SET Asylum

Single Ended Triodes (SETs), the ultimate tube lovers dream.

Return to SET Asylum

Message Sort: Post Order or Asylum Reverse Threaded

Need some opinions and maybe a lot of help with 300B design.

75.132.155.75
Posted on March 14, 2016 at 10:38:13
gavinhaley
Audiophile

Posts: 29
Location: St Louis
Joined: January 31, 2016



Hi all. I have had a 396A linestage half built for a few years. Just couldn't find the time to finish it. I had all the parts already procured for a 12B4A amplifier to go with it. Both designs were designed around Magnequest iron. After buying a SE 6L6G and listening for a few weeks I feel I need more watts per channel. I'm using a pair of FE206Es in reflex cabinets so I would guess they are about 97 dB 1w/1m in a fairly small room but the 6L6s just don't have enough grunt as nice as they sound. So I came up with a design for a 300B amplifier that utilizes parts I already own. I will need to buy some chokes and a power transformer but I have most of it. Here is what I have so far. A 396A transformer coupled to a 12B4A which is driving a 300B. Initial thoughts?

Gavin

 

Hide full thread outline!
    ...
RE: Need some opinions and maybe a lot of help with 300B design., posted on March 14, 2016 at 13:44:58
Paul Joppa
Industry Professional

Posts: 7295
Location: Seattle, WA
Joined: April 23, 2001
 A reasonable design, but you won't get a lot of power - maybe 3-4 watts. To get more, raise the 300B plate voltage or reduce the output transformer impedance (and raise the plate current). Or do both.

Here is the WE table of 300B operating points:


p-k	Bias	Current	Load	Power	2nd	3rd
Volts	Volts	mA	Ohms	Watts	db	db
200	-42	30	2000	 3.0	20	31
200	-39	40	2500	 2.6	26	38
200	-37	50	2500	 2.5	30	45
250	-55	30	2000	 4.9	18	27
250	-55	30	4500	 3.2	27	40
250	-52	40	3000	 4.0	26	36
250	-50	50	2500	 4.4	26	39
250	-48	60	2000	 4.7	26	38
250	-48	60	2700	 4.1	30	45
250	-45	80	1500	 5.0	26	41
300	-65	40	2500	 6.7	20	30
300	-63	50	2000	 7.2	21	29
300	-63	50	3000	 6.1	26	37
300	-61	60	2400	 6.6	26	37
300	-61	60	3400	 5.6	30	44
300	-58	80	1700	 7.5	26	37
350	-76	50	3600	 7.8	26	38
350	-76	50	5000	 6.2	30	45
350	-74	60	2000	10.2	21	30
350	-74	60	3000	 8.3	26	38
350	-74	60	4000	 7.0	30	44
350	-71	80	2200	 9.6	26	39
400	-91	40	5000	 8.4	26	37
400	-89	50	3000	11.5	21	31
400	-89	50	4000	 9.4	25	38
400	-87	60	3500	10.5	26	38
400	-87	60	5000	 8.3	30	46
400	-84	80	2500	12.5	25	37
450	-104	40	6000	 9.5	26	38
450	-102	50	5000	10.7	27	39
450	-102	50	6500	 9.0	30	45
450	-100	60	4000	12.5	26	38
450	-100	60	5500	10.1	30	44
450	-97	80	2000	17.8	21	30
450	-97	80	3000	14.6	26	37
450	-97	80	4500	11.5	31	45

When the second harmonic distortion at full power is 26dB down, that's 5%. That's a good middle-of-the-road value IMHO.

 

RE: Need some opinions and maybe a lot of help with 300B design., posted on March 14, 2016 at 13:55:01
gavinhaley
Audiophile

Posts: 29
Location: St Louis
Joined: January 31, 2016
 Hi Paul. I guess I'm not getting it. I was aiming for 350 on the Plate, -74 on the grid which with a 5000 ohm load should give me around 6.5 watts or so. Am I looking at this wrong? I can increase the cathode voltage to -76 easily enough. That should give me at least 6 watts right? Thank you for taking the time to look at it by the way.

