diyAudio (
-   Solid State (
-   -   SMPS powered JLH 69 - quick and dirty results (

JZatopa 4th March 2013 11:19 PM

SMPS powered JLH 69 - quick and dirty results
I just wanted to share some things I found today while playing around. I recently bought a connexelectronics 48V SMPS500R-single to power a Pass Zen v1 amp that I have. Unfortunately when i tried the SMPS on the amp my outputs shorted internally(I am still not sure what happened) so no more Zen V1 until I figgure out the issue.

Today I decided to try my SMPS on a JLH 69 amplifier I assembled using MJ15003 outputs and an unregulated linear supply. With the linear supply the amp sounded good but it never sounded like something special. It was easily bested by my Decware Zen select when A/Bed on multiple occasions. It also always had a slight hum on my parker audio 98s (96db/watt/m) but you couldn't hear it from the listening position.

With the SMPS I had no hum and no noise from the speakers and it sounded better in every way. The soundstage was bigger, the sound was clearer (less veiled) and the bass was improved. The amp now sounds like a real high end amp to me. I have not A/Bed it with the Decware Zen select but if I were to make a guess I would bet they are now close to the same level but just different "flavors". I think I may order another smps with the proper voltage for my JLH and make this a permanent addition.

BTW i tried adding a 1mh coil between the smps and the JLH to clean up any noise from the smps. I previously tested this unloaded and the coil did lower the noise output on the smps. I found however that this had a negative impact on the sound of the amp. The bass seemed to be reduced and it gave the amp an overly forward sound. If I were to play with this again it would be with an inductor made of a much higher gauge wire. Honestly I really don't think it is needed.

I put this up so that any of you who still think SMPS is not for hifi. I think the technology is here and could be used to push some current amplifier designs even further. If you haven't played around with SMPS I suggest you give it a try.

nigelwright7557 4th March 2013 11:46 PM

One problem with an SMPS over a linear supply is its reaction to driving the amp too hard.
A SMPS will reset itself if too much current is drawn and you will need to power down then up again.
A linear supply will keep trying output current until eventually its fuses blow.
The linear supply will react better to short bursts of peak power.

This is what I have found in use.

JZatopa 5th March 2013 12:03 AM

The SMPS I am using can output 10A into 48V and the circuit I am using is class A so I haven't run into any reset issues. I will say that it seems like a SMPS seems to make things burn up quick if there is a problem and it doesn't see it as a fault/short. I believe this may be due to their ability to provide a large amount of amps in a very short period of time.

nigelwright7557 5th March 2013 12:10 AM

I would guess it is down to the over current setting.

The SMPS I built used a 2 volt limit across the lower mosfet to determine when reset mode occurs. This along with the mosfet Rdson determines reset current trip.

If your 10 amp supply is well inside what your amp needs to destroy itself then thats what will happen.

An SMPS will not give much 50Hz out as the switching frequency will be a 100KHz or so.
If the primary side smoothing is a bit poor then 50Hz can get through.

JZatopa 5th March 2013 12:56 AM

I'm in the US so it would be 60hz :) but I have negligible hum and noise. Really I was surprised it worked so well.

Now if I could figure out why it blew the outputs on my zen v1.

nigelwright7557 5th March 2013 01:01 AM

If the SMPS put out different voltages to what the zen was designed for then it might have affected the bias setting and fried the output transistors.

JZatopa 5th March 2013 07:05 PM

I need to check the bias circuit closely then. I just looked at the datasheets for the active devices to see if they could handle the V. I should have looked at how much current would be going through them as well, a careless mistake on my part.

hiktaka 30th March 2013 01:28 AM

I'm always fascinated about SMPS usage for pure class A amps, assuming it is done right (not for skimping purpose, as most SMPS are).
One important note is that SMPS shouldn't run between loaded/unloaded state. On average, it wants a rather constant load. For class A, this isn't much of a problem. For class A/B or Gainclones, tho, it means no good.
For class A/B and Gainclones, you can bleed SMPS output for about twice or three times the bias current, so that SMPS won't go unloaded.

Loudthud 30th March 2013 03:13 AM

Some SMPS's don't like big capacitive loads. They may have several hundred uF on the output for filtering, but regulation might be upset if too much is added at the point of use. Does your amp have a big cap near the output transistors?

What's really needed is a regulation and stability test where frequency sweeps are run through the amp at several power levels and the output of the SMPS monitored.

Doug Kim 14th July 2013 02:30 PM

I just finished 5200-based JLH 1969 this morning, using 30V-8A SMPS. It might be premature, but the sound is pretty satisfactory.

I have a few questions:
1. There is absolutely no noise when I use Ipod as the source, but there's some high-pitch noise when I use DAC. It is not 60Hz AC noise. The frequency is higher. What can this be?

2. I used very cheap parts for the construction, but still sound quality is very good. Is there anyone who uses JLH 1969 as their main amp? Then what are the input and output capacitors?

3. How's 1969 version compared to other versions such as PLH, 1996, etc?


All times are GMT. The time now is 08:03 AM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2