It would be nice if we were all using the same set up but any reliable system will do. I'm using a behringer ECM8000 and ARTA (also free but you can't save data) and will probably move to HolmImpuls soon.
In most cases we will be comparing A to B and exact calibrated measurements are not required.
Hi
@Alexberg
The Triad toroid in Your pdf features 4 equal windings. The maximum U will be 1:4, which is by far too less for ESL usage. A 500VA core is far more than You need. The Amplimos for example use a 100VA core.
But what makes it really problematic are the 4 separate windings. You need not only to know up to which voltage prim-sec windings are tested, but to which voltage prim1-prim2 and sec1 to sec2 are tested. As in the case of the Triad the latter two values are not specced. Typically only the voltage rating between prim and sec is high enough for ESL usage. The inter-primary and inter-secondary voltage ratings are usually much lower (e.g ~500V).
So if You don´t know those specs the only safe way is to connect two toroids with single primary/single secondary winding together. Since now the highvoltage ESL-side of the trannies is connected in a series connection the prim-sec voltage rating is doubled. If this value is specced for example 4kV the output voltage to the ESL may reach 8kV peak-peak, equal 2.8kVrms.
jauu
Calvin
@Alexberg
The Triad toroid in Your pdf features 4 equal windings. The maximum U will be 1:4, which is by far too less for ESL usage. A 500VA core is far more than You need. The Amplimos for example use a 100VA core.
But what makes it really problematic are the 4 separate windings. You need not only to know up to which voltage prim-sec windings are tested, but to which voltage prim1-prim2 and sec1 to sec2 are tested. As in the case of the Triad the latter two values are not specced. Typically only the voltage rating between prim and sec is high enough for ESL usage. The inter-primary and inter-secondary voltage ratings are usually much lower (e.g ~500V).
So if You don´t know those specs the only safe way is to connect two toroids with single primary/single secondary winding together. Since now the highvoltage ESL-side of the trannies is connected in a series connection the prim-sec voltage rating is doubled. If this value is specced for example 4kV the output voltage to the ESL may reach 8kV peak-peak, equal 2.8kVrms.
jauu
Calvin
Dear Calvin,
I did not get what kind of transformer CharlieM needs: I definitely assumed that trafos are tested primary to secondary - otherwise it would be custom. Inter primary and intersecondary usually are not specified and limited to wire breakdown voltage, as you mentioned. The pdf was just an example (power and ratio wise). As soon as primaries are in series/secondaries in parallel or vice-versa, two on N trafos will share currents/voltages in proportion to respective turn ratios... which you surely know
If I needed standard transformer with specified (and tested) secondary to secondary and primary to primary voltage, I would check C-core transformers: secondaries and primaries on the different legs definetly have the isolation needed, but they are not TOROIDS.
What I also do not get - why is high isolation voltage needed between secondaries (better say low voltage)?
On the other hand, secondary winding is usually an outside one - it is easily removable/rewindable with proper isolation. Number of turns is not that high as well.
Alex
P.S. Was he looking for the toroidal trafo with single primary single secondary? Even then one can not safely apply 2400 VAC @ 500Hz instead of 240 VAC @ 50Hz
Thanks for posting about this guys. Finding usable trannies at a reasonable price is a HUGE problem for many builders. I posted the query hoping someone would know a U.S. distributor (or source that would ship to U.S.), manufacturer and part number for toroids comparable to those described in one of Calvin's threads, which I believe were 30VA or 50VA, 230v/6v with single primaries and secondaries and wired in tandem to yield a 68:1 step-up ratio. I suppose we can't know for sure that a given [untried] toroid with dual primary/secondaries wouldn't short-out but I would hate to risk smoking my amps to find out. The few sources I was able to find for single primary/secondary toroids were in Europe and those sources either didn't respond to my emails or wouldn't ship to the U.S. I think what electronically-challenged new builders really need is someone to basically say "here's a reasonably good and inexpensive toroid that's tried-and-true and here's where you can buy it and this is how you wire it up"