Gavin

 

RE: Need some opinions and maybe a lot of help with 300B design., posted on March 14, 2016 at 13:56:49
gavinhaley
Audiophile

Posts: 29
Location: St Louis
Joined: January 31, 2016



Here is my attempt at a power supply for the above. Yes I know the 5AR4 is not drawn correctly but the liitle sketch-it web app @ Digikey is kinda limited. Best I could do.

Gavin

 

RE: Need some opinions and maybe a lot of help with 300B design., posted on March 14, 2016 at 15:25:19
Triode_Kingdom
Audiophile

Posts: 10044
Location: Central Texas
Joined: September 24, 2006
 The schematic you posted indicates cathode bias. That means the anode-cathode differential will be less than what the PS provides relative to ground.

 

RE: Need some opinions and maybe a lot of help with 300B design., posted on March 14, 2016 at 17:35:37
Paul Joppa
Industry Professional

Posts: 7295
Location: Seattle, WA
Joined: April 23, 2001
 Exactly right!

T_K knows this but for others who might read this post and scratch their heads I'll note that it's a classic mistake, probably everyone here has fallen for it at some point: There is no such thing as "plate voltage"; there are only voltage differences, e.g. plate to cathode in the table and on data sheets, plate to ground, etc.

 

RE: Need some opinions and maybe a lot of help with 300B design., posted on March 14, 2016 at 19:35:11
hifipaul
Audiophile

Posts: 735
Location: NY
Joined: December 22, 2008
Try diode bias on the 396.

 

RE: Need some opinions and maybe a lot of help with 300B design., posted on March 14, 2016 at 21:13:44
gavinhaley
Audiophile

Posts: 29
Location: St Louis
Joined: January 31, 2016
 What a dumb mistake on my part. What makes it even worse is that Edcor does not make a transformer that goes above 900v ct. HHHHMMMMMM.

 

RE: Need some opinions and maybe a lot of help with 300B design., posted on March 14, 2016 at 23:16:37
maxhifi
Audiophile

Posts: 584
Location: Alberta, Canada
Joined: August 4, 2004
 It's not so bad, just increase that 10uf input capacitor to say 40 or 50uF. Download PSUD and play around to get the right voltage level, you're on the right track. If you wanted a choke input power supply you would need a higher voltage transforner but considering this is a class "a" amp, which has little variation in current demand from zero to full load, the better voltage regulation of a choke input supply doesn't offer any real advantage and then you may have to deal with a buzzing choke.

Also, I have a couple other comments on the power supply

It's kind of arbitrary to go dual mono at the first choke - why not either go full mono with a separate power supply for each channel,l (considering the amount of iron in this amp, there's little reason not to) or just use a single power supply for both sides.

One more comment: layout is going to be really important here, there is tons of opportunity to pick up hum with all those transformers, so to be fool proof go for a big chassis with lots of space between the power supply and the amplifier transformers, and be careful not to make any ground loops.

Another thing I have to ask: what's the intention of the interstate transformer between the first two stages? It's a 1:1 transforner and isn't isolating DC due to the 1uf coupling capacitor.. I think it's a redundant component, couldn't you just use a grid reaistor on the second tube and cut that transformer out of the circuit ? Speaking of which, why use the choke on the first tube instead of a resistor for the plate load? It's not as though this stage needs to produce the extra output voltage a choke will permit, and it will introduce a low frequency pole which won't help the overall performance of the amplifier. Also, those coupling capacitors are way bigger than they need to be.. I didn't calculate values but I would expect 0,1uf - 0,47 to be absolutely fine, and still allow the transformers to dominate the low frequency roll off.

One final comment: a single ended 6L6 will make about the same amount of power as this amplifier, if what you're after is more "grunt" why not try say a 15 watt push-pull amplifier, I don't think 20% more or less power will make a material difference, you may want substantially more for the "grunt" you're after.

 

RE: Need some opinions and maybe a lot of help with 300B design., posted on March 15, 2016 at 00:50:49
gavinhaley
Audiophile

Posts: 29
Location: St Louis
Joined: January 31, 2016
 Thank you for your reply. I am beginning to rethink my plan of attack here. My funding is limited and this amp, even with the parts I already have on hand is going to be a killer on the wallet. I am considering a different design using different tubes. I really don't like the way PP amps sound so I may build another SE EL34 amplifier with just the 396A. I know what you mean about layout. I was going through what I would have to do in my head and it was giving me a headache.

