transformers are generally the most expensive parts in an amp, so I'm not all that amazed that when an amp comes this cheap, the trannies are low quality.
I don't think it would add much to the cost to add a few safety improvements to the transformers. Maybe $1 in material and time per transformer?
Even if they didn't use vacuum to impregnate the transformers, soaking them for 24 hrs and allowing them to dry for another 24 hrs shouldn't be that big a deal. Even a 12 hr soak (over night) in varnish and allow them to dry for 24 would be better than nothing and might have prevented this short. Even soaking the finished bobbin in hot pruified bees wax (ala 1940s) would be better than nothing.
Also there was a splice in layer 6 of the bottom primary that was left un-insulated. A bottle of clear fingernail polish could have insulated the joint after soldering and would have been better than leaving it bare to potentially short to another winding or layer.
Please understand I'm not just trying to throw stones but give constructive criticism. I think it is great that these amps are available at a resonable cost, but these are what I consider to be safety/quality issues that could be addressed without great expendeture.
Even if they didn't use vacuum to impregnate the transformers, soaking them for 24 hrs and allowing them to dry for another 24 hrs shouldn't be that big a deal. Even a 12 hr soak (over night) in varnish and allow them to dry for 24 would be better than nothing and might have prevented this short. Even soaking the finished bobbin in hot pruified bees wax (ala 1940s) would be better than nothing.
Also there was a splice in layer 6 of the bottom primary that was left un-insulated. A bottle of clear fingernail polish could have insulated the joint after soldering and would have been better than leaving it bare to potentially short to another winding or layer.
Please understand I'm not just trying to throw stones but give constructive criticism. I think it is great that these amps are available at a resonable cost, but these are what I consider to be safety/quality issues that could be addressed without great expendeture.
Last edited:
unfortunately, varnising of a transformer is NOT considered to improve the safety by any safety standard. varnishing is intended to fix the magnet wire to the location in order to eliminate hum and transformer noise.
I guess that only leaves adding margin tape at the edges to prevent the windings from migrating and slipping around the edge of the paper, resulting in a short like the one I found.
I can see that that would add labor and cost.
I was thinking varnish might have held the windings and prevented this but I guess not from what you say.
An externally hosted image should be here but it was not working when we last tested it.
I can see that that would add labor and cost.
I was thinking varnish might have held the windings and prevented this but I guess not from what you say.
transformers
in principle you are right by adding a margin tape. but there are also other ways if there is enough space. or the bobbin a 2 chamber bobbin is chosen. yes, this a can be done also for output trannies, although many people believe, it does not work good.
in principle you are right by adding a margin tape. but there are also other ways if there is enough space. or the bobbin a 2 chamber bobbin is chosen. yes, this a can be done also for output trannies, although many people believe, it does not work good.
What other techniques are there besides the margin tape or multiple bobbins? I think the transformer discussed in the Williamson UL Amp (1947 or 1948?) utilized split bobbins.
I got these numbers when I measured the transformer that was still good.
Primary Inductance Lp – 5.4H
Primary Leakage Inductance Ls – 7.7mH
Primary Resistance Rp – 56 Ohm/62 Ohm
Secondary resistance Rs – 0.256 Ohm
Coupling Capacitance - 0.717nF
The frequency response in a 6P1P PP amp without Global Negative Feedback was almost as good as the Hammond and Olson transformers I have, so the transformers work pretty well. -3dB was at 30Hz compared to 20Hz for the hammond, so not bad at all. With a little GNFB it should clean up to 20Hz or better.
Considering the cost constraints the transformers are quite nice.
Hopefully improvements will continue to be made.
I got these numbers when I measured the transformer that was still good.
Primary Inductance Lp – 5.4H
Primary Leakage Inductance Ls – 7.7mH
Primary Resistance Rp – 56 Ohm/62 Ohm
Secondary resistance Rs – 0.256 Ohm
Coupling Capacitance - 0.717nF
The frequency response in a 6P1P PP amp without Global Negative Feedback was almost as good as the Hammond and Olson transformers I have, so the transformers work pretty well. -3dB was at 30Hz compared to 20Hz for the hammond, so not bad at all. With a little GNFB it should clean up to 20Hz or better.
Considering the cost constraints the transformers are quite nice.
Hopefully improvements will continue to be made.
TheGimp, I think it needs to be made clear that the transformer you dissected is not from a Suppo model, just in case some readers may be thinking this, it is a Chinese transformer used in a Meng model. Suppo OPT's do not look the same, and the quality issues could be similar or not. More scary, would be quality issues with power transformers.
