New Wave Guide Study

[oneoldude]

Hi all,

I have been working with EdL (thanks Ed) for the last couple weeks in preparing to measure some DIY WGs for use with non-compression drivers. That is, the tweeters will be standard off-the-shelf dome tweeters. Ed has turned some examples and sent them to me. Soon I will be doing some measurements (I have to go out of town for work). The measurements and related WG profiles will be posted as done.

We are currently shooting for a guide to match the RS 180 7" Dayton MidWoofer. The principle idea being to control the sound field to match the RS 180 at crossover and to provide low end boost so that the crossover can be below 1200 Hz without heroic crossover designs.

At this stage of the game we are still not certain where a good foundation would be for a base line WG to compare results against. Since there are many of you out there that have seen and heard (and hopfully measured :T ) Genelec, Amphion, SP Tech, or other fine high quality WGs for dome tweeters, we are asking for some help.

So, has anyone out there physically measured any commercially successful Wave Guides for dome tweeters or have a well founded opinion of how they should be shaped and sized?

From looking at pictures of commercial WGs (and seeing the SP Tech speakers but not measuring them) it seems that basic dimensions for RS 180 use might be 7" to 7.5" mouth diameter, throat to match the tweeter used, and a conical flair rate of perhaps 100 deg (more? less?) total included interior angle for the Guide.

Can anyone confirm or deny these basic assumptions?

How about measured round-over radius at mouth and throat?

Does anyone have any actual physical measurements?

If you have taken physical measurements or acoustical measurements, please describe what it was you measured. Right now Zaph has the only data I am aware of. Great work Zaph! :T

If you wish to provide an opinion, please give a commercially successful example product that uses a standard dome tweeter and solid support (perhaps acoustic measurements?) for the opinion.

BTW we do not intend to clone someone else's work. But we would like to start at a good point and develop what we feel works for us.

We hope that with a little help from our friends we can find and deliver some useful data for our own projects and for the DIY community. Hope you can help.

Thanks everybody :T

P.S.

Attached is a pic of one of the WG's we are working on. It is sporting a Vifa silk dome. Can't wait to measure it!

[JoshK]

I have no measurements to add. I have seen both the GedLee Summa and the earlier version of the SP Tech monitor. Both seems to use roughly the same type of profile (obviously of different sizes), that is the Geddes style OS with both a roundover at the mouth and the throat (just a bit at the throat).

The spreadsheet that Ed and I worked on will draw out some of these profiles.

[Dennis H]

Yeah, Geddes's oblate spheroid formula works out very close to:

1. A cone going to zero diameter at the back side of the horn
2. Bore out the throat diameter
3. Round a smooth transition at the throat between the bored throat and the cone
4. Round the outside of the cone at the mouth (not part of the OS equation)

I have a spreadsheet that can do these but it sounds like Josh already fixed you guys up. Another thing, most standard tweeters already have something of a throat curve so just doing whatever it takes to smoothly blend between that and the conical part of the horn sounds like the hot ticket.

[fcserei]

New Wave Guide Study

My take on the Waveguide:

modified 12" EV woofer with a Peerless soft dome and the 12" steelsound waveguide in a ~2 cu ft sealed box. The woofer crossover has one coil and a notch filter, the tweeter has a resistor and a 1uF cap. Crossover is around 1000Hz.
Have a couple of 1/3 octave measurement only. It is quite flat in the 400-20k band. It needs BSC below 400 and starts to roll off at 50 Hz. With an additional 2 cuft "backpack, it starts to roll off in the 30s.
Dispersion is well controlled on and off axle, vertical and horizontal.
BSC and Linkwitz transformation is done in the digital domain. With the LT it is good down to ~25Hz.
I did not hear any difference with changing the throat interface or using Geddes like foam plug ( and equalization), but I have a very different opinion on diffraction than Mr Geddes.

[JoshK]

Hi Frank,

Very cool! I didn't know you were working on such a project. I had given some thought to putting something similar together after visiting Dr. Geddes. I was thinking a 12" or 10" woofer and WG might be a bit better, especially when used with a seperate sub.

Curious, what is your opinion on diffraction? You tried the foam insert? Are these being used in your downstairs system?

[fcserei]

New Wave Guide Study

Josh,

Yes, I use 10 or 12 of them downstairs. I did not like the separate subs (2 * 18" BTW) rather went for the LT.
I've tied to file the vaveguide throat to differents diameter and profiles but I did not hear any difference when level ard FR matched.
Tried the foam plug with the necessary FR correction, but also did not hear difference, but I did not use compression drivers
My opinion on diffraction is that many are happening at the live event too ( musicans, instruments, chairs, scores, seats around you in the audience, the audience itself etc). They are not "disturbing" the perception, rather merge with the ambience.
In a correct reproduction chain additional diffraction and the possibly low level of the listening room ambience should merge with the ambient trail of the reproduction. This should cause no bigger difference than wether the next seat to you is empty or not on a live event.

[JoshK]

interesting... never thought about it like that.

[noah katz]

"My opinion on diffraction is that many are happening at the live event too ( musicans, instruments, chairs, scores, seats around you in the audience, the audience itself etc)."

Whatever diffraction happens around the instruments, performers, etc. id by definition what the live signal is.

Adding further diffraction effects in the playback chain will not faithfully reproduce the original event.

[fcserei]

Adding further diffraction effects in the playback chain will nit faithfully reproduce the original event.

So you are listening in an anechoic chamber

Othervise IMHO there is no such thing as the original event, and the live signal ( even a perfect 2 mic feed) is not a complete representation of the soundield.

[noah katz]

I don't follow your logic.

Your'e bascically saying that because everything isn't perfect, adding further response anomalies is doesn't hurt anything.

[noah katz]

oneoldude,

Would you be willing to measure one of the DDS waveguides if I supplied it?

It's polar response looks really nice

http://www.ddshorns.com/catalog.php?page=ENG190Pro

In contrast, I'd never guess from the off-axis response plots of the SP Tech that it used a waveguide

http://www.4sptech.com/version2/revelation-mr1_specifications.html

It doesn't get to -6dB @45 deg until 7 kHz - so much for directivity.

What tweeter(s) are you going to try?

You can email me at noah katz lmco com

Thanks

[Dennis H]

Okay, got a commercial in the Superbowl so I can reply.

Every so often I trot this graph out when diffraction comes up. I used Yavuz's line-array spreadsheet (tipped on its side if you will) to show the off-axis comb filtering of playing a mono signal through two speakers in an 8 ft. stereo triangle. Full scale on the graph is about +/- 1 ft. off axis so each division is about an inch. Things start to get nasty once you get more than an inch off axis yet the effects are fairly inaudible. Compared to this, a bit of cabinet diffraction is small potatoes.

Image not available

[oneoldude]

Compared to this, a bit of cabinet diffraction is small potatoes.

FWIW, I agree. I think SL agrees too.

[oneoldude]

In contrast, I'd never guess from the off-axis response plots of the SP Tech that it used a waveguide

http://www.4sptech.com/version2/revelation-mr1_specifications.html

It doesn't get to -6dB @45 deg until 7 kHz - so much for directivity.

I think the directivity issue for a box speaker like the SP Tech needs to be explained. I may not explain it well, but I will try.

In a box two-way, like the SP, the MW is fully omnidirectional at low frequencies. As frequencies rise, the MW becomes more and more directonal until it starts to beam like a flashlight when the wavelengths being reproduced are very short vs the MW diameter. Meanwhile, the T is very small in relation to the wavlengths it can reproduce at its bottom end. So at its low frequencies the T is trying to be omnidirectional.

Therefore, when you apply a crossover to the MW and T of say 2000 Hz, the MW is beaming and the T is broadly radiating. So the projected wave front of the MW is very different from that of the T. If nothing is done about this there will be a difference between the merged on-axis response of the two drivers and their off-axis response. Therefore, the on-axis and off-axis response may look ok, but the power response into the room will not be smooth across the crossover.

You might say, "So what?" Well, it has been shown that this type of "disconnection" between the radiating pattern of drivers and the resultant non-smooth power response is responsible for a great deal of what is called "speaker" sound from super high quality speakers. They can be spot on with frequency response, but still not sound as good as they otherwise could.

So, by using a Wave Guide you may be able to kill two birds with one stone. First, you might be able to make the T's radiation pattern match the radiation pattern of the MW. And second, you might be able to lower the crossover freq. to better integrate with the MW (like below breake up resonances) and still not overstress the T. Even if you accomplish both of these goals, the overall radiation pattern of the box speaker will not become highly directive like a horn used in a football stadium or pro-audio venue. It will still be omnidirectional at low frequencies and narrow as frequencies rise.

In looking at the plots presented by SP Tech, you cannot tell what is going on re: the WG. What needs to be seen is the polar plot at and around the crossover frequency for both drivers. If they are the same, then the Wave Guide has done its work and the crossover will merge well with both radiation patterns being the same at the crossover. The power response at crossover should be matched and the speaker should sound fine. Notwithstanding that you cannot tell if this is happening, the plots look pretty darn good to me for an MTM.

I might add that one of the reasons dipole speakers sound so good (among other reasons) is that by virtue of the dipole cancellation, the big fat omnidirectionality at low frequencies is reduced, power response is improved, and the polar pattern of radiation is more constant than in box designs. So, a Wave Guide would work very well there too (perhaps even better). The WG would have to be tailored to the M just like in a box speaker, but the radiation pattern target might well be different.

I will pm you re: the rest of your post.

Thanks

[cotdt]

what angle work best for waveguides? and shape?

[morbo]

Therefore, when you apply a crossover to the MW and T of say 2000 Hz, the MW is beaming and the T is broadly radiating. So the projected wave front of the MW is very different from that of the T. If nothing is done about this there will be a difference between the merged on-axis response of the two drivers and their off-axis response. Therefore, the on-axis and off-axis response may look ok, but the power response into the room will not be smooth across the crossover.

One question, how does a fullrange driver fit into this? I mean something like the 3-4.5" jordan, CSS, tangband, hivi etc. They don't have this discontinuity, and become more directive as frequency increases, but is their directivity desireable, or is there something about the power response of 'controlled directivity' 2-way designs that is preferable?

[oneoldude]

what angle work best for waveguides? and shape?

Er....that's what I am trying to find out!

[noah katz]

ab (just noticed that in your sig for the first time),

Yes, I understand all that re the benefits of the WG.

"In looking at the plots presented by SP Tech, you cannot tell what is going on re: the WG. What needs to be seen is the polar plot at and around the crossover frequency for both drivers...."

Compatible directivity at XO is certainly one of the benefits. Although no one ever seems to mention the vertical lobing from the MW+WGT being twice as tall as wide in the XO region.

Anyway, SP gives FR for 15, 30, and 45 deg off axis, which while not as comprehensive as full polar plots, gives enough info to conclude that it's a far cry from CD.

[oneoldude]

One question, how does a fullrange driver fit into this? I mean something like the 3-4.5" jordan, CSS, tangband, hivi etc. They don't have this discontinuity, and become more directive as frequency increases, but is their directivity desireable, or is there something about the power response of 'controlled directivity' 2-way designs that is preferable?

We are talking about apples, oranges, pineapples and potatoes here. I will again try to explain some difficult concepts that perhaps others would be better able to explain. So bear with me.

First. There is a fundamental power vs SPL problem with all loudspeakers. First let us presume we are in an anechoic chamber and all measurements are taken at say 1 meter. If I have a speaker that is beaming a high frequency on axis, I can point a mic at it on axis and get a reading, say 90 db. But if I go off axis, say behind it, I will read near 0. If I point a mic at all points around and over the speaker, I will get a bunch of different measurements from 0 db to 90 db. The average of all possible SPL at all points will be the total power put out by the speaker at that high frequency. But, if I change the frequency to a very low frequency where the speaker is omni polar and set the SPL to 90 db on axis, I can move the mic all around the speaker and I will read 90 db everywhere, including at the rear. Clearly, in the high freq beaming case I will have to average many points from 0db to 90db so the average power into the room will be below 90db. But in the low freq omni polar case I will have to average many points from 90db to 90db so the average power will always be 90db. Notice the results are different.

Now if I ring a triangle in the middle of the room it will send out sound in an omni directional pattern. Same if I ring a cow bell and the same if I beat a bass drum. So, depending on what instrument (or frequency) I am reproducing on the speaker, the power response and the on axis response will not be like real life. Hence, “speaker sound” rears its ugly head.

Second. No matter what driver you use, you will get increasing directivity and a shift in power response at some point with rising frequency. If you have a crossover to a much smaller driver you will have a discontinuity. If can you avoid it, better. But you still do not fix the first problem.

