tima's DIY RCM

[Neil.Antin]

Tim @tima (and Neil @Neil.Antin ), would you still recommend running the P120H at 70% power? Is that enough or too much?

David,

Here is a thought if the machine will allow:

-75% power at 35kHz. This effectively becomes the pre-clean step. The lower kHz is good for thicker soil films and larger particulate.

-100% power at 80 kHz. This effectively becomes the final clean step. At the higher kHz, it takes more power to develop cavitation and more power to get the best possible from the smaller (with less cavitation energy) bubble. But note that while at 80 kHz the cavitation bubble is smaller - there are more of them. Using 100% power then maximize the benefits of the smaller (gentler) 80 kHz bubble to get into the groove and release the really small particles and very thin films.

Good Luck,
Neil

 

[tima]

I use the P60H, mainly because my Wife does the food shopping and she can pick up a single 1 or 2.5 Gallon Jug of distilled water on a trip. I only clean 2 records at a time, works well for me. I haven't compared the 120H vs. the P60H.

Nice job!

 

[tima]

I'm considering getting a 6L Elmasonic P60H unit for the main cleaning and leaving the existing unit for rinsing. The rinsing really kills me.

My initial design and goal were to clean without a rinse step under the notion that keeping the wash tank water very clean obviates the need to rinse and no rinse step saves time. And I got good results with that approach - better than with my AudioDesk. But like a lot of things in audio one may think they're doing well until they try something else. Neil and I had a discussion (teacher/student) over at The Vinyl Press (following publication of the original edition of his 'book') about non-volatile residue (NVR) remaining on a washed record. (If you look, try starting around May 24, 2020.) That got me thinking about giving up my original approach and seriously considering a rinse step. David had the same idea.

I'm convinced now that rinsing perfects the wash. The difference is not dramatic but it is consistent and audible. Based on the Elma approach the rinse step added roughly 2/3 the cost of the wash tank setup while using my previous version's pump and filter for the rinse tank.

As I said somewhere, to make an additional step more tolerable one of the keys to rinsing is the ease of taking a group of records out of the wash and putting them in the rinse. This is one way the relatively expensive Kuzma RD rotisserie earns its keep. With 4-5 records mounted on a spindle you can go from wash to rinse to dry simply by taking the spindle to the next station - and do it with ease and safety. The RD comes with a great stand to hold the spindle straight up for mounting and at a 45° angle for drying.

Jarek, a DIY rotisseire alternative (to the Kuzma RD) built for the P60H is shown in @advanced101 's post #378 (above). From his picture it looks like it would be easy to move the spindle of two records to a rinse tank with a second DIY rotisserie. Rinse to daylight!

 

[Stacore]

@Neil.Antin Thanks! Pump was not so difficult to find unlike the abs-rated filters! Nothing locally :O Fortunately the UK company you found has 5" filters (we have a 5" filter housing).

@tima Thank you for the info! I will pass on Elmasonic - for sure a great machine but cannot justify the cost at the moment. Will prob go for that AudioRevita as it is (almost) plug n play and local and time is sth I've been missing lately. BTW what is your rinsing procedure? What time, freq and power?

 

[Stacore]

I've checked what I have available locally from AudioRevita: 10L/240W and 15L/360W both 40kHz and both have 24W/L, same as our current one (6L/150W). I'm leaning towards 10L/240W as it is smaller, unless 15L/360W offers some leap forward.

From your experience, what would be the optimal usage of 6L/150W and 10L/240W? From what I understand the latter offers more cavitation power as the volume is bigger at the same wattage per liter (pls correct me Neil if I'm wrong). This would suggest:

10L/240W, 5 LP's, (or 15L/360W, 7LPs) for cleaning
6L/150W for rinsing

Comments welcome
Thanks

 

[Neil.Antin]

Pump was not so difficult to find unlike the abs-rated filters! Nothing locally :O Fortunately the UK company you found has 5" filters (we have a 5" filter housing).