Thanks for posting about this guys. Finding usable trannies at a reasonable price is a HUGE problem for many builders. I posted the query hoping someone would know a U.S. distributor (or source that would ship to U.S.), manufacturer and part number for toroids comparable to those described in one of Calvin's threads, which I believe were 30VA or 50VA, 230v/6v with single primaries and secondaries and wired in tandem to yield a 68:1 step-up ratio. I suppose we can't know for sure that a given [untried] toroid with dual primary/secondaries wouldn't short-out but I would hate to risk smoking my amps to find out. The few sources I was able to find for single primary/secondary toroids were in Europe and those sources either didn't respond to my emails or wouldn't ship to the U.S. I think what electronically-challenged new builders really need is someone to basically say "here's a reasonably good and inexpensive toroid that's tried-and-true and here's where you can buy it and this is how you wire it up"
Hi,
@Alexberg
the idea of using two standard power toroids is to offer a solution to someone who:
- wants to drive a hybrid-panel (above ~100Hz)
- wants to use standard off the shelf parts
- wants to use easy to source parts (at least as we learned its easy in europe)
- wants to use parts which don´t burn craters in Your wallet
- wants parts he can be sure to withstand high voltages (the ones I used had a rating of 4kVeff -->5.7kVp, times two: 8kVeff -->11.3kVp. sec-to-sec would have been only 500Veff-->700Vp drive)
- wants parts that are high quality by manufacturing (specified data)
- wants parts with high quality ´behaviour´
- doesn´t want to rewind a tranny himself, test rewind again, test...throw away
It is to understand that this solution is not necessarily the best or that it suits Your demands at all. It is one solution out of many with the advantage of low needed effort and very good technical and acoustic results.
jauu
Calvin
@Alexberg
the idea of using two standard power toroids is to offer a solution to someone who:
- wants to drive a hybrid-panel (above ~100Hz)
- wants to use standard off the shelf parts
- wants to use easy to source parts (at least as we learned its easy in europe)
- wants to use parts which don´t burn craters in Your wallet
- wants parts he can be sure to withstand high voltages (the ones I used had a rating of 4kVeff -->5.7kVp, times two: 8kVeff -->11.3kVp. sec-to-sec would have been only 500Veff-->700Vp drive)
- wants parts that are high quality by manufacturing (specified data)
- wants parts with high quality ´behaviour´
- doesn´t want to rewind a tranny himself, test rewind again, test...throw away
It is to understand that this solution is not necessarily the best or that it suits Your demands at all. It is one solution out of many with the advantage of low needed effort and very good technical and acoustic results.
jauu
Calvin
Dear Calvin and CharlieM,
working in US I have never had problems finding something,
but it would be fare to say that the trafos were not always in this category.
On the other hand people here may not have an electronic background.
NEWARK/FARNELL is kinda pricey, nonetheless:
MULTICOMP|MCFG080/15|TOROIDAL TRANSFORMER VOUT: 15V | Newark.com
230 single primary, 80 VA, $21.56 + $20 cargo surcharge to continental US
Is it cheap enough... I do not know.
Best regards,
Alex
I would also go to the flea markets (fairs) or try to talk to HAM Radio guys...
Last edited:
Sorry, I'm a bit behind the curve here...
is there a link to the "beam splitter"? (wazzat?)
quick cursory read says to me that the requirements for a HF horn loaded ESL may be different than a direct radiator (WRT stators) since the horn has to behave like a compression driver of some sort?
there was a third thing that came to mind, but it is AM here just got up, and I forgot it! heh...
Edit: oh yeah - what if ur using wire stators??
is there a link to the "beam splitter"? (wazzat?)
quick cursory read says to me that the requirements for a HF horn loaded ESL may be different than a direct radiator (WRT stators) since the horn has to behave like a compression driver of some sort?
there was a third thing that came to mind, but it is AM here just got up, and I forgot it! heh...
Edit: oh yeah - what if ur using wire stators??
Last edited:
Hi,
as far as I remember the term ´beam splitter´refers to R.Sanders´s Cookbook. The idea behind is to avoid that the bass casing behind the panel doesn´t reflect soundwaves immediately back through the panel, which would mean disturbing early reflections. Instead the splitter should distribute or scatter the soundwaves.
jauu
Calvin
as far as I remember the term ´beam splitter´refers to R.Sanders´s Cookbook. The idea behind is to avoid that the bass casing behind the panel doesn´t reflect soundwaves immediately back through the panel, which would mean disturbing early reflections. Instead the splitter should distribute or scatter the soundwaves.
jauu
Calvin
Hi Bear,
stator open percentage is just one of the examples. There are many more aspects of ESLs that need more looking into (in my opinion). This includes wire stators and ultimately how they perform versus perforated plate stators. I think there's a lot to be said for wire stators but without proper testing it's more or less guess work. But how to test this properly?