 

Many many problems, posted on March 15, 2016 at 10:21:24
drlowmu
Manufacturer

Posts: 9730
Location: East of Kansas City
Joined: January 10, 2005
 Very ordinary ho-hum, mediocre at best, it begs for redesign ...

(1) Too many stages

(2) TWO coupling caps...... needs to be ZERO !!

(3) Can not use electrolytics on Rks, ever in such a design.

You likely need to better-wire your ENTIRE audio chain.

Have fun,

Jeff Medwin

 

RE: Many many problems, posted on March 15, 2016 at 22:52:38
gavinhaley
Audiophile

Posts: 29
Location: St Louis
Joined: January 31, 2016
 Thanks. That was most helpful. Ta Ta.

 

RE: Need some opinions and maybe a lot of help with 300Bsign.ree, posted on March 16, 2016 at 07:58:08
maxhifi
Audiophile

Posts: 584
Location: Alberta, Canada
Joined: August 4, 2004



Another thing to consider is using a well known design as a starting point, for example the JE labs circuit, it's more or less the most ubiquitous entry level 300B amplifier on the 'net, and won't cause any major headaches during construction. If you decide it's enough power you could later modify it with fancy stuff like plate load chokes and interstage transformers. But this will be a quick path to a proven amp design and will also let you know if it's enough power without breaking the bank.

 

RE: Many many problems, posted on March 16, 2016 at 09:10:12
drlowmu
Manufacturer

Posts: 9730
Location: East of Kansas City
Joined: January 10, 2005
 Hi Gavin,

Not sure what Ta Ta means !!

If SE 6L6s on 97 dB speakers in a small room does not have enough "grunt", it is simply because you have designed and executed your amp sub-optimally.

Poor transfer efficiency. Really !!

A 300B is NO big deal, and is not a particularly good tube.

You need better amp building skills, especially amp ( and system's ) wiring, and a re-do of your amp's power supply.

Jeff Medwin

 

RE: Many many problems, posted on March 16, 2016 at 10:33:27
maxhifi
Audiophile

Posts: 584
Location: Alberta, Canada
Joined: August 4, 2004
 I think that a couple things could be at play there. But one of them is that there's no way that speaker is actually 97dB sensitive across the board - see link below.

If we are dealing with a 94.6 dB speaker, and let's say the single ended 6L6 puts out say 8 Watts (being generous), this gives a maximum SPL at 1m of about 103.6dB , but less so at bass frequencies.

Out in the middle of the room this is going to fall off to a much lower level, (for example, theoretically it would drop to about 93dB at three meters) depending on the listening position, and the amp is going to clip during peaks, if the music being played has any sort of dynamic range.

A 300B will put out a similar amount of power, which doesn't solve the "grunt" problem at all.

I bought Klipsch La Scalas to overcome the lack of dynamic range with my single ended amplifier, and it worked out well.

This is why I suggested a push-pull amplifier, the good ones sound excellent and really this speaker isn't sensitive enough for a single ended amp except for low level or non complex music of limited dynamic range.

Also note that this speaker is rated for 30 Watts, and capable of 90 Watt peaks. I bet if you connected it to a 50W amplifier it would really wake up, and you could use it to its potential instead of being stuck listening to amplifier clipping.

 

RE: Many many problems, posted on March 16, 2016 at 15:10:32
drlowmu
Manufacturer

Posts: 9730
Location: East of Kansas City
Joined: January 10, 2005
 OK, I TOTALLY agree on the need for efficient speakers, as being THE priority. Allows one to use the best sounding tubes.

I owned LaScalas long ago, and presently am setting up ALTEC A7 VOTTs in my 12 by 18.5 foot listening room. It allows me to effectively hear a lovely sounding Type 45 SE amp, with all the volume I can stand.

Jeff Medwin


 

RE: Many many problems, posted on March 16, 2016 at 17:24:47
Tre'
Industry Professional

Posts: 17294
Location: So. Cal.
Joined: February 9, 2002
 "Out in the middle of the room this is going to fall off to a much lower level, (for example, theoretically it would drop to about 93dB at three meters) depending on the listening position..."

