity.
Audio ground and safety earth. Keep them separate. |
|
|
| Bob2 |
Hi Andrew
Leach states the following under the construction section
The central ground point can be made with a #4 or #6 machine screw through the bottom panel with several #4 or #6 solder lugs and a nut over it inside the box. The solder lugs must make good electrical contact to the bottom panel.
I'm still learning this stuff myself, but that sounds like the central ground is the chassis and the safety earth connects to the central ground.
Bob |
|
|
| AndrewT |
| quote: | Originally posted by AndrewT
There is a mistake in the PSU schematic where he shows the same symbol for the safety earth and the PSU 0volt. This implies these should be connected. Don't.
Keep the safety earth separate from the audio ground.
Run a wire from PSU 0v to audio ground.
Run a wire from safety earth to a disconnecting network and then a wire from the other side of the disconnecting network to audio ground. |
Note what I said in post8.
If you want to avoid hum loops and buzzing coming from the PSU or from the mains then I very strongly suggest you do not connect chassis to audio ground. |
|
|
| Bob2 |
Andrew
Could you explain what a disconnecting network is. Like I said I'm still learning this stuff.
Bob |
|
|
| AndrewT |
Hi Bob,
the exposed conductive parts of mains operated equipment MUST be permanently connected to safety earth.
That rule NEVER varies.
It's there to keep you and other users alive!!!!
Except in approved, double insulated equipment which are designed to be safe without the third safety earth wire.
As always there has to be an exception.
With that safety message in the back of your mind, but not forgotten, because we must come back to it, the audio ground should be a clean reference for all the audio side of the amplifier circuitry.
But there is a risk of calamity if the audio side is completely separate from the safety earth. The failure senario is as follows:-
The input live connection breaks free and touches some other internal metal component (eg. PCB track) and makes it live!
Now pull out the input RCA and inadvertently touch the RCA socket barrel (the nice gold bit sticking out and just waiting for a toddler to touch). Explain that one to the fatal accident inquiry.
So we need to isolate the audio ground from chassis (and mains) to keep all the horrible noise and spikes and hum and all else we don't want sounding in our speakers while still keeping the metal box safe to touch when something goes wrong.
That's the purpose of the disconnecting network. To keep you alive. Not to make the amp work and not to make the amp quiet.
The various components used in the disconnecting network must be robust enough to blow the mains input fuse when the mains fault develops. These fault currents while the fuse is melting and before the arc extinguishes can run to many kA. yes thousands of amps.
The network and it's connecting cables must survive long enough to take the fault current to earth and keep the exposed parts at a voltage that is low enough to prevent electrocution. I believe this voltage is around 50Vac (when in non hazardous areas).
The disconnecting network can consist of a parallel set of some or all of the following:-
Power bridge diode connected in inverse parallel,
Power resistor,
Power Thermistor,
High frequency capacitor.
Some commercial equipment also add a switch to bypass all these components. I would only ever close the switch as a debugging tool to try to isolate/identify a miswired accessory unit. These switches are often referred to as earth (or ground) lift switch. They MUST leave the permanent connection between chassis and safety earth wire intact whichever position the switch is in. They must NOT break the safety earth connection. I will never fit the switch. |
|
|
| Bob2 |
Andrew
Thanks for the explaination, very informative. Now as Leach stated, would R51 then qualify as the disconnecting network, seperating the two grounds, power from signal.
Bob |
|
|
| AndrewT |
Hi bob,
I don't think so.
It is certainly not the safety connection, but if one bolts the audio ground to the chassis that obviates the need for the safety disconnecting network.
However, there is still a connection from signal ground to chassis and power ground to chassis. I think this may give rise to buzzing and mains borne interference at the amp output.
But worse, as soon as one connects a mis-wired ancilliary with the safety earth connected to the audio ground you are guaranteed to close an earth loop and that in turn guarantees hum (a lot of it) at the amp output. The creation of an earth loop requires two pieces of mis-wired equipment. Don't go half way there by mis-wiring your power amp when it is clearly avoidable.
Leach explains why he has put the signal to power ground resistor on board| quote: | | is small enough to look like a signal short circuit between the two grounds |
and this demands it is physically short to minimise inductance.
It is not doing the same job and was never needed to do the same job in his version of mis-wiring the chassis to audio ground. |
|
|
| Gold_xyz |
Thankyou now I have understood :)
On the pcb there is the connection of the mass of signal and the mass of power.
Both go connected to the central ground.
(to see the PTH...)
