That's not the putting compound I think of when I hear the term 'silicone', but I should perhaps stop frequenting the Hollywood region I guess?
Although I believe it's Altronics that are the poor sods who ended up with the task of getting the Currawong kit out - no doubt regretting it.
For Aussie readers who want an Aussie designed 10W per channel valve (tube) amp that would leave this design for dead l have no hesitation in self promotion by recommending the Baby Huey instead. Parts cost would be very similar or less.
Baby Huey PP EL84 amplifier - diyAudio
Hi there,
If I wanted to build/use two of these in monoblock power amp form with my existing Grounded Grid pre-amp, presumably I can run the input signal straight through the 1K resistor to pin 2 of the ECC803S tubes?
Does an 8ohm speaker load affect the design, or just change the output power?
Thanks
TT
Baby Huey PP EL84 amplifier - diyAudio
Hi there,
If I wanted to build/use two of these in monoblock power amp form with my existing Grounded Grid pre-amp, presumably I can run the input signal straight through the 1K resistor to pin 2 of the ECC803S tubes?
Does an 8ohm speaker load affect the design, or just change the output power?
Thanks
TT
The Baby Huey design has stupid distortion-worsening constant current cathode biasing.
And people build it with an output transformer unsuitable for the high HT voltage and wondering why it goes into Class B.
Far better to build a Mullard 5-10 or a tube Playmaster. They were designed to be low cost and simple. But they were designed by people who knew what they were doing. Either will outperform a Baby Huey.
And people build it with an output transformer unsuitable for the high HT voltage and wondering why it goes into Class B.
Far better to build a Mullard 5-10 or a tube Playmaster. They were designed to be low cost and simple. But they were designed by people who knew what they were doing. Either will outperform a Baby Huey.
This has been trashed/thrashed out in several threads.
To give a full explanation of all the stupid features of the Baby Huey would take many pages and I'm getting tired of doing it.
Here's a short version:-
A Class A amplifier such as the BH increases power stage current draw in proportion to signal level, becasue of the curvature of tube characteristics. The CCS forces the current to remain at the idle level, by increasing the cathode bias voltage. Thus the tube bias voltage rises in proportion to signal level.
Tubes have a grid-cathode voltage vs anode current transfer characteristic that is increasingly curved the closer you take them to cutoff. Therefore a CC biasing block increases the non-linear distortion the amplifier produces.
Further, the increase in cathode voltage that occurs in heavy drive charges the cathode bypas capacitors. When the heavy drive ceases, the capacitor is left charged and can only discharge at a rate determined by the amount by which the tube is less than the CCS current. During this time the amplifier remains over biased and is partly paralysed - soft passages following loud passages are briefly badly distorted.
So, an unusual added complexity - const current bias blocks - makes performance worse! What a stupid thing to do.
The BH incorporates a system of feeding the driver triodes from the output stage anodes instead of from HT. This is sometimes done in cheap and nasty SE designs and in them is a form of negative feedback.
In the BH, the output transformer forces the two output anodes to be equal and opposite in signal voltage. So in theory the midpoint of the 16K cross-coupling resistor sees zero signal voltage and the end result is drastically reduced feedback. But while the output stage anodes will be equal and opposite, the two feedback resistors (47K) will due to tolerance be slightly unequal. With 5% tolerance the max error in any one amp is 10%.
The effect of this will be to force the output tubes to be unbalanced by up to 10% above whatever the other resistors and the tubes themselves will have and will increase distortion. Luck enters into it: In some amplifiers the tube unbalance will tend to cancel the feedback unbalance - in others it will add. This is NOT the way to competently design a 4-tube push pull amplifier.
Again, an added complexity - albeit small - that worsens performance. Crazy.
There is a balance pot in the driver cathode circuit. That's the wrong place. A balance pot would not really be needed if the stupid cross-coulpling in teh drive anode circuit wasn't there. But if you want the best possible performance, you trim the output stage idle currents for static balance FIRST. Then after doing that, for the theorectically best possible performance you THEN trim the output stage drive for dynamic balance. Usually, there is no audible difference and precious little measurable difference in attaining dynmaic balance.
The BH runs on a high HT voltage. The power transformer primary impedance for the voltage specified should be 10 kohm anode to anode. This is unusually high and is difficult to find in push-pull form with UL tappings. So people use an available lower impedance primary. This means either output tube cannot pull their anode voltage fully down without cutting off the other tube (a situation which is agravated by the stupid CCS bias circuit). With one tube cutting off on each peak, the result is efffectively on high signal levels the power stage gain halves - the result is large amounts of distortion. Or, people can track down a 10Kohm transformer. It will be designed for a much lower power output - added distortion again.