Jatek,

You are lucky,. I missed that you were using a 5" filter. This 5" abs rated filter is pure coincidence - its the first I have ever seen - Spectrum PPP Spectrum Premier Pleat Sediment Filters 47/8in Open Ended AAS From £24.00 - Spectrum Water Filters by Wrekin Water Filtration. It will be fine with the new pump - it has more than enough pressure-head to accommodate the filter higher pressure drop.

WRT to the AudioRevita - manual_arc_02.cdr (home.pl), couple of items to note:

-If using alcohol, my book Chapter VIII has a detailed discussion of alcohol and its flammability & explosion risk, and the Danger on page 72.

-The spin speed of 2.5 rpm is fast. Assuming this is a 6L tank, the max spin for cleaning 1 record is 3 rpm, 2 records is 1.5 rpm and 3-records is 0.7 rpm. The book Section XIV.5 provides the details. If the spinner is a 12VDC motor - a 3-12VDC variable power supply can be used to slow down the motor - 3-12V Adjustable Power Supply with USB Port Multifunction Charger Portable 12W Voltage Regulator Switch Power Adapter : Amazon.pl.

Good Luck,
Neil

 

[Stacore]

Jatek,

You are lucky,. I missed that you were using a 5" filter. This 5" abs rated filter is pure coincidence - its the first I have ever seen - Spectrum PPP Spectrum Premier Pleat Sediment Filters 47/8in Open Ended AAS From £24.00 - Spectrum Water Filters by Wrekin Water Filtration. It will be fine with the new pump - it has more than enough pressure-head to accommodate the filter higher pressure drop.

WRT to the AudioRevita - manual_arc_02.cdr (home.pl), couple of items to note:

-If using alcohol, my book Chapter VIII has a detailed discussion of alcohol and its flammability & explosion risk, and the Danger on page 72.

-The spin speed of 2.5 rpm is fast. Assuming this is a 6L tank, the max spin for cleaning 1 record is 3 rpm, 2 records is 1.5 rpm and 3-records is 0.7 rpm. The book Section XIV.5 provides the details. If the spinner is a 12VDC motor - a 3-12VDC variable power supply can be used to slow down the motor - 3-12V Adjustable Power Supply with USB Port Multifunction Charger Portable 12W Voltage Regulator Switch Power Adapter : Amazon.pl.

Good Luck,
Neil

Hey Neil,

I change the motor to 1rpm anyway and make my own cleaning solution according to your suggestions, so this is out of the picture.

What would you suggest: 6L/150W for rinsing and 10L/240W for cleaning or vice versa?

Thanks

Edit: You didnt miss anything, the 5" were not there, I asked them if they had them

 

[Neil.Antin]

I've checked what I have available locally from AudioRevita: 10L/240W and 15L/360W both 40kHz and both have 24W/L, same as our current one (6L/150W). I'm leaning towards 10L/240W as it is smaller, unless 15L/360W offers some leap forward.

From your experience, what would be the optimal usage of 6L/150W and 10L/240W? From what I understand the latter offers more cavitation power as the volume is bigger at the same wattage per liter (pls correct me Neil if I'm wrong). This would suggest:

10L/240W, 5 LP's, (or 15L/360W, 7LPs) for cleaning
6L/150W for rinsing

Comments welcome
Thanks

Jarek:

If you want to clean more than 2 records at time, the 10L/240W would be a good choice. At 10L to clean 5-records at once, you would need to slow the spinner down to only 1-rpm. I suspect that they are using a single design transducer and then increasing the number as the tank size increases, so they install three 60W for the 6L (180W), four 60W for the 10L (240W) and six for 15L (360W). Also, these power ratings can be wishful. If the transducers are not mounted optimally, some power is lost, and I am not familiar with whether power output is steady with electronics/transducer temperature.

FYI - For the 10L/240W unit, the UT transducers should only heat the water max 0.6F/min (assume ideal and most of the power is transferred to heat so the 10L = 22-lbm water, and the 240W = 13.7 BTU/min, heat capacity of water 1-BTU will increase 1-lbm water 1F, then 13.7-BTU/22-lbm = 0.6).

Using the 6L for rinse is OK if spinning too fast. You will loose cavitation intensity but you are mostly rinsing to remove any cleaner residue. Its a cost/performance compromise.

The one item you should consider is noise & reliability. These lower cost UT tanks as you probably know can be very noisy and contrary to the advertising are not designed for heavy use. If they were there would not be warnings to not use >30 min - does it have a temperature warning, some do. Units intended for continuous operation often have cooling fans and much more robust construction (with commensurate $$$).

Before any decision to buy a new UT, why not make the low cost changes first that can be used for any UT tank - so there is no money lost:
-Change the chemistry to as recommended.
-Install improved filter/pump
-Monitor TDS appropriately
-Clean just 1 record - how good is it, clean 2 records at a time - how good are they.

Then put together a list of objectives/requirements you want such as:
-What is the initial cleanliness of the record(s) you are cleaning. The Elmasonic P-series has two operating frequencies (35kHz & 80kHz) with variable power for each. Used effectively, it should be able to avoid vacuum-RCM (or similar manual wash) for all except the dirtiest of records.

-How many records to clean at-once. This will dictate the cleaning UT size - volume.

-How many records to clean during a single session. If you have to secure every 30-min for 30-60 min for the unit to cool, you may not be happy. This can make the decision to wait, and save-up for a better unit.

-Desired process time for combined clean & rinse. How easy is it to move a record from the cleaning UT to the rinsing UT.

-What is your access to DIW,. If you have lots, the rinse tank can be refreshed frequently. Will both the clean & rinse tank be filtered - if so, each needs its own filter/pump. You could use a 0.5 micron for the rinse tank and re-purpose your existing pump.

I have to echo what Tim (@tima) has said a few times that if the process becomes cumbersome, then you will not use it as frequently as you should.

Good Luck,
Neil

 

[Stacore]

Jarek:

If you want to clean more than 2 records at time, the 10L/240W would be a good choice. At 10L to clean 5-records at once, you would need to slow the spinner down to only 1-rpm.

This is exactly my plan - to install a 1rpm motor and clean 5 LP's at a time. What power and time would you suggest for 5LP's, 240W max power @ 10L, 1rpm ?

Using the 6L for rinse is OK if spinning too fast. You will loose cavitation intensity but you are mostly rinsing to remove any cleaner residue. Its a cost/performance compromise.

I think you meant "not too fast" ? I plan to use the same 1rpm motor for rinsing. How long should the rinsing take? 5min@1rpm @ 100% power (150W/6L)?

Still thinking if the 10L/240W machine should go for cleaning or rinsing or it doesn't matter? From what you say, a bigger tank offers more stable cavitation action.

The one item you should consider is noise & reliability. These lower cost UT tanks as you probably know can be very noisy and contrary to the advertising are not designed for heavy use. If they were there would not be warnings to not use >30 min - does it have a temperature warning, some do. Units intended for continuous operation often have cooling fans and much more robust construction (with commensurate $$$).

I'm aware of that but since I do a vacuum pre-clean, there will be a time for the machines to rest.

Then put together a list of objectives/requirements you want such as:
-What is the initial cleanliness of the record(s) you are cleaning. The Elmasonic P-series has two operating frequencies (35kHz & 80kHz) with variable power for each. Used effectively, it should be able to avoid vacuum-RCM (or similar manual wash) for all except the dirtiest of records.

All over the place: from nice n clean to "omg" Recently brought to life the original 1953 recording of Tosca with Callas at La Scala. Was very very dirty.

Pre-clean is actually pretty fast and removes all hair, fingerprints, fluff. I want to keep it.

-How many records to clean at-once. This will dictate the cleaning UT size - volume.

-How many records to clean during a single session. If you have to secure every 30-min for 30-60 min for the unit to cool, you may not be happy. This can make the decision to wait, and save-up for a better unit.

