I need an amp for a first time speaker project. So far the P3A has been chosen. It seems to be suitable for the speakers and possibly to build for a newbie.
The amp must be compact and able to output around 30-35w continously into a 8 ohm load. This translates into a 25v rail voltage.
Given these criteria I decided to build to seperate mono amps. I should be able to stuff a proper transformer (toroidal 100-120VA) and the amp section into this nice case:
I guess the heat dissipation from the output transistors is rather moderate at this voltage. Will I be able to use the cases as heatsinks by simply mounting the transistors directly on the sides?
I've been looking at different toroidal transformers, but haven't yet decided. It must rated 100-120VA and have a diameter of no more than 90mm. Any suggestions? The transformer must be available in Europe.
Does anyone in here use the above case and/or know the internal dimensions?
Thanks
Regards
Soeren B.
The amp must be compact and able to output around 30-35w continously into a 8 ohm load. This translates into a 25v rail voltage.
Given these criteria I decided to build to seperate mono amps. I should be able to stuff a proper transformer (toroidal 100-120VA) and the amp section into this nice case:
An externally hosted image should be here but it was not working when we last tested it.
I guess the heat dissipation from the output transistors is rather moderate at this voltage. Will I be able to use the cases as heatsinks by simply mounting the transistors directly on the sides?
I've been looking at different toroidal transformers, but haven't yet decided. It must rated 100-120VA and have a diameter of no more than 90mm. Any suggestions? The transformer must be available in Europe.
Does anyone in here use the above case and/or know the internal dimensions?
Thanks
Regards
Soeren B.
Hi, I had a look about on RS, and although I couldn't find a toroidal model with would fit in the 90mm space with suficient power rating, I did find a 150VA open frame model which measured only 100 X 82 X 82 mm which should fit (provided the case is tall enough). anyway, hope that is of some help, and if you could still et me know where thcase is from I would be very greatful.
Just guessing, but based on my experience building a modest power P3A, --- anything big enough to holdthe PCB and transformer, etc will probably be ok for a heatsink -- provided it is fairly heavy aluminum. The picture you show is encouraging since it has fins and looks like a casting. Somewhere on the Rod Elliot website, he has some formulas to calculate thermal resistance using the surface area. You might look at those - you would have to do a lot a measuring and some arithmetic but nother exotic.
Hi!
I used this case (one size smaller) for my GainClone (which now lies unused upon my carpet in front of my P3A
), they can really provide some cooling. But one drwaback is that there are no holes in the case to make air ventilation possible, so you will suffer from heat buildup.
I don't know though, if you can find a suitable transformer. The toroids from RS Components are probably not usable, since they waste space for being potted...
Maybe use a "standard" transformer, not a toroid? Maybe out of an old disfunct amp?
Try it... I can not rely on such a case as a heatsink for my P3A... 42 V rails, 6 ohms load, MJ15003 / 4 transistors...
Bye,
Arndt
I used this case (one size smaller) for my GainClone (which now lies unused upon my carpet in front of my P3A
), they can really provide some cooling. But one drwaback is that there are no holes in the case to make air ventilation possible, so you will suffer from heat buildup.I don't know though, if you can find a suitable transformer. The toroids from RS Components are probably not usable, since they waste space for being potted...
Maybe use a "standard" transformer, not a toroid? Maybe out of an old disfunct amp?
Try it... I can not rely on such a case as a heatsink for my P3A... 42 V rails, 6 ohms load, MJ15003 / 4 transistors...
Bye,
Arndt
The case is sold at Conrad.com (data sheet here). No internal dimensions are given. Generally this is one of those data sheets that do not serve a purpose.
Farnell sells a toroidal transformer from Avel Lindberg, which is rated 100VA and has a diameter of 91mm. I cannot find this trafo on AL's website. Instead they have a product range called Y23, which is even more compact (100VA: diameter 85mm; 120VA: diameter 89mm). I should be able to fit into those cases. But who sells this Y23 series?
Cradle22, how serious is this heat buildup? Will it be a problem? Should I be able to use a 22vac type (corresponding to approx. 30vdc per rail) giving 50w? Remember, there's only one channel in each case. How powerfull is your GC?
Farnell sells a toroidal transformer from Avel Lindberg, which is rated 100VA and has a diameter of 91mm. I cannot find this trafo on AL's website. Instead they have a product range called Y23, which is even more compact (100VA: diameter 85mm; 120VA: diameter 89mm). I should be able to fit into those cases. But who sells this Y23 series?
Cradle22, how serious is this heat buildup? Will it be a problem? Should I be able to use a 22vac type (corresponding to approx. 30vdc per rail) giving 50w? Remember, there's only one channel in each case. How powerfull is your GC?
Hi, I know this is a bit of an indirect way of getting the internal dimensions fo the case, but the case is designed to hold eurocards, therefor, it must have an internal width of 100mm. So from this, you can tell that each wall has a thinkness of 5mm, and the internal dimensions of the model you are looking at must be:
162 X 100 X 70 mm (the ends are only 1.5mm thick).
Hope this helps you out a bit, as it should mean you will be able to use larger transformers.
162 X 100 X 70 mm (the ends are only 1.5mm thick).
Hope this helps you out a bit, as it should mean you will be able to use larger transformers.
Hi!
Internation relationships have their benefits...
I have the conrad catalogue (and my GC case) lying in front of me, and the one you depicted on your picture must be the biggest one available, which (external measurements) is 165 * 110 * 105 mm big (length * width * heigth), this means 163 * 100 * 95 mm internal measurements (judging from my own, smaller case, but the thickness of the material doesn't vary).
I would say that using one case for one side would be OK (I thought that you meant implementing both channels into one case including transformer, whcih would have meant pushing your luck...)
Using a case, and a single trabsformer for each channel should be possible. I mounted both OPA548 in the smallest version of those cases, with external transformer, and I run them at 35 V rails (25 V transformer). I did not mount them to the sides, but to the top of the case, and did not try anything else. With my setup, the case only gets hot when pushed really hard into my 6 ohms speakers. I would say 30 V rails, with one case per channel, at normal listening levels (that means not constant PA action...), at loads > 4 ohms you are on the safe side. Just use standard mica washers with heatsink grease (I used Arctic Siver III from computer supplies, don't know if that makes such a difference).
You can also try to press the transistors (which ones to you inted to use) to the case (with a c-clamp) prior to drilling, and therefor try finding the best place to mount the transistors.
And please tell me the outcome of your experiments! I really like those cases (first saw them on the "GeenKlone"-project), and the smallest one is rather empty with my stereo GC inside (about 35 % space used)...
Bye,
Arndt
Internation relationships have their benefits...
I have the conrad catalogue (and my GC case) lying in front of me, and the one you depicted on your picture must be the biggest one available, which (external measurements) is 165 * 110 * 105 mm big (length * width * heigth), this means 163 * 100 * 95 mm internal measurements (judging from my own, smaller case, but the thickness of the material doesn't vary).
I would say that using one case for one side would be OK (I thought that you meant implementing both channels into one case including transformer, whcih would have meant pushing your luck...)
Using a case, and a single trabsformer for each channel should be possible. I mounted both OPA548 in the smallest version of those cases, with external transformer, and I run them at 35 V rails (25 V transformer). I did not mount them to the sides, but to the top of the case, and did not try anything else. With my setup, the case only gets hot when pushed really hard into my 6 ohms speakers. I would say 30 V rails, with one case per channel, at normal listening levels (that means not constant PA action...), at loads > 4 ohms you are on the safe side. Just use standard mica washers with heatsink grease (I used Arctic Siver III from computer supplies, don't know if that makes such a difference).
You can also try to press the transistors (which ones to you inted to use) to the case (with a c-clamp) prior to drilling, and therefor try finding the best place to mount the transistors.
And please tell me the outcome of your experiments! I really like those cases (first saw them on the "GeenKlone"-project), and the smallest one is rather empty with my stereo GC inside (about 35 % space used)...
Bye,
Arndt
Hi!
Don't know. At least not 50 posts a day in the "Not just another GC" thread... any more...
I never intended my GC to be more than an experiment. And when my main amp failed (OK, I helped him at that, letting my probes slip...
), I decided that it was time to go after the real thing.
Class A was out of the question at that time (money, loudspeaker considerations), and I wanted to have something with power (regardless of what some watt freaks say, I still consider 90 W into 8 ohms with just four output transistors somewhat powerful), so I decided to build the P3A.
Apart from having to implement a soft start at sometime (every 10th to 20th switch on, my mains fuse plops out ), and ordering a computer-CNCed frontplate my P3A is finished...
Maybe I will consider exchanging the standard Welwyn resistors with RC55 precision type, but that will have to wait...
Does someone want to buy a GC? Just 1000 US $$$
Bye,
Arndt
JoeBob said:
Don't know. At least not 50 posts a day in the "Not just another GC" thread... any more...
I never intended my GC to be more than an experiment. And when my main amp failed (OK, I helped him at that, letting my probes slip...
), I decided that it was time to go after the real thing.Class A was out of the question at that time (money, loudspeaker considerations), and I wanted to have something with power (regardless of what some watt freaks say, I still consider 90 W into 8 ohms with just four output transistors somewhat powerful), so I decided to build the P3A.
