Check this circuit Please

[Juan]

This is a lot to process and my head is spinning a bit . I am not that smart anyway .
This is the best forum on the net because I have tried most of them and got almost no usable help from the others .  Thanks to everyone and I will get back to you on any results when I can .
Feel free to keep posting if any new ideas come up and I will check in when I can .
Obrigado .

[Juan]

I have redrawn the circuit with reference to the advice as best I can and changed the power transistors to higher current devices which I can buy . I could not find BD131 or BD132 Transistors .
Does this look a better circuit ? Hope I have it right now .  I will try this one first before making any radical changes.

[DrGeoff]

Do you really want to drive those transistors without anything to limit the base current? Poor 555's might cook a bit.

[Ian.M]

Yes, BD131 and BD132 are no longer in production by NXP (Philips), and although there is still limited production from lesser known 3rd party manufacturers, there's certainly no point in paying the premium for Philips N.O.S. for a new build.  The only issue is the lower current gain (BD175: h_FEmin. 15 @1A vs BD131 h_FEmin. 20 @2A) but odds are you'll get away with it for a one-off as that's a limiting value and the typical h_FE is usually a bit higher - the graph in the datasheet says h_FEmin|30 @2A, and worst case it would simply have a bit less output power.

Have you got an oscilloscope or at least some means of measuring frequency up to say 100Khz? (e.g. a decent DMM with a frequency range)  If not you are going to have a hard time verifying the circuit is operating correctly.   The easiest option if you don't is temporarily replace C1 and C4 with 100nF parts which will bring the frequency down by two orders of magnitude to where you can certainly hear it.  10nF for a one order of magnitude reduction would *probably* be enough, unless you have high frequency hearing loss.   

@DrGeoff: they are emitter followers, so the base current is limited to I_load/h_FE - simply don't overload the output!

[Juan]

Sorry no oscilloscope . The data  sheet for the BD 175 /176 says it can handle 3 amps collector current and they will be on large heat sinks . As I said before the exact frequency is irrelevant as long as it's reasonably close .
Thanks for the heads up on the 555 not overheating in the circuit that had me worried again .
If I place a candle in front of the tweeter horn it should tell me if it's working or I can detune it as you mention .
What load could I put on the output ? One 8 ohm tweeter or two ?

[Ian.M]

If they are moving coil tweeters, try 2x 8 ohm in series initially (for nom. 16 ohms) , and check how much current it draws from the battery.  If its under about 1.2A, you could remove one of them.   I don't think it could ever drive two in parallel.

[Juan]

If they are moving coil tweeters, try 2x 8 ohm in series initially (for nom. 16 ohms) , and check how much current it draws from the battery.  If its under about 1.2A, you could remove one of them.   I don't think it could ever drive two in parallel.

OK that a really important tip . I would have made a bad mistake there ,  thanks . I am not sure about the moving coil part . I was going to use a compression horn tweeter that shows a good response at 30khz something like a Motorola KSN1025A or B , a Piezo device ,  if available or a replacement ,  other than  that I don't know much about speakers in general .
What other tweeters do you think would be suitable ?   I thought a horn type device would project the sound better than a dome tweeter .  I could be wrong . I am sure you will steer me straight . I checked my Digitech multi meter and it does indeed have a frequency test . 5 khz to 10 mhz 1.2% accuracy at 50khz . How I use that properly I don't have a clue . The instructions are very poor. I guess I put the negative probe on the negative side of the circuit and the positive probe on the output pin of the 555 . Does that sound right ?

[Ian.M]

For a Piezo horn, I'd just connect one.   N.B. a Piezo element that isn't tuned to resonance is typically a very capacitive load - so it is likely to pull large current spikes from the supply every time the H-bridge switches, and may overheat the transistors.

Check the supply current - it nearly all ends up going through the transistors, with one digonal pair on at a time, so if that's under 1A and the  transistors aren't heating up (especially the upper ones) you are probably OK.

[Juan]

Ok ,  so if a speaker is moving coil type I can use two 8 ohm  in series connection for approx 16 ohm total .
If speaker is a Piezo type device then only connect one speaker .
Monitor transistors for overheating , check the supply current . They will all be on large isolated heat sinks anyway with conducting paste applied . I learnt that from a previous  project .
In your esteemed opinion what type of speaker  should I use as too many choices tend to confuse me .
However I feel I am getting somewhere now , thanks ever so much .