Mounting Power Resistors

[samej]

Hi,

I'm building a cab with peizo tweeters, and it seems to be recommended to add a power resistor to protect some amps from the low impedance at very high frequencies. I presume that because the resistor is very low resistance compared to the tweeters high impedance at normal frequencies (where there is some power), that it doesn't get very hot and can be mounted, for example directly to the plywood with cable ties. Or is this a bad idea? Should i use a heat sink or mount it away from anything that might burn? If so any recommended method? On stripboard?

I haven't really been able to find anything discussing this.

Thanks

[Al Heeley]

The crossovers on my Marshall cabs had the power resistors mounted on their legs maybe 1/2 an inch proud of the pcb to aid cooling. Trouble with this was there was greater leverage on vibrations which then caused the solder joint to crack. So mounting proud of the board is an acceptable way of tacking it but try to mount the pcb with some vibration absorbtion and make sure the connections to the board are really strong.

[flyfisher]

+1 on the mechanical considerations.

You don't mention the resistor wattage but they can be quite physically large. The bigger ones will have mounting holes specifically to provide mechanical support and it's a generally good idea to use them but even more so in a strongly vibrating environment like a bass cab.

If you're worried about anything burning then use a decent heatsink. If you're really worried about things burning then choose a heatsink with the correct thermal characteristics to keep the resistor within its working temperature range even at maximum rated dissipation.

[samej]

I'll try mounting it off some stripboard. It has no mounting holes on the resistor. I've now decided I don't think it will get very hot so it's more the mechanical aspect I guess. I'll try it for a while and see if it does get hot.

Cheers

[Alien]

I've used silicone adhesive to glue resistors directly to the back of piezo drivers in the past. I tend to use the type with aluminium cases like [url="http://uk.rs-online.com/web/search/searchBrowseAction.html?method=getProduct&R=1073611"]this[/url] as they have good solid solder tags to connect the leads to.

A

[flyfisher]

Yes, those are the type of resistors I had in mind. Good solid solder tags (as Alien said) but also solid mechanical mountings as well.

The OPs resistors might not have mounting holes but don't forget that it will do harm to use a device with a higher rating if mounting holes are important (but keep the ohms the same though!).

Also, the higher the resistor power rating, the less need for a heatsink.

With a convenient mounting hole it might be possible to bolt the resistor to the speaker somewhere.

[thinman]

I've built a Bill Fitzmaurice cab with piezos. The resistors are there to make sure the cab impedance remains above a reasonable value should the piezos oscillate.

I secured mine using some small zip ties which keeps them nice and secure. A blob of hot melt glue would probably work well. I wouldn't rely on stiff wiring to keep them secure as they're subject to a lot of vibration.

Edited June 27, 2010 by thinman

[thepurpleblob]

You need to read the spec sheet for the piezos you are using for a start. Some have quite fancy protection built in and don't need a series (or parallel resistor or even a capacitor) some do. You can't guess this bit or bad things might happen.

Somebody said the higher the power rating the less need for a heatsink. I think you'll find that the opposite is true. You can work out a rough guess at the power dissipated by the resister using Ohm's Law but it all goes as heat and if it's a lot you might need a heat sink.

This is an interesting read if you have a few minutes... [url="http://www.partsexpress.com/pdf/piezo.pdf"]http://www.partsexpress.com/pdf/piezo.pdf[/url]

[flyfisher]

[quote name='thepurpleblob' post='879273' date='Jun 27 2010, 09:54 PM']Somebody said the higher the power rating the less need for a heatsink. I think you'll find that the opposite is true. You can work out a rough guess at the power dissipated by the resister using Ohm's Law but it all goes as heat and if it's a lot you might need a heat sink.[/quote]
It's not the power [u]rating[/u] that determines how much power is disspiated. The circuit configuration will determine that and, yes, you can calculate this if you know the relevant figures.

My point was that there would be no harm using a higher-rated resistor than necessary and it might be mechanically advantageous.

Say the circuit design was such that the resistor would dissipate 1 watt, then it would do no harm to use, for example, a 25W resistor - it would still only dissipate 1 watt and it might be able to do so without requiring an additional heatsink (you'd need to do the thermal calcs to be sure), but it would be physically larger and probably have good mechanical mountings as per the link in Alien's post.

[thepurpleblob]

[quote name='flyfisher' post='879391' date='Jun 27 2010, 11:57 PM']It's not the power [u]rating[/u] that determines how much power is disspiated. The circuit configuration will determine that and, yes, you can calculate this if you know the relevant figures.

My point was that there would be no harm using a higher-rated resistor than necessary and it might be mechanically advantageous.