I still have about 1000 LP's to clean. Your tips (dilute the solution, 10mins at max power instead of 20 on 75%) has already cut the time considerably. Rinse UT should make it even faster even if 10mins needed to cool the machine down (30min needed after operating for 30mins, I do 10mins cycles).

-What is your access to DIW,. If you have lots, the rinse tank can be refreshed frequently. Will both the clean & rinse tank be filtered - if so, each needs its own filter/pump. You could use a 0.5 micron for the rinse tank and re-purpose your existing pump.

No problem with DI, but I'll filter both cleaning tank (with the 0.2u absolute filter you found in UK) and rinse tank (with the 0.5u nominal I have).

 

[Neil.Antin]

This is exactly my plan - to install a 1rpm motor and clean 5 LP's at a time. What power and time would you suggest for 5LP's, 240W max power @ 10L, 1rpm ?

As written in my book.

"XIV.10 Step 2 - Final Clean: ...Exposure time should be similar to the pre-clean-UCM so that no part of the vinyl record is in contact with the final-cleaner for greater than 5 minutes. Assuming an exposure time of 2-3 minutes, with only 33% of the record exposed at any time, a total rotation time of about 6-9 minutes would be recommended subject to record rotation previously discussed paragraph XIV.5 " At 1-rpm, 0.33 of the record is exposed/min, so 5-min accumulated exposure is 5-min/0.33 = 15 min on the high side.

As far as power, I make no recommendation and allow the available hardware at your acceptable price to dictate. I only provide guidance to maximize what your have unless detailed technical data such as that provided by Elmasonic or equal or such as the data published for the pump/filter are available.

The time for the rinse should be 6-9 minutes with no need to extend to the high side of 15 min.

I think you meant "not too fast" ? I plan to use the same 1rpm motor for rinsing. How long should the rinsing take? 5min@1rpm @ 100% power (150W/6L)?

Per my book XIV.5.3 five-records in a 6L tank should not spin any faster than 0.5 rpm. At 1 rpm, cavitation intensity will be deceased and can be as reduced by as much as 25-50%. But this is only for rinsing, so you have a comprise.

Still thinking if the 10L/240W machine should go for cleaning or rinsing or it doesn't matter? From what you say, a bigger tank offers more stable cavitation action.

Yes, use the 10L for cleaning. Item to consider is tank width and record spacing. The 10L tank 'should' be wide enough to space the records far enough apart to get good cavitation in-between. But that spacing 'may' be too wide for the 6L tank. What is the optimum spacing - good question. There is currently no good detail other than anecdotal/subjective observation. But the spacing is most important for cleaning - you want to maximize cavitation intensity. There are number of different variables in-play, and 'some' are addressed Section XIV.3, and I have been contemplating others. I may try to coalesce some kind of guidance similar to what I did for record rotation this winter. But knowing that XIV.5.3 overwhelms some people, record spacing could make that pale in comparison. For now, space as best you can.

Take care,
Neil

 

[tima]

@tima ... BTW what is your rinsing procedure? What time, freq and power?

Rinse tank: I use a Kuzma RD rotisserie and Elmasonic S120H with the Little Giant 1-AA-OM pump and polypropylene 0.35 micron (nominal) Flow Max FM-0.35-975 filter in a Pentek 158116 ¼” #10 Slim Line housing.

At the start of a cleaning session I degas the rinse water (distilled) for ~10 minutes.
As noted earlier, I pick up a spindle of records from the wash tank RD and place it in rinse tank RD.

The S120H operates at a single frequency of 37Hz. I use Power: 80-90%. Time: 10 minutes. The pump/filter operates during the rinse cycle. W/out using the heater I let the water temperature rise as a product of the ultrasonic action; by the end of 10 minutes it's typically 30° C I'm still experimenting with rinsing.

 

[dminches]

Rinse tank: I use a Kuzma RD rotisserie and Elmasonic S120H with the Little Giant 1-AA-OM pump and polypropylene 0.35 micron (nominal) Flow Max FM-0.35-975 filter in a Pentek 158116 ¼” #10 Slim Line housing.