Are you forgetting about the reverberate field?

I can walk all around my listening room and the overall spl changes very little.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Need some opinions and maybe a lot of help with 300B design., posted on March 16, 2016 at 18:55:32
Paul Joppa
Industry Professional

Posts: 7295
Location: Seattle, WA
Joined: April 23, 2001
 This should give you much more than 375 volts, closer to 500 volts. With that, you can run around 60mA (using a 1500 ohm cathode resistor) and drive 8+ watts through the Robin Hood 5K output transformer.

 

RE: Many many problems, posted on March 16, 2016 at 19:38:54
maxhifi
Audiophile

Posts: 584
Location: Alberta, Canada
Joined: August 4, 2004
 This is a really good, and somewhat surprising point, which I had never considered before, and cannot say I fully understand. I think it merits further reading, thanks for raising it.

 

RE: Many many problems, posted on March 16, 2016 at 21:53:27
Tre'
Industry Professional

Posts: 17294
Location: So. Cal.
Joined: February 9, 2002



A 6db loss for every doubling of distance only applies outdoors (in a wide open space) where there is nothing for the sound to bounce off of and return to the listener.

I have mixed a lot of live rock shows outdoors but only once, that I can think of, was in a wide open space with nothing for the sound to bounce off of.

That was the Cherry Festival, Traverse City Michigan. The stage was set up at the end of the peninsula facing back towards the city. **The concrete all around the peninsula is lower than the grass area and the one small building is about 250ft from the stage.**

This is a case where a recording made off the 2 bus (the signal from the mixing console that drives the amplifiers that drive the speaker system) can sound good.

Normally the mixer has to compensate for the "ills" of the room and a 2 bus recording (known as a "board recording") sounds awful.

Being totally outdoors, the only thing I was compensating for was the speakers. The board recording actually sounded a lot like the show.

Note, a normal SPL meter measures only direct sound. They do make special SPL meters that measure direct plus reverberate sound.

But you don't even need one. Just walk around your room and listen. 6db is a lot, 12db is huge. You will hear spl differences as you move around your listening room but they will be due to standing waves within the room, not distance from the speakers. And unless you have some really bad standing waves in you room you will not find large spl changes. Changes in the sound, yes, but not large changes in overall spl.


BTW "If we are dealing with a 94.6 dB speaker, and let's say the single ended 6L6 puts out say 8 Watts (being generous), this gives a maximum SPL at 1m of about 103.6dB" I think we all have stereos.

Two 94.6db speakers, each with 8 watts driving them to 103.6db spl, would yield 106.6db spl total.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Many many problems, posted on March 17, 2016 at 07:01:37
maxhifi
Audiophile

Posts: 584
Location: Alberta, Canada
Joined: August 4, 2004
 This makes a whole lot of sense. Concert sound looks like a real challenge, especially outdoor concerts!

The reason I didn't account for the sum of two speakers, is I tend to listen to a fair niumber of recordings with exaggerated stereo separation, so I didn't want to consider that the two channels are contributing equally. Under ideal circumstances they would be but I prefer to have my system designed for worst case.

I still think 8W is inadequate for those speakers under a wide range of conditions, based on empirical experience, but the how and why of it acoustically are not so much within my grasp as I thought.

One thing I do wonder - if it's the case that the volume doesn change much with distance,
Why is it that when I use arrays instead of usual speakers, that the volume seems more consistent with respect to distance? I mean empirically if I have a line array playing I can walk right up to it and to the other side of the room and the volume is the same everywhere,
Whereas when I listen to, say, a bookshelf speaker it seems louder up close.

Is it just that the reverberant field begins farther away for the bookshelf speaker ?


 

Bass VS La Scalas, posted on March 17, 2016 at 08:25:29
Triode_Kingdom
Audiophile

Posts: 10044
Location: Central Texas
Joined: September 24, 2006
 I'm not sure any of this is relevant given the fact that La Scalas are so bass deficient. In fact, lack of bass response is the fly in the ointment relative to most "high efficiency" speaker systems. Unless you want to bi-amp, or you have room for a huge straight or folded horn, it won't be a full-range system.

