Can using an active sub cause HF drivers to blow???

[jdurango]

Paul, good point!

Also, don't want this to turn into a JW thread. As far as "upgrading" goes in the home hi-fi and pro audio world, I think you can do a lot worse than Jim Williams. His work is highly regarded and reviewed by some legit professionals and his client list speaks for itself. Again, all I care about is the end result, and I can tell you this Adcom amp is definitely an improvement over the stock implementation (again, I'm coming from a Bryston 4B and have also owned a stock Adcom 535).

Anyhow, I guess I'm still not sure how to check that the RF levels are at a safe level. Obviously there will be SOME level of RF, which is perfectly safe. Are we talking under 5% of nominal levels in the 10-20K range? 20%? 50%? Is removing RF protection as simple as bypassing an inductor somewhere in the amp?

Since the monitor is taken apart, I can directly probe the leads going to the HF driver as well.

 

[Gold]

Anyhow, I guess I'm still not sure how to check that the RF levels are at a safe level.

I’d put a 1k test tone in and look for hair on the scope. I’d guess under a millivolt or two would be okay.’

Every time I have heard of a JW mod being removed it was for reliability issues. Not sound issues.

 

[jdurango]

Upon further inspection, I noticed that it's a GFA545, NOT 555....so only 100W of power per side @ 8ohms. So yeah, really not sure what's going on here. I'll run signal through with my trusty HP function gen and check for noise with my trusty Tek scope on the morrow! Thanks everyone!

 

[JohnRoberts]

RF/HF oscillation is only one of several possible fault vectors. Easy to see with a scope, often audible as a "frying" sound (caused by rectification).

The most obvious is just cranking the volume too high.

How much power (SPL) is the active sub? If the sub is very powerful you may be tempted to try to crank up the amp to keep up, inadvertently clipping the amp. 

A 100W amp can still make a lot of power driven into saturation. A square wave can deliver 200W from a 100W amplifier. Besides 2x the power, the clipped signal contains excessive HF energy.

Last but not least maybe ask the speaker manufacturer.  We clearly do not know...  :

JR

 

[jdurango]

Again, no clipping was ever apparent at any time through any measurement apparatus, either audibly, visually via clip indicators on the amp, or from the DAW outputs. Dealing with USA tech support, they're telling me some studios are replacing both their HF AND LF Amphion drivers every 45 days. That.....is......insane. Time to move to ATC's!

Anyone want a deal on a pair of lightly used Amphion One18's

 

[trobbins]

Again, no clipping was ever apparent at any time through any measurement apparatus, either audibly, visually via clip indicators on the amp, or from the DAW outputs.

Can you elaborate on your measurements and apparatus connected to the amplifier output?

The 545 has a distortion alert indicator - is that what you are referring to?  That indicator doesn't monitor the output of the amplifier, only the input stage signal level.

 

[JohnRoberts]

Again, no clipping was ever apparent at any time through any measurement apparatus, either audibly, visually via clip indicators on the amp, or from the DAW outputs. Dealing with USA tech support, they're telling me some studios are replacing both their HF AND LF Amphion drivers every 45 days. That.....is......insane. Time to move to ATC's!

Anyone want a deal on a pair of lightly used Amphion One18's

Is that the amplifier tech support, or the loudspeaker tech support?

It is not unheard of for vendors to do the finger pointing dance to shift blame. If that is loudspeaker tech support, what are they smoking (drivers apparently)?

JR

 

[jdurango]

Can you elaborate on your measurements and apparatus connected to the amplifier output?

The 545 has a distortion alert indicator - is that what you are referring to?  That indicator doesn't monitor the output of the amplifier, only the input stage signal level.

Amp is configured with a signal balancing and impedance matching buffer (henry engineering matchbox ii). It is configured so that a +4dBu 1k sine at the balanced input yields -10dBu  at the unbalanced output, which is connected to the amp.

Yes, the visual meters I'm referring to are the ones on the front, but if the input wasn't clipping, I'm not sure how the output would be clipping unless there's a serious problem with the design of the amp....and it's even more puzzling how to output could more than DOUBLE (to exceed the 200W rating of the speakers) or distort so badly to deliver adequate DC to blow the drivers with zero audible distortion.

I'm going to hook the amp up now to look for DC or RF on amp output, as well as output from the xover. Will report back.

 

[jdurango]

Okay, I used an old 260W 8 ohm PA woofer as a load.

Pics here: https://drive.google.com/drive/folders/1Doc1K3k3WiEg-Eulxei8Tgw0_yrVQYPz?usp=sharing

Signal is as follows:

HP Function gen set to 1k @ 300mvPP (and louder was just too taxing on the ears and I don't have a "quiet" high wattage 8 ohm load laying around)

Two way split BNC on function gen output.

One split is BNC to RCA cable attached directly to amp input.

Other split is BNC to BNC attached to Tek 2232 Chan B Input

Tek 2232 Chan A input is attached to Tek 100mhz 1x probe attached to speaker + and -

And of course amp output is also attached to speaker + and -

Took several pics at different resolutions in the HF to ultrasonic range. The left or top trace is always Chan A (from amp), the right or bottom trace is always Chan B (from function gen) Again, pics are here: https://drive.google.com/drive/folders/1Doc1K3k3WiEg-Eulxei8Tgw0_yrVQYPz?usp=sharing

As you can see, there is no noticeable "fuzz" on the amp channel until the very highest res (20mv @ 1microsecond)....there appears to be maybe 5mv of ripple? This certainly seems reasonable and is probably partly attributable to less that ideal probing (ie. I don't have BNC to banana cables)

Also, basic listening test with the amp is super clean and gorgeous. No noticeable distortion at all.

Unless there's a problem with probing or testing setup, I don't think the amp is the issue.

I should mention I cranked resolution all the way up to 0.1 microsecond (WELL into very high ultrasonic range) with similar results, no noticeable fuzz and no DC offset. What say you hivemind? Thanks!

 

[trobbins]

Given some testing has not shown an amplifier problem, then one suggestion is to temporarily replace the tweeter driver with say a 20W 8 ohm resistor (assuming the driver is 8 ohm), and then  use your normal music to see if the 20W resistor heats up abnormally.
If you had a nice lab meter then also measure the rms voltage (frequency response to at least 100kHz) across the 20W resistor .  That could be a way to check if the source music has unexpected hf content, or if the crossover is faulty, but assumes you can get to the tweeter driver leads and cankeep the speaker intact enough to play music.

 

[Gold]

Amp looks good. Mea Culpa.