Say the circuit design was such that the resistor would dissipate 1 watt, then it would do no harm to use, for example, a 25W resistor - it would still only dissipate 1 watt and it might be able to do so without requiring an additional heatsink (you'd need to do the thermal calcs to be sure), but it would be physically larger and probably have good mechanical mountings as per the link in Alien's post.[/quote]

Yes... I see what you mean now. A certain amount of power will result in the same amount of heat dissipation regardless of the (maximum) power rating of the resistor. However, a dirty great big one will fins and stuff will dissipate that more easily, yes. I've never used piezo tweeters and don't know the gory details but it does sound as though the power dissipation of the series resistor is not a big deal anyway.

[Alien]

[quote name='thinman' post='878797' date='Jun 27 2010, 01:27 PM']I secured mine using some small zip ties which keeps them nice and secure. A blob of hot melt glue would probably work well. I wouldn't rely on stiff wiring to keep them secure as they're subject to a lot of vibration.[/quote]

I'd avoid hot melt glue to secure something that, pretty much by definition, is going to get hot.

A

[samej]

It's a 20 W 10 ohm resistor. It's just a cement thing with no heatsink. My rough calculations suggest that if the speaker is using 300W the resistor will produce about 0.006 W @ 1000Hz and 1.4 W @ 25 KHz. I've assumed the woofers impedance stays at 8 ohm. I have 2 tweeters in series with the resistor, and that is all parallel to the woofer. The tweeters are 1kohm @ 1 KHz and 60 ohm @ 25 KHz.

I guess to achieve that power the cab as a whole will be being driven pretty hard. Certainly as a bass cab it shouldn't be an issue, but as a PA speaker it might, particularly with loud recorded music? Even still 1.4 W isn't much....



edit for resistor wattage correction

Edited June 28, 2010 by samej

[Alien]

Should be fine. If you want to keep the resistor in clear air just hang it between the woofer and the tweeter - I've done that in the past too.

A

[thinman]

[quote name='Alien' post='879822' date='Jun 28 2010, 01:19 PM']Should be fine. If you want to keep the resistor in clear air just hang it between the woofer and the tweeter - I've done that in the past too.

A[/quote]


Hmm - I'd be a bit concerned about hanging a heavy-ish component on its wires. It wouldn't worry me with hi-fi speakers but a bass cab is subject to a lot more vibration and handling shock. It may well be OK but solder joints are quite prone to cracking under vibration and stress. Personally I'd fix them to something solid.

[flyfisher]

+1 to all that.

Solder joints are for electrical connectivity, not for securing large components in a highly vibrating environment.

Better safe than sorry.

[Alien]

20W ceramic resistors don't really weigh much at all. A short length of 4mm CSA cable (like you'd run from the input socket to the speaker) probably weighs a fair bit more, and likely moves around a fair bit too, but nobody bats an eyelid about the stresses on those solder joints do they?

A

[flyfisher]

[quote name='Alien' post='880545' date='Jun 29 2010, 12:22 AM']but nobody bats an eyelid about the stresses on those solder joints do they?[/quote]

Perhaps they should. A heavy cable is just as susceptible to movement and vibration as anything else and should ideally be mechanically fixed rather than just soldered and allowed to flap around. How many people have experienced "socket" problems at some time or another?

These are the sorts of construction details that can determine long-term reliability.

[Phil Starr]

I've used these a lot and being lazy and usually prototyping/fiddling tend to just solder direct. the resistors don't usually get that hot and I've even got away with 1W resistors.

It is unlikely that there will be any inbuilt protection or internal resistor. As far as I know Motorola and CTS have stopped making piezo's and all I've been able to find recently are the cheap clones. Having said that I've had a lot of success with this one [url="http://www.bluearan.co.uk/index.php?id=SKT902490&seeimage=YES&img_name=902490_CO1"]http://www.bluearan.co.uk/index.php?id=SKT...name=902490_CO1[/url] which is unbelievably cheap for the sound it makes. If anyone out there finds a supplier of better quality piezos I'd be pleased to find out.

I'd recommend using silicone to fix the resistor down to the back of the tweeter.

[Al Heeley]

[quote name='Alien' post='880545' date='Jun 29 2010, 12:22 AM']20W ceramic resistors don't really weigh much at all. A short length of 4mm CSA cable (like you'd run from the input socket to the speaker) probably weighs a fair bit more, and likely moves around a fair bit too, but nobody bats an eyelid about the stresses on those solder joints do they?[/quote]
Cable will be tethered at both ends. Its not so much the weight as the moment-lever action. In my marshall crossovers the power resistor was jutting out horizontally from a vertically mounted board and all the oscillating force from vibrations during play and transport was focused on one small area- the solder joint. 4 separate failures of the units were experienced before I threw in the towel and sent them back. 2 were related to the resistor joint, 2 related to the inductor coil coming loose for exactly the same reason.

Silicome sealant is a good answer as fixative and vibration absorber.