See you soon :D |
|
|
| AndrewT |
Hi,
is there a reason to use version 4.4 rather than 4.5? |
|
|
| samoloko |
I have build leach 4.5 amplifier with standard components
I like the sound but feel that It can be better
Would some of you which have done upgrade of components to the amplifier tell me what components need to be changed (for better sounding) compared to http://users.ece.gatech.edu/~mleach/lowtim/part3.html
Your choise of transistors, condensators and etc |
|
|
| samoloko |
OK no reply
if you have a minute wright just the type of C6 you are using
and optionaly type of the potentiometer for volume |
|
|
| jacco vermeulen |
Samoloko,
the trick of the low-tim amplifier is to have a very linear amp, aka large bandwidth amp, with the lowest number of stages.
Which is achieved by employing a triple darlington output stage, the combined high current gain enables the relatively low fT output devices to perform at audio frequencies.
Secondly, by linearising the Vas stage with the aid of a fast small helper, Q8/Q9 in the Leach schematic.
Making it sound better means making the amp even faster or reduce the number of stages, while keeping it unconditionally stable.
Download a schematic of an early ElectroCompaniet if you like to know more.
What you can do is discussed a zillion times at the large Leach thread.
Just tossing in some fancypants parts will not make it sound any different.
C6 should be non-polar, as advised by Prof Leach, or leave it out and use a quality DC-servo.
A stereo blue velvet Alps RK27 still is best value for money, anything better costs much more, you are the only one who can look inside your wallet. |
|
|
| tryonziess |
| Hi, After assembling my Leach amp I found that increasing the bias made a major improvement in the sound quality. Dr. Leach has some comments on the Leach website about this. Watch out for the heat and give everything time to settle in on each attempt at adjusting the bias. It can take awhile for the outputs to stabilize. Also, allow good air circulation inside the enclosure not just around the heatsinks fins. Tad |
|
|
| samoloko |
| 10 K , 50 K , or 100K Potentiometer Is best to use for volume |
|
|
| AndrewT |
| quote: | Originally posted by samoloko
10 K , 50 K , or 100K Potentiometer Is best to use for volume |
depends on what your pre-amp is capable of driving. |
|
|
| samoloko |
I am not using preamp at now
cd - amp |
|
|
| AndrewT |
| I'll rephrase it to what can your source drive? |
|
|
| tryonziess |
I have a 100k potentiometer on my Leach. It does not do to good. Dr. leach recommends a 10k pot in one part of the TIM article.
However, Like Andrew has stated it depends in part on the preamp and source. I am driving from a CD player with passive preamp --- just the pot. Tad |
|
|
| Andy L. Francis |
| quote: | Originally posted by samoloko
10 K , 50 K , or 100K Potentiometer Is best to use for volume |
A standard 47/50 kOhm volume pot should work fine! You could try 20/22 kOhm too. |
|
|
| pollypocket |
Hello again;
As per the original thread, I have a commercial Leach amp: The Leach Amp manufactured by LSR&D back in the 80's. In the course of having the amp re-capped, it became evident that the power supply electrolytics are no longer available in the same size as originally supplied in the amp. The problem is that the space available in the case only allows for the same diameter and length as the original (it is a very tight fit). Any suggestions on where to look for caps in the original size? I would like to replace them, as I am sure they are getting tired.
thanks,
Paul |
|
|
| djk |
| Do we have to guess the size, or are you going to give us some clues? |
|
|
| pollypocket |
djk
Sorry mate, I thought the amp and parts were common knowledge. I will pull the lid off again and get my calipers out. Stay tuned.
thanks,
Paul |
|
|
| djk |
| I think they're 2-1/2" dia, but I don't know the length. |
|
|
| duekfx |
| Does anyone have a list with decent parts? Im having trouble getting those out-of-date parts listed. I was thinking about MJL4281A (NPN) and MJL4302A (PNP)(end-stage), and MJE15034 (NPN) and MJE15035 (PNP) (drivers). Not sure about them, and I have no clue how to replace the others. |
|
|
| jaycee |
The drivers are fine. Flat pack devices are not recommended with this PCB, but there is a redesign by Jens Rasmussen that allows this: http://www.delta-audio.com/Leach-Clone.htm
MPS8099/8599 are available from On Semiconductor. If you are having trouble getting them then you can use MPSA06/56, they are not as good, but they will work.
I suspect the devices you are having trouble with are the 2N3439 and 2N5416. For these, MJE340/350 is often substituted but they are a different package and not easy to use with Leach's PCB. |
|
|
| tryonziess |
If you have a spare 33.00 Advanced Circuits will make you a couple of proto boards in a couple of days. Kind of like a special for small orders.
Then you can download the gerbers from Delta-Audio.com ie, the Jens Rasmussen boards in either the 6 or 10 output devices and send them to Advanced Circuits. These boards use modern devices and are VERY well layed out and constructed.
I have built 2 fantastic Leach amps from these boards and can personally attest to how well they sound. I am currently running 1 amp bias per rail with 70 volt rails. The sound is amazing.
Just something for you to think about. They really turned out nice. Tad |
|
|
| duekfx |
| Thx for the replies, gonna check that. |
|
|
|