Not only is this gain halving agravated by teh CS biasing, it agravates the sliding bias caused by the CSS baising (because the output tube not cutoff is forced to do twice the work in Class B) - still more distortion.
What a stupid design!
Last edited:
Design Engineers really don't like having to fix up a circuit after the PCB has been finalised, the board layout often restricts what can be easily done to address problems. The Currawong Amp is quite obviously designed down to a price rather than up to a standard BUT needs work on the Phase Splitter to correct its balance.
The practical fix for this is to remove the 6K8 in the cathode circuit and give the differential amp splitter a decent "tail" by inserting a current source returned to a -ve voltage rail. The negative rail is required to give enough "compliance" voltage for the current source to work.
Set both anode resistors to the same value (say 120K) and up the current in both sides of the diff amp to say 0.6mA. You will need to generate the negative supply from the 12V AC feed. Because of the low impedance looking into the diffamp cathodes and the high impedance of the current source (which will divide any negative rail noise by a lot), and the low current draw, the negative rail supply can be quite simple.
A half wave recified supply would probably be OK. Ring of Two transistor current source set for 2 x 0.6 = 1.2mA would do but I would probably go a little more sophisticated and use a cascoded 2 transistor CCS with LED reference.
Cheers,
Ian
The practical fix for this is to remove the 6K8 in the cathode circuit and give the differential amp splitter a decent "tail" by inserting a current source returned to a -ve voltage rail. The negative rail is required to give enough "compliance" voltage for the current source to work.
Set both anode resistors to the same value (say 120K) and up the current in both sides of the diff amp to say 0.6mA. You will need to generate the negative supply from the 12V AC feed. Because of the low impedance looking into the diffamp cathodes and the high impedance of the current source (which will divide any negative rail noise by a lot), and the low current draw, the negative rail supply can be quite simple.
A half wave recified supply would probably be OK. Ring of Two transistor current source set for 2 x 0.6 = 1.2mA would do but I would probably go a little more sophisticated and use a cascoded 2 transistor CCS with LED reference.
Cheers,
Ian
All well and good, GingerTube, but why persist with the Currawong in the first place?
A UL tube Playmaster will do a lot better (about the same power, lower distortion) and be far simpler to build. And uses cheap TV triode-pentodes too. There must be a few folk who persist with Playmaster designs, because the transformers are still available.
A UL tube Playmaster will do a lot better (about the same power, lower distortion) and be far simpler to build. And uses cheap TV triode-pentodes too. There must be a few folk who persist with Playmaster designs, because the transformers are still available.
mach1
What is the part number of the OT. I may have a match for your single tranny on my shelf amougst my "orphans". Fairly sure I have a single Furguson OP308/15, will need to check.
Else I definitely have a pair of Furguson OPM7A.
Oz Tranny Data
Australian Made Transformers
Cheers,
Ian
What is the part number of the OT. I may have a match for your single tranny on my shelf amougst my "orphans". Fairly sure I have a single Furguson OP308/15, will need to check.
Else I definitely have a pair of Furguson OPM7A.
Oz Tranny Data
Australian Made Transformers
Cheers,
Ian
That is coz it is A&R Tranny.
Keep going down that linked page and you will get to the A&R pages. You did'nt get duped.
It seems like Keit's linked source is doing A&R copies but I did'nt like the prices.
Probably cheaper to grab some ST35 Trannies from the States or buy Hammond 1608 (8K Raa) or 1609 (10K Raa). I've used both of these Hammond trannies and was quite satisfied with their quality. Seen good reports of the ST35 trannies but never tried one.
Will also check to see if I have an A&R 2672.
Off the air for next few days, or possibly only one, NBN installed my fibre modem and battery backup inside my house 2 weeks ago and Telstra come tomorrow to fit my router/WiFi Modem and change my land line phone connection to work on the NBN fibre. New Dell Laptop turned up yesterday so about to join the 21st century at home rather than relying on work facilities.
Cheers,
Ian
Keep going down that linked page and you will get to the A&R pages. You did'nt get duped.
It seems like Keit's linked source is doing A&R copies but I did'nt like the prices.
Probably cheaper to grab some ST35 Trannies from the States or buy Hammond 1608 (8K Raa) or 1609 (10K Raa). I've used both of these Hammond trannies and was quite satisfied with their quality. Seen good reports of the ST35 trannies but never tried one.
Will also check to see if I have an A&R 2672.
Off the air for next few days, or possibly only one, NBN installed my fibre modem and battery backup inside my house 2 weeks ago and Telstra come tomorrow to fit my router/WiFi Modem and change my land line phone connection to work on the NBN fibre. New Dell Laptop turned up yesterday so about to join the 21st century at home rather than relying on work facilities.
Cheers,
Ian
Last edited: