The Cycle Satiator, universal charger for the enthusiasts

[BVH]

Justin, I am thinking in the same realm on how much more Ah or mAh will go in btw the two diff terminations. That's why it seems to me, anyway, such a no-brainer to terminate at 3.5 or even a little less on LiFeP04 to obtain as much cycle life as possible.

That's good to know about Coulometric efficiency on Lithium cells. How about Peukerts' Factor? Is it as much of a capacity effector on LiFeP04 as with Lead, ect?

 

[Warren]

Justin,

"Here's the windows version of the suite:
http://www.ebikes.ca/downloads/Satiator_Suite_Windows_v0.95.exe"

I have tried several times to open the new file linked above. I put it directly into C: drive on my old IBM, running Windows XP. Norton hates it and wants to remove it. When I attempt to open it anyway, Windows says it is not a valid Win32 application. :-(

Warren

 

[amberwolf]

It's a self-extracting Winrar file, and it does execute here on WnXP, with no 3rd party security packages installed.

However, it's default extraction folder is
C:\Program Files\Mozilla Firefox
even though the folder it's actually in is
C:\md\Downloads
if i run it directly from the Firefox download window.

If I run it from the folder it is actually in,
C:\md\Downloads
then it's default extraction folder is
C:\md\Downloads

Perhaps when creating the Winrar file, it would be a good idea to give it a specific target folder name at root directory level of C: or the %foldertype% stuff to stick it on the desktop or MyDocuments, so that when Grin Tech support is helping people with any problems, they'll know for sure where it is putting the files, and also so that any documentation on the web can tell people exactly where the files will be. (even though they will still probably have to have the person "find" the downloaded file itself in some cases, then at least if that person finds and runs it thru their browser instead of explorer, and doesn't pay attention to following dialogs as is common, it'll still be easy to help them find the stuff for the next stage).

 

[amberwolf]

How about Peukerts' Factor? Is it as much of a capacity effector on LiFeP04 as with Lead, ect?

No, lead-acid is MUCH worse about this than any other chemistry I've ever used. It still makes some difference in others, but not that bad.

 

[BVH]

Thanks for that info.

Got my Satiator Saturday. Just a couple questions:

1. When using the profile editor on my PC, I had thought that the communication line would remain open during the processes of read-in, editing and writing but it appears that once you read-in the profiles, the Comm window must be closed and Comm terminated before editing can take place and one must re-enter Comm mode to write the edited profiles back to the charger, correct? This is not an issue, I just want to be sure it is functioning normally.

2. There is one mention in the manual about "hiding" profiles. I don't see any way to hide them in the unit nor in the profile editor. I have only one use for my charger currently and would like to hide all the standard profiles and show only the 3 custom profiles I created. I know I can delete standard profiles but is there a way I'm not seeing to "hide" them?

3. Is the field "Complete Amps" the current rate at which termination occurs in the CV phase or is it the current flow after termination in the Lithium chemistry profile? I think its the former but I thought I read somewhere (I can't find it now) it was the latter in what was maybe an earlier version of the manual? If it's current rate at termination point, it seems like a low default value in my case - .12 Amps for a 12 ah pack. In my other hobby chargers, I'm used to seeing C/10 (normal charge) with an option of C/20 (accurate charge).

The charger is about 10-15% smaller than I had guessed and thank you for providing a nice, long USB cord! I like the ability to set the output line resistance. I have a new Agilent 6.5 digit bench DMM with 4-wire Kelvin capability for reading Ohms and I have the 4-wire test probes. So if I want to extend my output leads, which I will probably do, I can precisely set my new leads resistance figure. This is a nice feature.

 

[justin_le]

1. When using the profile editor on my PC, I had thought that the communication line would remain open during the processes of read-in, editing and writing but it appears that once you read-in the profiles, the Comm window must be closed and Comm terminated before editing can take place and one must re-enter Comm mode to write the edited profiles back to the charger, correct? This is not an issue, I just want to be sure it is functioning normally.

Yes, this is correct. At the moment anything that that involves the bootloader reading or writing is ending with a device reset which restarts the Satiator. This behavior could and probably should be changed, so that you have an option to restart the Satiator or stay in bootloader, so I'll see that we change that in the next software update.

2. There is one mention in the manual about "hiding" profiles. I don't see any way to hide them in the unit nor in the profile editor. I have only one use for my charger currently and would like to hide all the standard profiles and show only the 3 custom profiles I created. I know I can delete standard profiles but is there a way I'm not seeing to "hide" them?

If the profile is not active, then it is effectively "hidden". So only the profiles that have the "active?" radio box checked are going to show up on your profile selection list when the device is turned on.

3. Is the field "Complete Amps" the current rate at which termination occurs in the CV phase or is it the current flow after termination in the Lithium chemistry profile? I think its the former but I thought I read somewhere (I can't find it now) it was the latter in what was maybe an earlier version of the manual?

From the manual:

If it's current rate at termination point, it seems like a low default value in my case - .12 Amps for a 12 ah pack. In my other hobby chargers, I'm used to seeing C/10 (normal charge) with an option of C/20 (accurate charge).

With lithium profiles this is just the threshold current at which point the text across the satiator says "charge complete". It doesn't have any actual effect on the actual charge algorithm. The charger goes into CV mode and once it does that the current decays, and at some point in the process you can choose when it indicates the charge is finished.