Third. If you have pinpoint directivity (like in electrostatics or some box speakers) you may have such a narrow sweet spot that you loose your sound stage if you move your head. So you do need dispersion to give you a good sweet spot. How much is the question that must be answered to tell you how much “controlled directivity” you want or need. In any event you want the directivity the same for all frequencies without change or discontinuity for best results.

So you see it is all a matter of compromise because nothing you can do will make it perfect.

Hey anybody got some measurements on those Wave Guides?

**For additional info, see this thread https://www.htguide.com/forum/showthread.php?t=13074

[oneoldude]

Noah,

Each and every design has it benefits and drawbacks. Speaker design is nothing more than a complex mass of compromises and often misguided beliefs.

While I am not an apologist for the SP Tech design, you must note the the crossover freq is a very low 600 Hz. That effectively shrinks the C to C spacing issue on the MTM which in any event is not related to width but to crossover frequency. And if you will note the vertical MTM notch in the response plots, you will see is quite nicely pushed back. That indicates good design. But I would rather have seen a sharper slope in the crossover to shrink that interaction even more.

I agree the SP Tech do not even approach CD. But who said they had too?

BTW, anybody got some WG measurements?

[oneoldude]

Noah,

Sent you email and it was kicked back undeliverable.

[noah katz]

ab,

"I agree the SP Tech do not even approach CD. But who said they had too?"

SP Tech claims that "Sound is radiated into the environment with a constant 90° angle of dispersion from 500 to beyond 10,000Hz. both vertically and horizontally."

http://www.4sptech.com/version2/dispersion.html

In my situation with asymmetrical L/R speaker environments, I'm banking on CD to improve things.

[oneoldude]

Noah,

As I said, I am not an apologist for SP Tech. Notice they do not describe how they come up with that 90 deg number. I think you used a -6db reference. Who knows what reference they used?

A read of the language and phrasing they use on their site will give you a good idea of what to expect though.

[noah katz]

ab,

"I am not an apologist for SP Tech"

Didn't mean to imply you were (although you do seem intent on defending them); my point, which perhaps I should have made more explicitly, is that I have a greater interest in waveguides that more truely approach CD behavior.

"Notice they do not describe how they come up with that 90 deg number."

Doesn't matter; their off-axis response curves enable extracting the relevant info.

[ergo]

I have a plan to go to Finland in 13-14 of February. I will try to go by the HiFi Talo shop in Helsinki. They have a DIY speaker kit using custom WG.

http://www.hifitalo.fi/shop/showsub.php?id=20

They sell it separate also so I will try to buy a few and post results later on.

Ergo

[oneoldude]

Thanks Ergo. :T

Waiting on ya.

Have a nice trip!

[elefantino]

Hi,

I have used the Finnish Chorus waveguides Ergo mentioned. Their flare seems rather similar to the flare of Genelec 1031 studio monitor waveguides. Unfortunately they weren't available seperately anymore last time I checked.

One guy on a German DIY forum ("Marvin") did some 3D modelling of these:

http://www.audioavid.com/wbb2/attachment.php?attachmentid=1270

http://www.audioavid.com/wbb2/attachment.php?attachmentid=1272

I started a thread about my experiments here:

http://www.hifi-forum.de/viewthread-104-4573-1.html

I have used these waveguides together with Seas 27TDC and modified Vifa D25AG35 tweeters. There are more measurements with different tweeters but the same waveguides here:

http://www.hifi-forum.de/viewthread-104-5884.html

Monacor sells a very cheap plastic horn for their DT-300 tweeters which can be modified (i.e. shorted) for waveguide use with other tweeters as well. You will also find measurements for these in the first thread above.

I hope you can make some sense of these threads despite the language. I'm sorry that I don't have the time to do an extensive translation.

Have fun

Peter

[AJINFLA]

Hi Peter,

I wish the Monacor was around a few years back when I was searching ! It looks much like what I had to fabricate. It probably would have made life a whole lot easier :banghead: .
For some reason my Google translation stopped halfway down the last page of your thread - the post with the D25AG35 waveguide, what size (diameter) is that? It looks rather compact. Where did you get it from?
I ceased development of my XT19 WG a while back, but I do have a pair of D25AG35's collecting dust that I might resurrect sometime.
Tell Christoph/Matthias (Thanner) thank you for me. The next time someone asks me for measurements I give them that link :W .