 

RE: Many many problems, posted on March 17, 2016 at 08:35:11
Tre'
Industry Professional

Posts: 17294
Location: So. Cal.
Joined: February 9, 2002
 "Is it just that the reverberant field begins farther away for the bookshelf speaker ?"

For those particular speakers it must be...I guess.

I do know that the more directional the speaker is, the farther away the reverberant field begins.

"I still think 8W is inadequate for those speakers under a wide range of conditions"

That could very well be. I didn't meant to commit on that position per se, just the technical reasoning behind it.

P.S. I had another thought about the line array. It's kind of hard to get close to all the drivers at the same time vs. a small 2 way bookshelf speaker.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: Many many problems, posted on March 17, 2016 at 09:29:41
maxhifi
Audiophile

Posts: 584
Location: Alberta, Canada
Joined: August 4, 2004
 "could very well be. I didn't meant to commit on that position per se, just the technical reasoning behind it."

And thank you for doing so, more knowledge is always better!

 

RE: Bass VS La Scalas, posted on March 17, 2016 at 09:38:23
maxhifi
Audiophile

Posts: 584
Location: Alberta, Canada
Joined: August 4, 2004
 I suppose it wouldn't be life without compromise... I like the la scalas a lot but I can see why someone who is after deep bass response would be turned off them. I haven't got the space or budget for a more involved horn loaded system with better LF extension, and I really enjoy the sound of the la scalas. Plus, I don't like to imagine moving them again :)

 

RE: Bass VS La Scalas, posted on March 18, 2016 at 05:23:31
Mr_Steady
Audiophile

Posts: 2042
Location: North Florida
Joined: August 19, 2014
 Audio seems to be a series of compromises. With folded W bass bins you get tight and low distortion bass, but it doesn't go very deep, but that can be easily solved by adding a subwoofer. However, a 15" direct radiator in a ported box can never be made to be as low distortion and as tight sounding as horn bass. You get to choose.


---------------------------------------------------------------
Big speakers and little amps blew my mind!

 

Spot on assessment... horn bass embodies... Accuracy.~nT, posted on March 20, 2016 at 08:55:00
Cleantimestream
Audiophile

Posts: 7550
Location: Kentucky
Joined: June 30, 2005
 ~!
The Mind has No Firewall~ U.S. Army War College.

 

Not if the horns are short (nt), posted on March 21, 2016 at 20:31:44
Triode_Kingdom
Audiophile

Posts: 10044
Location: Central Texas
Joined: September 24, 2006
 nada aqui

 

No Sir, shorter horn merely less bass... No increase in distortion whatsoever, posted on March 23, 2016 at 08:39:10
Cleantimestream
Audiophile

Posts: 7550
Location: Kentucky
Joined: June 30, 2005
 !
The Mind has No Firewall~ U.S. Army War College.

 

RE: No Sir, shorter horn merely less bass... No increase in distortion whatsoever, posted on March 24, 2016 at 08:21:26
Triode_Kingdom
Audiophile

Posts: 10044
Location: Central Texas
Joined: September 24, 2006
 You can believe that I understand the working principles of horns. Your remark wasn't about distortion, it was about accuracy. A speaker isn't accurate if bass is down 5 dB at 50 Hz. I don't think that even qualifies as hi-fi.


 

RE: No Sir, shorter horn merely less bass... No increase in distortion whatsoever, posted on March 24, 2016 at 11:43:35
kyle
Audiophile

Posts: 1839
Location: London Ontario
Joined: September 29, 1999
 That would exclude a large number of stand mount "monitors", most Lowther designs, smaller Magneplanars some JBLs and Altec A5s & A7s.
You're a tough listener / measurer.

 

you are welcome to believe that Triode... music lives in the midrange and I have owned speakers that went to.., posted on March 24, 2016 at 14:14:21
Cleantimestream
Audiophile

Posts: 7550
Location: Kentucky
Joined: June 30, 2005
 20 Hz, the Khorn bass bins stomp the flab and gratuitous grumbling of any bass reflex design you care to name. As for the LaScala's ... and no bass below 50 CPS... I hope your room is BIG enough to swallow a 56 foot wave if you are so adament about having your 20 HZ.
The Mind has No Firewall~ U.S. Army War College.