At the start of a cleaning session I degas the rinse water (distilled) for ~10 minutes.
As noted earlier, I pick up a spindle of records from the wash tank RD and place it in rinse tank RD.

The S120H operates at a single frequency of 37Hz. I use Power: 80-90%. Time: 10 minutes. The pump/filter operates during the rinse cycle. W/out using the heater I let the water temperature rise as a product of the ultrasonic action; by the end of 10 minutes it's typically 30° C I'm still experimenting with rinsing.

How does one run the S120H under 100% power? I didn’t realize it had that capability.

 

[tima]

How does one run the S120H under 100% power? I didn’t realize it had that capability.

Clearly one does not. My error. Thanks for catching it.

 

[dminches]

Clearly one does not. My error. Thanks for catching it.

Whew! I thought I was missing something.

I am going to take Neil's advice and trying running the 10 minute clean at 80 Hz with full power.

 

[Neil.Antin]

@tima & @dminches,

Here are some other items to consider that may further improve cleaning.

Fundamentally, in the tank there are layers of peak energy with peak cavitation intensity that for your bottom firing UT transducers should be in horizontal layers Ultrasonic Energy Distribution - Zenith Ultrasonics (zenith-ultrasonics.com). Ideally, the peaks will be spaced at ~1/2 of the wavelength lamda Microsoft PowerPoint - 1200_L_22_VWS_3.pptx (uiowa.edu). For DIW with a speed of sound of 1498 m/s, and the 80 kHz, the peaks will be spaced at (0.5)(1,498,000 mm/80,000 Hz) = ~19 mm.

The Elmasonic P-series has a sweep function that is intended to shift the operating frequency +/- some kHz to broaden the zone of peak UT energy (ergo-cavitation intensity); but it may reduce the absolute peak. They (and others) do this since the normal use of UT cleaning is with parts that are static in the tank. So the sweep function is intended to even-out the cavitation intensity in the tank. The value of sweep is debated in the industry.

BUT, cleaning records rotates the record and this means that the record is constantly passing through these peak-layers. As it the record passes from lower to higher to lower zones of cavitation intensity, a scrubbing type action should occur. BUT, the Elmasonic also has a Pulse-mode - that increases the UT energy by 20% which means the peak cavitation intensity should be higher.

So, the question that comes to mind, which is best? No-Sweep/Pulse, With-Sweep or With Pulse. Since you are not running full power at 37-kHz, this is probably more applicable to the 80-kHz mode. Intuitively for a record and how its being rotated, w/o sweep should yield better cleaning because the peak cavitation intensity zones should be higher. So the question may be down to with or without Pulse.

Not to make this too complicated, but record spacing may be factor, but with you cleaning only 6-records at time with good spacing between, I would think unlikely. But closer spacing is uncertain.

For you who have the best UT systems, there may still be room for improvements in the quest for best achievable.

Good Luck,
Neil

 

[tima]

@tima & @dminches,

Here are some other items to consider that may further improve cleaning.

Fundamentally, in the tank there are layers of peak energy with peak cavitation intensity that for your bottom firing UT transducers should be in horizontal layers Ultrasonic Energy Distribution - Zenith Ultrasonics (zenith-ultrasonics.com). Ideally, the peaks will be spaced at ~1/2 of the wavelength lamda Microsoft PowerPoint - 1200_L_22_VWS_3.pptx (uiowa.edu). For DIW with a speed of sound of 1498 m/s, and the 80 kHz, the peaks will be spaced at (0.5)(1,498,000 mm/80,000 Hz) = ~19 mm.

The Elmasonic P-series has a sweep function that is intended to shift the operating frequency +/- some kHz to broaden the zone of peak UT energy (ergo-cavitation intensity); but it may reduce the absolute peak. They (and others) do this since the normal use of UT cleaning is with parts that are static in the tank. So the sweep function is intended to even-out the cavitation intensity in the tank. The value of sweep is debated in the industry.