Cheers,

AJ

[elefantino]

Hi AJ,

I saw your project a while ago (on Madisound board?) and I could have bet your work was worth the trouble. Why did you cease development? Weren't you satisfied with the result?

All of my measurements (and Thanner's too) were done with the compact Finnish waveguides that are provided with Chorus speaker kits. You can find some images of these at http://www.chorus.fi . Personally I don't have any experience with the Monacor horns yet.

I think these shallow Finnish waveguides are optimised for directivity control. Lower frequency boost slowly decreases from 2k to lower frequencies. I would not use them for crossovers below 1.5k.

I forgot to give you the basic poperties of the waveguides:

mouth diameter 130 mm (plus 2*5 mm flange)
throat diameter 38 mm
depth 22 mm
transition from throat to mouth could be best described as nearly constant circle segment

Hope this helps.

I will tell Christoph and Matthias. Thanks!

Peter

[AJINFLA]

Hi Peter,

I assume you are talking about these (pic) on the Mad board a while ago. I was (am) very pleased with the sound :T . It is the best that my ears have heard in my own living room - and I've owned many hi end commercial designs. It was really my first serious attempt at an open baffle dynamic driver full range loudspeaker system.
But I am a hopeless DIYer at heart, so great is still never good enough :W .
Each speaker I build teaches me a little more (I hope). I had built plenty before with various configurations. The platform that the Orion uses and variants such as mine, is the one I will continue to use.
Note that Jon's designs here also are much the same platform (dipole bass/mids/monopole tweeter, etc.). That won't change.
I have however long believed that using a WG like I did is still not the optimal configuration. It brings one closer to emulating point source behaviour for reasons you surely know. But no matter how hard one manipulates this configuration (MT, MTM,etc.) there is still no better way to emulate point source behaviour than with a point source. Or for practical purposes, transducers with near point source behaviour. Such as coaxials or triaxials.
That will be the focus I return to as far as the 100/350hz~20k region is concerned. Directivity control of the HF using WG's will still be involved of course, just not in the fashion that I used with my Orion mutants .
The goals are much the same. I noted previously that Hifitalo also has a coaxial kit now, along with the WG's. It's no coincidence :B .
Keep up the great work. I'll get the Google translator to work somehow when I keep returning to your site. Otherwise I'll have to learn some german :W .

Cheers,

AJ

Image not available

[noah katz]

Peter,

Thansk for the links.

I'm puzzled to see that many of the graphs seem to show significant directivity down to 500 Hz, which is even more surprising after seeing that they're only 130 mm dia!

Any idea why that is so?

Oh, and how did they sound?

Thanks

[elefantino]

Hi AJ,

I always thought that a tweeter waveguide is just what the Orion design needs and your implementation is admirable. Unfortunately our rooms in an old building are too small for dipole designs. Otherwise I would certainly give them a try, too.

I'm also thinking of a design with controlled directivity and point source behaviour. But I'm not too happy with coaxial units like those by Seas or Thiel. Their moving 'waveguides' are likely to create IM distortion. On the other hand I like the ME Geithain studio monitor quasi-coaxials with their seperate tweeter baffles in front of the woofers. I currently consider how a waveguide solution could be integrated into a similar design. It will certainly need a big midwoofer which may compromise midrange resolution. Another idea may be an array of small midrange drivers that surround the WG tweeter.

Hi Noah,

I'm puzzled to see that many of the graphs seem to show significant directivity down to 500 Hz, which is even more surprising after seeing that they're only 130 mm dia!

Any idea why that is so?

This is an effect caused by small front baffles. I suppose that off-axis sound with long wavelengths gets 'lost in space' (not really as long as there are room boundaries, but reflected sound doesn't show up in gated measurements) rather than being directed to the mic as with a wide baffle. In addition there's a slight on-axis overshoot at the top of the 'baffle step'.

Oh, and how did they sound?

So far I have only listened to my prototypes with their mediocre Nokia midwoofers. I need to replace these with drivers with less distortion and better resolution. Peerless HDS Exclusive 180 or Dayton RS180 look like an obvious choice, my only gripe is their inefficency. With small studio monitors I'd like to avoid thermal compression as much as possible.

