Discussion in 'Solid State' started by K7sparky, Feb 26, 2013.
What's a kill-a-watt?
A nifty inline device to monitor line conditions.
What makes you say that?
Are these insulated tab triacs using a BeO/ceramic tab or a metal tab? I don't trust metal tabbed insulated triacs chassis mounted on non-earthed equipment- do you?
Haven't looked at this thread in awhile sorry.:tears:
The isolated tab is regular metal. Nope doesn't bother me a bit. We all have our worry factors and they change over the years. Alfred E Newman moment - "What me worry" for those that forgot.
Never seen or heard of a isolated tab shooting through, but sure as God made little green apples it's happened. All the failures I have seen were benign or nothing left but the tab, leads and beach sand.
One of my worry factors causes me to opening up every amp I have with a non polar plug, trace the white lead back to the plug and use white out to stripe that side of the plug so I can put that blade into the wide side of modern outlets.
I am used to seeing 1/2 to several meg resistor from the white AC lead to chassis on vintage with non polar plugs. Plugged in one way chassis is tied to ground the other is 120 VAC. Reason to worry? Me yes. Never even gotten a tickle from a flipped plug.
The TRIACS are cheaper with a non-isolated tab and you add the mica and the #4 screw isolator. Personally I prefer the ease of isolated, but that's just lazy old me. Like everting in life, yous pays yo monies and takes yo chances.
Thanks for the great traces you posted!:grnbounce Like you my power is stiff. Around 50 feet from the pad mount to my shop, 200A service and I didn't scrimp on the copper inside runs 20 to 50 feet around the electronic area.
Thanks for the link to the Kill A Watt. That's a new one for me.:thmbsp:
You are right - lots of amps don't need help for the power switches. My Pioneer SX-1050 as another example of doing it right. There were also plenty of amps that had issues. Don't remember if my RS660 still has the relay I installed. Any of the amps I did a relay mod to get the TRIAC mod if I need to go back under the hood.
The RS660 / LR 9090 series are another example of a bad idea using a power strip for switching AC on and off. The 660 uses a wafer on the power switch and the 9090 uses a contact in the AC switch of the Power / Speaker switch to dump the protective relay on shut down. The loud thump of the woofs on power failures out here on the farm are something that can't be good. No need to get one every night when I secure the amps.:twak:
Aren't these switches the same as the mentioned Lafayette/Setton switches?
I don't have speakers hooked up to my LR-9090, I only use headphones. I've been using the power-strip method of turning it on and off for about 6 months without issue. Aside from the pop in the headphones upon powering off are there any other adverse effects I should expect?
Post 19 so-ein-pech confirms the Alps switch is the same on small SX Pioneers as I pictured from a Laffer / Setton series.
I suggested removing the snubber as:
It is no longer needed as there is no arc to quench.
It is another failure point gone
It leaks a minor amount of AC - just picking nits.
If there is a pop "power down" on the Lafayette / Setton the protection circuit is not functioning correctly. The electronics, welded relay contacts or switch (Alps / Wafer) not making contact to common would be the top guesses.
Pop on "power up" the protection circuit is also not working correctly. Possibly the power on delay / hold off cap. On the Laffers / Settons I increased the capacitance to increase the time on delay. I am not in a hurry for power on and wanted the power supplies and PA stable before hooking up the speakers. The PAs oscillate a bit on power up. Some longer / worse than others. I know I watched the PA voltage before the protection relay contacts on a scope and may have dug out my old Brush 2 channel pen dragger and run a few charts.
Only adverse effect that comes to mind is the pop you mention. The +/- rails decay at different rates. The + has a slightly higher load so decays faster. Have not checked for oscillations while shutting down. Not sure how much you have invested in your cans, or if the DC voltage across them on decay will hurt them or not. The headphone jacks have 270 Ohm resistors in series which may be enough to protect your cans.
With the AK knowledge base I expect someone here to know that answer!:scratch2:
The Lafayette uses the N/C contacts on the power switch to trip the protection circuit relay at the same time the N/O contacts open the AC power thus opening the speaker (and phone) leads.
I believe you also mentioned somewhere you do not use the switched AC outlets on the back so no issue there.
There are a few times I like the switched AC socket. Always a possible path to trouble. At first I used extra contacts on an ice cube relay. The TRIAC is way cheaper than a relay if you buy new and fine for the switched AC socket also. Either way the switch is protected from dumb mistakes / Murphy.
I use Monoprice DJ8323 headphones [$30] with the Lafayette LR9090 so I'm not concerned about harming them. The turntable that I have hooked up to the Lafayette is pretty much exclusively for playing mono records so the Monoprice cans are just fine.
Here's an odd question:
I have a dimmer switch that automatically ramps up the power to full over the span of about 1.75 seconds.
Would using that on an untested vintage receiver (or even one thatis in regular use) have any benefit?
The post about triacs in dimmers made me think of this.
No. Do. Not. Use.
Such devices are useful for resistive loads only.