Regarding the freq response of the OPT, the Hammond 1608/1609 have 25% more mass than the Meng tranny you tested, that is, if it weighs the same as mine 🙂 So the Meng OPT does not do too bad at all, sound-wise that is?
Ian.
Regarding the freq response of the OPT, the Hammond 1608/1609 have 25% more mass than the Meng tranny you tested, that is, if it weighs the same as mine 🙂 So the Meng OPT does not do too bad at all, sound-wise that is?
Ian.
Last edited:
sSppo transformers
we now are working together to improve the Suppo transformers safety and performance.
the tests are already finished.
we now are working together to improve the Suppo transformers safety and performance.
the tests are already finished.
Sorry if I did not make that clear Ian, I was hoping to add a specific example of how a transformer could fail and add to the discussion of how to improve them.
transformer again
this is a tricky thing: for transformers there is the european standard EN61558, for US its' different. there are many considerations inside that standard, how a transformer should be constructedfor its intended purpose. this field is just too wide to be discussed here on the forum in details.
this is a tricky thing: for transformers there is the european standard EN61558, for US its' different. there are many considerations inside that standard, how a transformer should be constructedfor its intended purpose. this field is just too wide to be discussed here on the forum in details.
Anybody tried other modifications, such as a CCS? For both channels I used a LM317 as CCS. After listening to it for several days I still like the sound.
Maarten
Maarten
Adding a picture, cathode resistors R11 (two of 270 Ohms) are removed from the PCB as well as the bypass capacitors C4 of 220 uF. btw: I put two resistors of 47 Ohms in parallel to obtain a good value for the CCS.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Nice Mod there Maarten. Could you describe the improvements/changes in the sound after your CCS mod? BTW the unit I have uses two paralleled 220uf 25V caps for each channel's cathode bypass. Did your unit use the same combination? I had originally wanted to replace the single 220uf cap/channel as shown the the schmatic with a 1500uf 50V cap/channel. So should I still replace the 440uf combo with the 1500uf cap? I really dont want to make any changes if it would not make much difference. Now, if your mod is a big improvement then maybe another round of surgery is in order.
Regards,
David
Regards,
David
What's the (technical) advantage in using a CCS instead of a common cathode resistor in this Class AB PP amp? 😕
Hello,
you won´t have to check the bias settings very often, especially when the tubes get old and wear out.
George
you won´t have to check the bias settings very often, especially when the tubes get old and wear out.
George
Ccs
as the tubes may age differntly in gm (mA/V) using a single CCS, this may cause a problem of unbalanced working increasing hum and transformer saturation
as the tubes may age differntly in gm (mA/V) using a single CCS, this may cause a problem of unbalanced working increasing hum and transformer saturation
Current will be constant for both halves but indeed the balance might shift if one tube ages faster than the other. This is the same for the original resistor bias setup as well.
Can change that (add a balance pot or use a separate CCS per channel, but then I'd need to add a bypass capacitor again) but at this moment there is absolutely no humm. And as I plan to build a new amp from scratch soon, I leave it as is at the moment.
(only did put some crimp tube over the two resistors for better electrical shielding)
Ccs
constant in the sum of both tubes plates and including (changing) screen grid currents. so, somewhen, the stronger tube ( higher gm) will be forced to worke harder, to compensate, and even, maybe even over its limits of dissipation. sharing the same autobias or same CCS bias is not always a good idea. i know, there is not a perfect simple solution. best is then: SE, no balancing act.
constant in the sum of both tubes plates and including (changing) screen grid currents. so, somewhen, the stronger tube ( higher gm) will be forced to worke harder, to compensate, and even, maybe even over its limits of dissipation. sharing the same autobias or same CCS bias is not always a good idea. i know, there is not a perfect simple solution. best is then: SE, no balancing act.
new
for the forum there is another thing to fiddle and play around with: search for "miniwatt" amplifier. besides that, the miniwat is a copyright brand , and prtocted still. its a SE mini amp, manufactured: inChina!where else.
And it incorporates a switch mode power supply. I am keen, to have a look to the SMPS etc. How far we can go down? I guess the test result already, at least for safety. It is marketed via Hong Kong. interested DiY may look here, but keep fingers insulated after delivery: MiniWatt Tube Amplifier
for the forum there is another thing to fiddle and play around with: search for "miniwatt" amplifier. besides that, the miniwat is a copyright brand , and prtocted still. its a SE mini amp, manufactured: inChina!where else.
And it incorporates a switch mode power supply. I am keen, to have a look to the SMPS etc. How far we can go down? I guess the test result already, at least for safety. It is marketed via Hong Kong. interested DiY may look here, but keep fingers insulated after delivery: MiniWatt Tube Amplifier
- Status
- Not open for further replies.