16 or 8 ohm speaker ?

[Mark Hennessy]

A quick search suggests it might be similar to the 2038. If so, that's running from 24V off-load, and the 18VAC supply is coming from an overwind on the motor. That'll sag considerably under load. The 3 watt rating is "speech and music", which means "peak power", which means the average sine wave power will be more like 1.5 watts.

The output capacitor is 400uF, which gives a -3dB point of 50Hz into an 8 ohm resistive load. That's plenty low-enough for something like this. Assuming that capacitor is still healthy, of course

[Dick.1]

Hi Mark

"Fleabay" is just my "affectionate" nick-name for ebay & has no other meaning

Link here

http://www.ebay.co.uk/itm/pair-of-6x4-sp...SwGiRTrXCP

Doesn't give much away

I've already used my portable mp3 player to check things out via the tape/radio sockets

Out of curiousity I hooked up a Celestion "Tube 10" guitar speaker (8 ohm) I had to hand & the results even at 1/2 volume where ear splittingly loud mostly trebly but the speaker was just lying beside the player on the table...........they are an efficient speaker & work very well in a semi sealed cab.
I do have a 12" "neo magnet" cele (16 ohm) that would work very well I imagine but not that much room left inside the cab for the turntable ...............

Dick.

[Dick.1]

A quick search suggests it might be similar to the 2038. If so, that's running from 24V off-load, and the 18VAC supply is coming from an overwind on the motor. That'll sag considerably under load. The 3 watt rating is "speech and music", which means "peak power", which means the average sine wave power will be more like 1.5 watts.

The output capacitor is 400uF, which gives a -3dB point of 50Hz into an 8 ohm resistive load. That's plenty low-enough for something like this. Assuming that capacitor is still healthy, of course

Hi Mark

The rail is measured by me at 23.5v (off load) which would appear in line with your thinking.

The docs I downloaded agree that 2028/2038 are  more or less the same as is the Ultra 6024.

Dick.

[Mark Hennessy]

Link here

http://www.ebay.co.uk/itm/pair-of-6x4-sp...SwGiRTrXCP

Thanks for that - they look quite respectable, and not a bad price for a pair, including delivery. Tempted to pick up a set...

I might be wrong, but that looks like a Mitsubishi logo on the back? As they are shielded, perhaps there were made for CRT TV sets or similar (the usual market for elliptical speakers in recent times)? As you suggested earlier, PC speakers also seems likely - perhaps more so, thinking about it...

The cloth surround suggests a reasonable amount of travel, so an appropriately tuned enclosure is most likely the missing link here.

So, definitely some potential for a future project. Perhaps a bench amp? My one gets a lot of use when fault-finding and designing stuff...

[Dick.1]

I had thought Mitsubishi as well ...paper cone & decent doped surround, 55mm magnet housing & weighing in at 360gms

It should work better in a "box" of course, there's some room 5x4x6 H.D.W. (inches)....something to look into with the potential to slope the back a bit

Very interesting "bench amp" by the way enjoyed reading the back story as well.

Dick.

[Murphyv310]

Hi.
I think the OP has answered his own question in the last few posts.
Fitting an 8 ohm speaker of dubious efficiency is the crux of the matter. If you need to crank up the volume to get a reasonable sound you could well be driving the amp too hard.
If the original 16 ohm speaker was efficient and required less drive for the same volume than the 8 ohm replacement then I'd say that it would be an idea to revert to the specified speaker.
Adding a matching transformer could be a safeguard or an idea if a good quality high efficiency 8 ohm speaker is available.

[Terry]

To calculate the peak output power, take the supply voltage and divide by 3 (strictly speaking it should be 2.828 but the supply will almost certainly sag at maximum output) and square the answer. Then divide by the speaker impedance:

P(W) = ((Vs/3)^2)/ZΩ 

That's not the peak power; rather it's the average sine wave power. What some folk erroneously call "RMS Power"

I was well aware of that Mark! However I felt it would be a better figure to compare with published data on power ratings.

I  did  include the peak current figure as well, though!

I ignored emitter resistors but I'd also rounded up the RMS conversion figure so the answer won't be 100% accurate anyway but as it will, in fact, be a higher figure than actually achievable it allows a degree of safety.

Any speaker available can be used to check the amplifier although the power should be kept down if the impedance is lower than specified - particularly if it is a 4Ω  one!

Higher impedance speakers will work without any restrictions but the output power will reduce as the impedance increases and would need to be  taken into account.

[Hartley118]

I'm guessing that the expired cartridge that you replaced was a crystal type. It may be that the replacement has a lower capacitance than the original and so may need a higher load impedance. A crystal or ceramic cartridge effectively incorporates its own rather low value coupling capacitor, typically of the order of 2nF and usually needs a load of at least 1Mohm to preserve its low frequency response

Achieving a high input impedance in transistor amplifiers of that vintage was always a challenge. If you've got plenty of gain, the first thing I'd try in order to get more bass response is to connect a 470k resistor (or more) in series with the input to present a higher load impedance to the cartridge.

Martin

[Dick.1]

Hi.
I think the OP has answered his own question in the last few posts.
Fitting an 8 ohm speaker of dubious efficiency is the crux of the matter. If you need to crank up the volume to get a reasonable sound you could well be driving the amp too hard.
If the original 16 ohm speaker was efficient and required less drive for the same volume than the 8 ohm replacement then I'd say that it would be an idea to revert to the specified speaker.
Adding a matching transformer could be a safeguard or an idea if a good quality high efficiency 8 ohm speaker is available.

Hi Trevor

The HMV came to me with this 8ohm speaker fitted by a previous owner, it is my declared intention (as posted) to look for the correct spec of speaker

The fitted 8ohm speaker appeared to be of mediocre efficiency & judging by the size of the magnet, will need more power to drive it properly than is available so "padding" the speaker to the correct impedance will rob to much volume.

This is the first transistor amp like this I've looked at in 40 odd years & my memory very obviously 'aint what it used to be, so some posts in this thread may appear somewhat "disjointed" to those who have been able to continue uninterrupted in the trade over many years


Dick.

[Mark Hennessy]

To calculate the peak output power, take the supply voltage and divide by 3 (strictly speaking it should be 2.828 but the supply will almost certainly sag at maximum output) and square the answer. Then divide by the speaker impedance:

P(W) = ((Vs/3)^2)/ZΩ 

That's not the peak power; rather it's the average sine wave power. What some folk erroneously call "RMS Power"

I was well aware of that Mark!

So why not say "take the supply voltage and divide by 2" rather than what you actually said (which amounts to "2-root-2 plus a bit")?

Clearly, you had "Watts RMS" in mind when you wrote that. Not "peak output power".

However I felt it would be a better figure to compare with published data on power ratings.

At that stage, we hadn't looked up the published ratings. That didn't happen until I did that (post #11).

I did  include the peak current figure as well, though!

Yes, which is what matters here for the sake of the output transistors. Which I didn't question, BTW. Indeed, I provided the analysis that you could've done...

I ignored emitter resistors but I'd also rounded up the RMS conversion figure so the answer won't be 100% accurate anyway but as it will, in fact, be a higher figure than actually achievable it allows a degree of safety.

Yes, but even with the "engineering rounding" - which I have no problem with - you were talking about "Watts RMS" rather than "Peak Power".

Either way, the question has been answered.