Discussion in 'Headphone Amplifier Measurements' started by atomicbob, Apr 8, 2018.
Any chance you'll test out some of the more popular LPS like Plixir, Teralink, HDPLEX & Uptone?
Most of those devices don't have the required voltage rating.
Not unless they make a 28Vdc 1.5 Amp or larger LPS suitable for use with the mcth. Maybe another thread. But there are other measurements in the queue awaiting posting and more devices to be measured.
Sorry, I should have specified the SMPS supplies, agree that the LPS should behave unless pushed beyond their rating.
mcth power on inrush current measurements
1. keysight 34461A with Benchvue
2. Electro Harmonix 6922 tube from TubeDepot (replacement for first EH6922)
3. mcth SMPS
4. BK 1627A LPS
mcth SMPS 28V power on inrush to normal run time current
Watch the Y axis on this one. It is 100 mA/div instead of 200 mA/div used on the following graphs. Also I did not zero the baseline.
It would appear the mcth SMPS has a soft start mechanism limiting the inrush current peak to approximately 700 mA
mcth SMPS 28V power on inrush current rise time
mcth SMPS 28V power on inrush current to nearly stable current time
BK 1627A LPS 28V power on inrush to normal run time current
With an LPS the mcth exhibits a little over 1.254 A with the EH6922 used for this test.
BK 1627A LPS 28V power on inrush current rise time
BK 1627A LPS 28V power on inrush current to nearly stable current time
Ah, that makes more sense. I'll try to work in the 1.0 amp load if time permits this weekend.
Bob, given your inrush measurements here, do you feel a 1.5A LPS has sufficient "wiggle room" (headroom? I'm not sure the technical term to use here)?
Using a 2, 3, or 5A+ LPS would be overkill rather than potentially damaging (ie: the amp will pull what amperage it wants at 28VDC)?
Margin or headroom are the typical terms used by engineers. If sticking with the stock EH6922 then it is probably ok. I went to 2.1A on the Acopian for additional margin. My two lab supplies will constant current limit at 1.25 A during power on but they don't seem to hit constant current mode at 1.3 A, for the particular EH 6922 I have.
Note: I only purchased a 3A model, A28MT300A, because I got a good used deal on a tested unit. The idle current seems to be lower than .4A. Double the onrush current is quite overkill and doubt it will provide any advantages sonic or otherwise. Like I said, I got a good deal (hint: there are still some out there)
I doubt the fans will kick on on my unit, unless they are just on all the time like a computer power supply.
Looks,like you can't buy the terminal strip protector separately from Acopian. I might could call them. I need some cables and a barrel connector more badly at the moment. I foresee a trip to the hardware store in my future.
I might have missed it, but what sized DC plug is used for the MCTH?
It is a standard 5.5mm x 2.1mm male dc connector:
Just picked this up at the hardware store along with a three pronged power cable with open tails. Looks this Sparta on electrical coating exactly what I need to keep this LPS safe.
You actually can order the terminal strip protector separately. You just have to give them a call and place the order over the phone.
Just be careful about the shipping. I was dumb and didn't ask for shipping prices (I just asked for UPS) and ended up paying $20 for shipping for a $5 item.
I think that stuff is designed to protect the exposed terminals from problems like corrosion, which isn't going to be an issue for an LPS used at home. Rather than to protect the user against electrical hazard.
Low cost Chinese 28V LPS ripple and FFT measurements.
Full description: Finished 65VA Ultra Low Noise Linear Power Supply 5V 9V 12V 15V 18V etc
This 28V LPS measured was acquired through ebay for $125US including shipping.
The same LPS is also available through AliExpress for $90US + $33.61US shipping.
Test equipment setup same as listed in first post.
Abbreviations used in labels below:
ULN = Ultra Low Noise
LPS = Linear Power Supply
Ripple 28V 56R load 20mV/div 5mS/div 100KHz - 65VA ULN LPS
FFT 28V 56R load 16MHz - 65VA ULN LPS
The 28V 65VA ULN LPS works well with the mcth. I think I hear a slightly better, very low frequency transient response with the BK Precision 1627A but the difference is minor.
This 65VA LPS appears to be well constructed, made with reasonable quality components. The threaded DC output connector is particularly robust. A great solution for those wanting an LPS without the DIY aspects of either the lab or OEM LPS.
It is indeed for protection against corrosion or other elements (fingers). It is essentially the same Flex stuff that annoying guy sells on TV.
One of the uses it mentions on the package is for sealing solder joints on a wire. It lasts longer than electrical tape. So I guess we can think of this as spray on liquid electrical tape, however it is very thin when applied and gets everywhere, so tape off what you don't want sprayed.
PS - I wound up putting my Cthulhu on the shelf next to the MCTH with the connectors out the back. So odd it vents to the side, but judging from the Acopian documents, it is designed to be side mounted. Weirdo industrialists.
Their spec sheet says 1,400V dielectric withstanding per mil (0.001") thickness, and they specify 5 mil minimum for insulation purposes, so if you follow the instructions, you'll be good.
@atomicbob, out of curiosity, what is the yellow rectangular thing sitting on the white plate?
My point was not that it is a poor insulator. It was rather that you do not want to rely on an incredibly thin layer of (sometimes transparent!) conformal coating applied from a spray can, to protect from shock.
The yellow thing is the load resistor.
As @Luckbad mentioned, it is a 56 ohm 50W resistor used to load the power supply during noise measurements. A resistor assures that all the noise measured is from the power supply. If the LPS was measured with the amplifier connected some of the noise measured would likely be from the amplifier.
Separate names with a comma.