The stator open percentage is not too hard to test. Someone has to build a number of the same stators but with different open percentage. These don't have to be full blown ESL panels, I think small 12x5 cm panels would be fine. For example constructed from circuit board. The required number of panels quickly explodes if we want to vary hole spacing and hole size and if we expect the two to interact. If we don't expect that, then we can first examine open percentage and later hole size separately.
If we start with varying open percentage I would expect to see changes in:
* sensitivity
* capacitance
* damping
* stator resonances
For example for the capacitance I suspect (based on observations in the past) that it will not change linearly with (stator surface area - hole area).
stator open percentage is just one of the examples. There are many more aspects of ESLs that need more looking into (in my opinion). This includes wire stators and ultimately how they perform versus perforated plate stators. I think there's a lot to be said for wire stators but without proper testing it's more or less guess work. But how to test this properly?
The stator open percentage is not too hard to test. Someone has to build a number of the same stators but with different open percentage. These don't have to be full blown ESL panels, I think small 12x5 cm panels would be fine. For example constructed from circuit board. The required number of panels quickly explodes if we want to vary hole spacing and hole size and if we expect the two to interact. If we don't expect that, then we can first examine open percentage and later hole size separately.
If we start with varying open percentage I would expect to see changes in:
* sensitivity
* capacitance
* damping
* stator resonances
For example for the capacitance I suspect (based on observations in the past) that it will not change linearly with (stator surface area - hole area).
Alex, thanks for the tip on the Multicomp trannys from Newark—this could be a solution.
Calvin, if I figure correctly, a pair of those 230v/12v trannys wired backwards in tandem with inputs parallel and outputs in series would yield only about a 30:1 step-up ratio, which would be lower than I want. However, I see that Newark also has a 50VA, 230v/6v Multicomp tranny that would yield the higher step-up ratio I’m looking for. Still, there is the question of flashover resistance for the dual windings on the 6v side. I’m wondering if a 250 watt amp would put out enough voltage to short across those windings-- and the higher voltage output side a single winding so perhaps these Multicomp trannys might be safe to use. I don’t see a data sheet for them though, and I don’t know what would be typical for the other characteristics like leakage, inductance, losses, etc.. I don’t really understand such requirements anyway. What are your thoughts on these 230v/6v Multicomp trannys for use above 200 hz?
Arend-jan , great topic .... Agree and would like to see this evolve into a proper platform to test and compare various topics and Opinions.
Suggestion to set a course instead of hoping around on different topics , moving on when finished for eg.
1. Comparing transformer types, wire configuration etc.
2. Wired ESL vs Holey, hole type
3. flat panel vs curved
4. Diaphram coatings
5. spacer gaps
6. Hybrid config type
regards,
Suggestion to set a course instead of hoping around on different topics , moving on when finished for eg.
1. Comparing transformer types, wire configuration etc.
2. Wired ESL vs Holey, hole type
3. flat panel vs curved
4. Diaphram coatings
5. spacer gaps
6. Hybrid config type
regards,
Hi,
ohh come on guys. Those toroid trannies, what You can use them for, which parameters to observe etc etc, have been described here and in their own thread. Believe the words or not, test such trannies or not, but don´t expect me to evaluate devices via forum, especially not devices that deviate from the recommendations. Besides, I´m sure this theme already spoils the thread.
jauu
Calvin
ohh come on guys. Those toroid trannies, what You can use them for, which parameters to observe etc etc, have been described here and in their own thread. Believe the words or not, test such trannies or not, but don´t expect me to evaluate devices via forum, especially not devices that deviate from the recommendations. Besides, I´m sure this theme already spoils the thread.
jauu
Calvin
Yes, but first we have to define models for the things we want to investigate. We could just jump in and start measuring whatever comes to mind, but that does not get us much more than results for the specific case at hand. Which could still be of interest nonetheless.
The scientific way would be to make a model, use it to predict behavior for a specific case, then confirm the model by performing controlled measurements.
Making the model is often the hard part.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.