Apart from having to implement a soft start at sometime (every 10th to 20th switch on, my mains fuse plops out
), and ordering a computer-CNCed frontplate my P3A is finished...Maybe I will consider exchanging the standard Welwyn resistors with RC55 precision type, but that will have to wait...
Does someone want to buy a GC? Just 1000 US $$$
Bye,
Arndt
Thanks for the answers all of you. And yes, I will post some pictures whenever the amps are done, but it might be 1½-2 months due to military service.
I've been looking at different transformers, and has (almost) decided for Avel Lindbergs Y23 range (date sheet). Will it do?
I've been looking at different transformers, and has (almost) decided for Avel Lindbergs Y23 range (date sheet). Will it do?
P3A Question
Hi!
Just want to ask if I can use my transformer that gives rail voltage of +/-63V to a P3A? Do I have to change anything? Output power is not an issue, I just don't want to junk my transformer and find a use for it instead.
I read that I can parallel output devices so can I use 4 pairs of 2SC5200/1943 for the output?
Regards,
Jojo
Hi!
Just want to ask if I can use my transformer that gives rail voltage of +/-63V to a P3A? Do I have to change anything? Output power is not an issue, I just don't want to junk my transformer and find a use for it instead.
I read that I can parallel output devices so can I use 4 pairs of 2SC5200/1943 for the output?
Regards,
Jojo
Info on P3A please
Hi Arndt
I just came over from the chip amp section and noticed your post here. Can you ( or someone ) tell me about the P3A, it sounds interesting. presumably class A ?
I would be interested to here your thoughts on the on how it out performs your gain card.
cheers
mike
Hi Arndt
Cradle22 said:Hi!
I used this case (one size smaller) for my GainClone (which now lies unused upon my carpet in front of my P3A
I just came over from the chip amp section and noticed your post here. Can you ( or someone ) tell me about the P3A, it sounds interesting. presumably class A ?
I would be interested to here your thoughts on the on how it out performs your gain card.
cheers
mike
Re: Info on P3A please
Hi!
No, the P3A is not class A, but class AB. I think it would be hard getting a class A amp producing 80 - 100 W into 8 ohms with 42 V rails voltage and just two power transistors...
But the P3A is one of the simplest, yet best sound class AB amps. Take a look at ESP, there you can find it (also boards you can order, but it can also be implemented on a prototype board).
Now, don't hit on me, I say it uses a somewhat similar approach like the AKSA amp (been through this discussion before).
I cannot really compare it to the GC, since I am not such an abled music listener like some others on this forum , I think both sound fine, and better than anything I've had before, but the P3A sweeps the floor with the GC when it comes to output power.
I think a P3A can be build from anything between 60 - 400 €, depending on what components & casing you chose... and I would say that you will have to look hard at finding a better commercial amp up to a price range of 1500 - 2000 €...
Bye,
Arndt
Hi!
mikelm said:
No, the P3A is not class A, but class AB. I think it would be hard getting a class A amp producing 80 - 100 W into 8 ohms with 42 V rails voltage and just two power transistors...
But the P3A is one of the simplest, yet best sound class AB amps. Take a look at ESP, there you can find it (also boards you can order, but it can also be implemented on a prototype board).
Now, don't hit on me, I say it uses a somewhat similar approach like the AKSA amp (been through this discussion before).
I cannot really compare it to the GC, since I am not such an abled music listener like some others on this forum
, I think both sound fine, and better than anything I've had before, but the P3A sweeps the floor with the GC when it comes to output power.I think a P3A can be build from anything between 60 - 400 €, depending on what components & casing you chose... and I would say that you will have to look hard at finding a better commercial amp up to a price range of 1500 - 2000 €...
Bye,
Arndt
Parallel outputs
What is the best way to rig up parallel outputs on a P3a? I got my MJ15003/4 pairs from On-Semi's sample program (5 pairs) and have been thinking about a parallel output stage for my amp.
Correct me if I'm wrong (I frequently am) - parallel output pairs allow lower R loads such as 4R speakers with the same rail voltage & allow lower dissipation per transistor. Would a parallel output have any other effect on the amplifier?
Cheers
Paul
What is the best way to rig up parallel outputs on a P3a? I got my MJ15003/4 pairs from On-Semi's sample program (5 pairs) and have been thinking about a parallel output stage for my amp.
Correct me if I'm wrong (I frequently am) - parallel output pairs allow lower R loads such as 4R speakers with the same rail voltage & allow lower dissipation per transistor. Would a parallel output have any other effect on the amplifier?
Cheers
Paul
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.