 

Half Wavelength Theory (nt), posted on March 24, 2016 at 17:51:06
Tre'
Industry Professional

Posts: 17294
Location: So. Cal.
Joined: February 9, 2002
 .
Have Fun and Enjoy the Music
"Still Working the Problem"

 

Sure, one side of the room greater than 28 ft, how about 20 Hz swamping out a real important area..., posted on March 24, 2016 at 18:16:14
Cleantimestream
Audiophile

Posts: 7550
Location: Kentucky
Joined: June 30, 2005
 Like 50 Hz... A known fact.
The Mind has No Firewall~ U.S. Army War College.

 

Common misconception, posted on March 24, 2016 at 22:32:53
Triode_Kingdom
Audiophile

Posts: 10044
Location: Central Texas
Joined: September 24, 2006
 "I hope your room is BIG enough to swallow a 56 foot wave..."

I don't think you understand how pressure waves work. 20 Hz is easily accomplished in the confines of a car if the drivers are capable. More to the point, it really doesn't matter if your musical tastes favor the midrange. Accuracy requires the ability to reproduce the entire range. If a speaker can't do that, it's less accurate than one that can.

 

RE: Common misconception, posted on March 25, 2016 at 03:49:40
Cleantimestream
Audiophile

Posts: 7550
Location: Kentucky
Joined: June 30, 2005
 The following in terms of import... The room Trumps all other considerations, followed by the speakers and I daresay the preamp before the amp I have used transformer driven volume controllers...not dead and lifeless but definitely not as realistic sounding... Overload a small room all you want... It is not realistic.and surely, not accurate.
The Mind has No Firewall~ U.S. Army War College.

 

The requirements for accurarcy , posted on March 25, 2016 at 05:48:18
Mr_Steady
Audiophile

Posts: 2042
Location: North Florida
Joined: August 19, 2014
 I personally think there is more to accuracy than bandwidth. It may not even be the most important.

This conversation is making me scratch my head. Any direct radiator that is flat down to 20Hz is going to be 18 to 21" in diameter. That's so large that it's going to be put in it's own enclosure, and that makes it a sub-woof-er.

Maybe we should start a thread on the speaker asylum, and see how many multiway speakers we can come up with that are flat from 20Hz to 20kHz. I can't think of any, and I bet the ones that do have serious accuracy issues of their own for reasons other than bandwidth. Not that any of this matters, as I'm sure you are just being pedantic for your own amusement.


---------------------------------------------------------------
Big speakers and little amps blew my mind!

 

RE: The requirements for accurarcy , posted on March 25, 2016 at 22:04:50
Triode_Kingdom
Audiophile

Posts: 10044
Location: Central Texas
Joined: September 24, 2006
 This issue isn't one of absolutes. All else being equal, a musical reproduction system that is 3 dB down at 25 Hz is inherently more accurate than one with a 3 dB point of 55 Hz. Yes, low distortion counts for something, as does SPL. However, it simply isn't meaningful to say that one or the other of these attributes is most important, and to claim a system to be more accurate as a result of that characteristic alone. That's why I responded to the statement regarding the accuracy of bass horns the way I did. As a blanket statement, it isn't true.

"Any direct radiator that is flat down to 20Hz is going to be 18 to 21" in diameter."

That's not true either.











 

RE: The requirements for accurarcy , posted on March 26, 2016 at 07:45:06
Cleantimestream
Audiophile

Posts: 7550
Location: Kentucky
Joined: June 30, 2005
 Here, something incontrovertible, a horn loaded woofer design {or any compression driver implemented correctly} is 10 {ten} times lower in distortion than any cone or dome. If your mind cannot count that as important, your ears surely will.
The Mind has No Firewall~ U.S. Army War College.

 

OK...:), posted on April 10, 2016 at 10:29:10
PakProtector
Audiophile

Posts: 12363
Joined: May 14, 2002
 That IT will likely do better just upstream of the 300B. Its secondary will also give you a beneficial low-DC resistance path for the minimal bias grid currents( the ones that give rise to maximum grid circuit resistance specifications ).

cheers,
Douglas


Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

Page processed in 0.046 seconds.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2024, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]