Bench power supply ringing/resonating with test signal

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have an AIM-TTI EX354RD dual bench power supply powering a +-35V prototype power amplifier. When I run a test sine wave through the amplifier I can hear the power supply ringing/resonating with the test signal. The ringing follows the amplitude and frequency of the test signal and it is fairly loud and annoying. This does not seem normal to me. Have any of you experienced this kind of thing before?
 
Yes I have a 300 MHz scope and I don't see any oscillations whatsoever. The current reading on the bench supply looks normal. The ringing becomes audible when the supply is drawing around 500mA per rail and gets worse as I increase the test signal (and the supply draws more current). At the onset of clipping the supply is drawing around 2A per rail.
 
I have an AIM-TTI EX354RD dual bench power supply powering a +-35V prototype power amplifier. When I run a test sine wave through the amplifier I can hear the power supply ringing/resonating with the test signal. The ringing follows the amplitude and frequency of the test signal and it is fairly loud and annoying. This does not seem normal to me. Have any of you experienced this kind of thing before?

My guess is it is an electromagnetic component that "sings" as the result of the lower frequency (audio band) current being handled by that electromagnetic component. It should not be very loud at low frequencies (100Hz).

Suggestion: Remove the cover and put a rubber-glove on your hand. Then, put a signal on your amplifier such that you have a clearly audible response from the SMPS. When touching (with the rubber glove) the chokes/transformers you should be able to locate the component that sings the most.
 
Hi ,i just shortly looked at those power supplies catalog datasheet ,they says this supply combines linear regulator and switching type pre-regulation ,which is actually singing in your case .You can perform small test ,try to load both channels ,lets say 0,5Amp ,but load must be constant, a resistor or some lamps .It should be then silent . So what you hear ,audio singing ,means that switching type pre-regulator reacting fast to load changes ,meaning load current is changing in wide range .So if you would like to reduce that singing sound ,you can add 10000 uf to each output ,like in simple 50hz unregulated supplies ,then you will hear singing only at bass frequencies. Amplifier will draw fast changing current from capacitors ,and when they have not enough energy ,from power supply .In short ,increase capacitors , and you will have similar effect like LPF for subwoofer . I made some kind of similar supply ,and there in some cases even ferrite rings can sing ,but not loud .
 
Is it possible the amplifier OR very often it's the dummy load that audibly sings.
If the bench SMPS supply was misbehaving badly.. IS IT? Then I think you could do more troubleshooting E.g is it near the current limits etc. The right tool to do this is an O- scope. EDIT> Sometimes a DSO can be misleading.. do always check the lower time base settings too.
 
Last edited:
SMPS and linear -- got a lot of control loops which may be badly compensated. The amplifier circuitry may be modulating one or more of these.

Bad error amplifier circuitry in the linear regulator control loop -- here's a simple test -- with a suitable resistor as dummy load, inject a 1kHz signal onto one of the power supply rails and take an FFT. If you see a lot of harmonics bingo.

There's no easy fix for that type of problem.

When testing amplifiers I use a variac and a standard bridge rectifier with 20,000uF capacitors (and a bleeder resistor).

Here's a failed unit tear-down on youtube: YouTube
 
Last edited:
AX tech editor
Joined 2002
Paid Member
I was shouting 'the diode, the diode' until he finally heard me ;-)

Funny they call it 'an eBay power supply' while it is clear it is a TTI product which is a well regarded UK company.
There are a lot of almost-identical looking far east implementations around, but TTI is a great company.
Still, parts do fail occasionally of course as proven here.

Jan
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.