That said, I'm pleased with their sound so far. Their midtreble is uncolored, the middle and upper frequency regions are liquid and lively. Localisation is definitely above average, the soundstage they create is very well defined but not exceptionally wide and spacious. Changing back to more conventional hifi speakers, these tend to sound more colored, sometimes even shrieky. That's the way how it should be.

Cheers,
Peter

[AJINFLA]

But I'm not too happy with coaxial units like those by Seas or Thiel. Their moving 'waveguides' are likely to create IM distortion

I agree with you there, to a certain extent. Here is what Andrew Jones had to say http://www.diyaudio.com/forums/showt...651#post766651 . Read the whole thread if you have a chance. But the Seas/Thiel are not the types I was referring to, although the newest Thiel looks interesting http://thielaudio.com/THIEL_Site05/P...preliminfo.pdf .
More like this for coincident http://www.tadhomeaudio.com/model-1.html or http://www.pioneer.de/de/product_detail.jsp?product_id=11428&taxonomy_id=36 7-409 -very much a waveguide type approach!
Or this for coaxial http://www.audiophysic.de/kronos/index.html http://www.audiophysic.de/caldera/index.html
Except with tweeter directivity control and no boxes!

Cheers,

AJ

Image not available

[cotdt]

AJ I hope you realize that your speakers are missing a tweeter...

[elefantino]

Hi AJ,

Here is what Andrew Jones had to say http://www.diyaudio.com/forums/show...6651#post766651 . Read the whole thread if you have a chance.

Good link, thanks! Yet I'm still a die-hard 2-way devotee...

Cheers,
Peter

[AJINFLA]

AJ I hope you realize that your speakers are missing a tweeter...

Whew...found it! Had me going there for a second Cotdt... :spin:

Image not available

Andrew's old handiwork.

Cheers,

AJ

[cotdt]

omg... is that what I think it is? How did you do that? pretty amazing...

[Patrick Bateman]

If you look at the curve on the SP Tech, it does NOT appear to be using the same curve as the Summa. An oblate spheroidal waveguide has an extremely gentle transition from throat to mouth (that's the point of it actually.)

The SP Tech has an abrupt transition about 1/3rd of the way down the throat.

Also, while it's good to pay attention to the curve of the waveguide, it's also important to consider the cabinet. Note the roundover on all of the cabinet edges to reduce diffraction.

It's interesting to note that JBLs monitors now use all of these features, from an oblate spheroidal waveguide to massive cabinet roundovers.

[Dennis H]

Since this thread got resurrected, there's also a thread going over at diyaudio that has a couple of links to new waveguide projects/experiments.

http://www.diyaudio.com/forums/showthread.php?s=&threadid=85910

http://www.aeronet.com.au/waveguide.htm

http://www.troelsgravesen.dk/C17_II.htm

[JoshK]

I just purchased a pair of DDS Eng 1-90's. With those I plan to use a compression driver, such as the B&C DE250 or the BMS 4552ND (haven't decided which yet).

I am still planning to use the MCM waveguides with a dome tweeter for my version of the Arvo Part.

[JoshK]

Btw, I am pretty sure JBL's new speakers came about as a result of work with Dr. Geddes. When I visited DrG, he mentioned he was in discussions with JBL.

[Dennis H]

As Noah noted over in the diyaudio thread, the Eighteen Sound waveguides look pretty sweet. He says Dr. Geddes gave a thumbs up to their polar pattern. And they use a flat mounting flange so no special machining or adapers are needed to use a conventional tweeter or a better-grade compression driver which tend to be bolt mounted rather than screw-in. Cheap too.

http://www.loudspeakersplus.com/html/18sound_horns.html

[JoshK]

I ordered a pair of the DDS Eng 1-90's from Assistance Audio recently (responded to emails btw) and was told it came in screw on or bolt on. I chose the bolt on because the BMS 4552ND and the B&C DE250/500 are all bolt ons. This will also lend itself well to playing with dome tweeters I'd imagine. So I guess we have a few choices now between the DDS, 18Sounds, MCM, etc.

[JoshK]

Btw, anyone have any suggestions for which compression driver to use? I am clueless about these. The B&C ones look to have much flatter FR, but I have no idea if the horns they are measured on are compairable to the ones the BMS are measured on.

[ThomasW]

I am still planning to use the MCM waveguides with a dome tweeter for my version of the Arvo Part.

They're on sale in the newest flyer