BUT, cleaning records rotates the record and this means that the record is constantly passing through these peak-layers. As it the record passes from lower to higher to lower zones of cavitation intensity, a scrubbing type action should occur. BUT, the Elmasonic also has a Pulse-mode - that increases the UT energy by 20% which means the peak cavitation intensity should be higher.

So, the question that comes to mind, which is best? No-Sweep/Pulse, With-Sweep or With Pulse. Since you are not running full power at 37-kHz, this is probably more applicable to the 80-kHz mode. Intuitively for a record and how its being rotated, w/o sweep should yield better cleaning because the peak cavitation intensity zones should be higher. So the question may be down to with or without Pulse.

Not to make this too complicated, but record spacing may be factor, but with you cleaning only 6-records at time with good spacing between, I would think unlikely. But closer spacing is uncertain.

For you who have the best UT systems, there may still be room for improvements in the quest for best achievable.

Good Luck,
Neil

Yes, higher frequency = smaller + more bubbles = greater and more even distribution than lower frequency but less force. I was not aware there is a banding effect (~ ¾ inch for 80kHz)), but as you imply moving the record through these bands should mitigate the result a fixed object would see. I'd be interested to learn how the banding effect lays out for horizontal firing transducers.

I have been using neither the Sweep nor the Pulse functions on a regular basis though I tried them out. I'll try the Pulse when I clean another batch. I did wonder what effect the +/- frequency variations might have on cavitator lifespan.

When you write "Intuitively for a record and how its being rotated, w/o sweep should yield better cleaning because the peak cavitation intensity zones should be higher." Are you saying consistently higher in frequency because there is no minus reduction, or ...?

For the while I'm doing 4-5 records on a spindle with two or three pucks between each to increase the space between them.

Thanks Neil - your contributions are always helpful !

 

[Neil.Antin]

Yes, higher frequency = smaller + more bubbles = greater and more even distribution than lower frequency but less force. I was not aware there is a banding effect (~ ¾ inch for 80kHz)), but as you imply moving the record through these bands should mitigate the result a fixed object would see. I'd be interested to learn how the banding effect lays out for horizontal firing transducers.

I have been using neither the Sweep nor the Pulse functions on a regular basis though I tried them out. I'll try the Pulse when I clean another batch. I did wonder what effect the +/- frequency variations might have on cavitator lifespan.

When you write "Intuitively for a record and how its being rotated, w/o sweep should yield better cleaning because the peak cavitation intensity zones should be higher." Are you saying consistently higher in frequency because there is no minus reduction, or ...?

For the while I'm doing 4-5 records on a spindle with two or three pucks between each to increase the space between them.

Thanks Neil - your contributions are always helpful !

Tim,

The sweep-function shifts the UT frequency say 2-kHz, so its primary operation is 37-kHz but then periodically sweeps to 35-kHz and 39-kHz. When they do this the average power may increase, but the peak power may decrease. Zenith has some thoughts on this - Sweep Frequency Ultrasonic Cleaners - Zenith Ultrasonics (zenith-ultrasonics.com). Zenith simultaneously produces 40-kHz and 80-kHz with many transducers - the effect being a very uniform cavitation intensity in the tank; and for cleaning of static parts is great. However, the more I think about this, the layering that occurs may be a significant advantage unique for cleaning records.

Companies have been using sweep-frequency for many years with no report of lifespan issues that I am aware of. However, good industrial units have active (fans) cooling systems for the transducers & electronics. The Degritter which is a powerful compact unit has fans and monitors temperature and will secure the transducers and run the fans to cool the unit off. The Elmasonic P/S-series makes no mention of operational time limits.