I agree. I learned that 20 years ago while hooking it up to the ceiling fan. The fan motor turns fine but was pulsing until I turn on a light along with the fan and problem solved but not a good way to leave it. So I removed the dimmer and back to switch.
Please allow me to revive this thread.
I've done this mod on a receiver and two amps, and it worked absolutely perfectly. Today, I did it on a Sansui AU-777A amp. I'm getting some non-volume-dependent hash through the speakers with the triac installed. It's still there if I turn the speakers off.
Moving the triac 18 inches away from the unit with some jumpers makes no difference. Switching the leads on MT1 and MT2 makes no difference. Taking the triac out of the circuit eliminates the hash.
I haven't noticed any hash on the two amps and the receiver I've installed triacs on. The only other change I've made to the AU-777A is a new polarized, two prong AC cord. I double checked that I have hot and neutral wired right.
Gotta give this some thought. Maybe I'll try the r-c snubber across MT1 and MT2, which I have never bothered with.
Did you use the recommended? It doesn't require a snubber.
A "snubber" may help to some extent, but the cutoff of the triacs at the point of polarity transition can set up an inductive kick in the transformer (like an ignition coil) and that may be difficult for an R/C to eat.
This is one reason I have usually advocated a relay for this duty instead of a triac. I got my fill of triacs and SSRs in the '80s eating the primary of power transformers due to the unintended consequences of how they switch. New transformers in some equipment have a better chance of being ESD shielded and wound with better insulated wire so that these voltage spikes do not punch through the varnish, but older transformers are unlikely to be designed for this.
These things switch at zero crossing, so I'd be surprised to see the device be the the source of trouble if it was working properly. Maybe a bad one...?
Could be a bad one, but remember ELI the ICE man. Transformers are inductive, and triacs are somewhat resistive and this has the greatest effect near zero volts - in both directions of the AC wave. Relay contacts are completely unaffected by the inductive lag of current behind the EMF. Not carrying cleanly through the full waveform can have unintended and detrimental effects, even if only dropping a few volts at the crossing. Remember that a millions of 6V vehicle ignition coils only used a few volts of primary exciter voltage from the magneto. That's the equivalent of +3V and -3V lost at the cross, or what would amount to a Vf of less than 3V. Most solid state devices rated for 200V are at least 1.0V Vf, which could translate to a small inductive surge at the cutoff point. I used to watch this via a storage scope and cursor over the duration of the gap, and compare to the corresponding "hiccup" in primary voltage when the triac was on the secondary side controlling an inductive load, such as a vibratory feeder or power control transformer. The spikes in primary voltage were impressive, and enough to explain the frequent loss of large KVa (37 and up) transformers or transformer banks when the primary windings suffered insulation breakdown.
I suspect that most of these vintage units with linear power supplies used the "old" conventional designs which are of similar construction to those 1980s, never-meant-to-be-switched line transformers, and lacked the shielding and "spike-resistant" insulation of modern transformers.
Granted, there ARE triac devices (not simple triacs) which use photo-coupling and driver circuits which "pre-fire" the output triac to minimize the phenomenon, but even those can have a wee bit of lag.
Ideally, an HD thermistor for inrush current limiting and a small relay with adequate inductive load charachteristics should solve the problem of marginal power switch contacts permanently, or at least up to the mechanical and electrical MTBF of the relay.
Thanks all for the thoughts. I'll try the snubber, but I don't see why that would make a difference with hash. It could be a bad triac, and I have some spares, so I'll pop another one in there if nothing else works.
Along the lines of what I did differently on this unit: On previous units, I put the triac very close to the power switch. I couldn't find a decent and easy spot near the AU-777A switch, so I mounted it on the underside of the chassis, at the back of the unit, on one of the power transformer bolts. Right where the AC line comes in. I snipped the hot wire at the switched outlet. I thought maybe proximity to the transformer might have something to do with it. But, I have the triac on jumpers, 18" away from the unit, and the hash is still there.
So if the snubber doesn't work, next thing will be mounting the triac temporarily at the front of the unit, near the switch, and seeing what happens.
As far as using a snubber with this particular triac, Littelfuse says none is necessary, since it's designed for an inductive load. EW recommends it anyway to avoid unintentional ons and offs due to power line dirt. This hasn't been happening to me.
This amp has cleaned up and fixed up almost too easily so far. If the triac is the biggest problem I hit, I'll be very happy.
The distance from the AC line, the switch, and the transformer are all but inconsequential. Electricity (arguably) travels at about 186,000 miles per second (6.7x10^8 MPH). I'll let you calculate how long it takes to traverse that extra 18" of copper and see if the resultant frequency shift/time represents anything near the microwave range, let alone the RF range (Forget about the audio range, where you can apparently hear the noise).
Well that sent me right to Google. Lead and lag. OK. I'll give that some thought. The mnemonic could be better.
I've been looking at these things. There are some threads over at DIY. But the attraction of the triac is that it's cheap and simple.
Separate names with a comma.