What about horizontal mounted transducers such as the Degritter. Good question. I suspect that vertical standing waves (or variations thereof) will be produced. Degritter says they they do use a sweep. The 1/2-lamda for 120-kHz is 6 mm, and varying record thickness from say 120g to 200g 'may?" have an effect. But with sweep, it would broaden the peak zone and with good engineering place the record in a near optimal position regardless of width. It will be interesting to see how the HumminGuru all-in-one ultrasonic vinyl record cleaner fairs. Its using 40-kHz and a very small fluid volume - 350 mL; the water is going to heat very quickly if it has any significant power which is not specified. And, because the water volume is so small, they may have to spin really slow (<<1 rpm) to avoid moving so much fluid as to cancel much of the cavitation intensity - but to many people who do only a simple brush cleaning, it will likely be a revelation.

 

[dminches]

One of the downsides to full power on the 80 Hz cycle is that it heats the water up much more quickly. I tried that today and will test the results later but I did notice that by the end of 20 minutes (10 @ 37 Hz @ 70% and 10 @ 80 Hz @ 100%) the temperature was 35 degrees C instead of 33 or 34. That in itself isn't a issue for 1 spindle of LPs but it does prevent me from doing a 2nd. Without any way to actively cool the liquid I am limited to cleaning 1 spindle at a time.

 

[Neil.Antin]

One of the downsides to full power on the 80 Hz cycle is that it heats the water up much more quickly. I tried that today and will test the results later but I did notice that by the end of 20 minutes (10 @ 37 Hz @ 70% and 10 @ 80 Hz @ 100%) the temperature was 35 degrees C instead of 33 or 34. That in itself isn't a issue for 1 spindle of LPs but it does prevent me from doing a 2nd. Without any way to actively cool the liquid I am limited to cleaning 1 spindle at a time.

There are a number of active cooling solutions. The easiest is to install a small radiator with fan in the outlet tubing between the filter and the tank. This way the filter keeps the radiator internal clean so the radiator does not clog and then the radiator always sees the lowest pressure. The Shurflo pump has more than enough pump head to handle any pressure drop caused by a radiator.

A very cheap option is an aluminum radiator which appears good to 120F (DI Water Compatibility Chart – Greg Reyneke (gregknowswater.com) with DIW - I want to double check, but copper is a no-no. I also need to double check the pressure rating - most are good for only 2-bar (~30 psi) but located after the filter should be fine. Here is an example: Amazon.com: DIYhz Water Cooling Computer Radiator, 12 Pipe Aluminum Heat Exchanger Liquid Cooling Radiator Heat Sink 120mm with Fan for CPU PC Laser Water Cool System DC12V Black : Electronics.

A much more expensive option using a high pressure stainless steel radiator is such as this AS04-05G01 Stainless Steel Tube-Fin Heat Exchanger | Boyd Direct (boydcorpdirect.com). This is the best you can buy. I have not searched for cheaper, and an aluminum radiator is likely fine.

The concept is that when the pump/filter is operated between cleanings, it also cools the fluid enough for the next batch. Since the pump may only be operated for say 10-min between batches, the radiator has to be sized accordingly.

If you can tell me the bath temperature at the beginning of the 100%-80kHz and then after the 10-min cycle I can estimate the power going into fluid. Its not difficult. The tank is ~12.5L = ~27.5-lbs of water. It takes 1-BTU of energy to raise the temperature of water 1F. So, if the bath water for example increases 5F, (5)(27.5) = 137.5 BTU/10min = 13.7 BTU/min = 237W. To this I add the pump-motor power (the pump adds heat to the fluid) to get the final heat input and then select the radiator accordingly.

But active cooling is only if the higher power provided cleaning benefits.

Neil

 

[dminches]

But active cooling is only if the higher power provided cleaning benefits.

Neil

Neil, thanks for the detailed response. I had tried a 3 fan computer radiator before but I found that the higher readings on the TDS meter indicated there could be some interaction between the cleaning solution and the radiator.

This is the radiator I used: https://tinyurl.com/89npepuc

It is copper so that may be the issue.

If I can just drop a radiator between the pump and filter and the pump will keep things flowing I am open to trying one of your solutions. Even with the 70% power the temps were too high to do multiple cleanings back to back.