You are right that the default threhold is on the low side for sure, so even with a fast bulk charge it can still seem like the satiator is taking a long time to charge a pack up, especially packs with a high internal resistance. A default of more like 0.3-0.4A would be more appropriate and would give the appearance that the charge is faster.

The charger is about 10-15% smaller than I had guessed and thank you for providing a nice, long USB cord! I like the ability to set the output line resistance. I have a new Agilent 6.5 digit bench DMM with 4-wire Kelvin capability for reading Ohms and I have the 4-wire test probes. So if I want to extend my output leads, which I will probably do, I can precisely set my new leads resistance figure. This is a nice feature.

Good to hear at least someone will be making use of it! I was excited to have control over this, since cable lead resistance can have an appreciable effect on the duration of the CV phase of the charge process in a typical setup. If you have a low resistance/high power battery that is like 50mOhm total pack resistance, but the charging cable is say 150mOhm, then the constant voltage phase is hit earlier and then takes about 4x longer than it would if you had zero lead resistance.

 

[johnrobholmes]

I second having the charger stay in boot loader mode while hooked to a computer. That would be handy. Also, the super long USB is really nice. I can't always pull my bike directly to the computer, and combined with the charge cable I get 8 or 10 feet to work with.


No complaints here though, its already a million times better than any other option.

 

[Ypedal]

A simple check box on the Profile Editor would be great,

- Stay connected to Satiator

- Restart Satiator after uploading

 

[Ypedal]

Also noticed this :



Words being cut off by one letter on the first 3 and OR on satiator at the end.

--

edit : It would also be great to have a " Check For Latest Firmware Version " button that would link to a web directory and report back the current version vs the last version flashed previously. Instead of customers going on the site and finding the link themselves. ( lazy.. i know.. but it would be a neat option ) ..

Another thought .. add the Firmware button on the tool bar instead of/or as well as going to " File " menu.

 

[BVH]

Great info, thanks Justin! I hooked up to my nearly full charged pack with a Watts-up meter in-line just because I enjoy seeing lots of data. Voltage and Amperage numbers were in perfect agreement to two decimal positions. Thanks for the explanation on what happens after "Charge Complete" appears and how the "Complete Amps" setting relates. As you said, I noticed that it took about 5 minutes for current flow to ramp down and terminate after "Charge Complete" appeared. I also noticed that when I cam back an hour later with the system still running, that a small additional amount of mAh had been put in so it looks like even after current flow completely stops, it can come back on.

I must have been doing something wrong before regarding hiding profiles because now only my customs show up, which is what I want.

Thanks again for a great charger!

 

[BVH]

Going to show a bit of my lack of electrical knowledge here but here goes. I would think that cable resistance should be measured for both the Pos and Neg wires and those figures should be added for total cable resistance. Correct or not?

Just for info, I combined the factory cable, a Watts-up meter with 3" pigtails and Anderson PP's on both ends plus about 30" of old Monster brand shielded, 18 AWG audio cable with a gold plated RCA connector. Both sides read .0795 Ohms, + or - 5 Milliohms. So in this case, I would set cable resistance to 0.160 mOhms, correct?

Measuring just the factory cables (both pieces connected), I got .0490 and .0498 for a total of .098 mOhms

 

[justin_le]

Going to show a bit of my lack of electrical knowledge here but here goes. I would think that cable resistance should be measured for both the Pos and Neg wires and those figures should be added for total cable resistance. Correct or not?

Yes correct. And so your result of 0.160 Ohm is right. But you could probably set it higher in the Satiator to like 0.2 Ohms or more and then speed up the charge time spent in the constant voltage phase.

Measuring just the factory cables (both pieces connected), I got .0490 and .0498 for a total of .098 mOhms

That sounds about right (with the correction that it would be Ohms, not mOhm). By both cables I presume you mean both the DC output cable and also the XLR->Anderson adapter plugged in together? The DC output cable by itself, just to the XLR plug, measures 80-84 mOhm typically, so I'd expect the extra connection and cable length with the XLR to anderson adapter would increase this a bit.

 

[BVH]

Yes, corrected to Ohms, not mOhms. Got a little redundant there using the Decimal and the word mOhms. And yes, my numbers include the XLR->Anderson adapter. And I'll bump it up a bit as you suggest for faster CV time.

 

[mrbill]

I was recently charging two batteries, each with separate but identical BMS systems, wired in parallel using firmware version 0.902, giving the net specification:

Li-Ion (4.2v/cell max), 10s, 25Ah

Battery 1: 10s, 12Ah
Battery 2: 10s, 13Ah

I set up a Lithium charge profile with 42.0 volts as the top of charge cutoff voltage, 8 Amps charge rate, and leaving the rest of the parameters at their defaults.

When I attempted to charge fully these two batteries (wired in parallel) from a state of half-charge, I should have put in about 12 Ah or more before the charger registered "charge completed". On my first attempt, the charger stopped charging ("charge completed") after exactly 4.00 Ah were put into the battery. I restarted the charge cycle (now from 66% SoC), and again the charger stopped after putting exactly another 4.00 Ah into the battery.

When I charged the battery from near 0% SoC or 50% SoC using an identical profile but with 4 Amps charge rate, the charger fully-charged the battery in one go.

Has anyone observed this behavior, either with parallel batteries or a single battery? I have submitted a bug report to Grin, but I thought I might post my experience in case I might have overlooked something in my setup.

Thanks.

 

[Ypedal]

If one of the BMS quit or cut the charge for any reason, the other BMS will get 8 amps and likely shut down as well ( most BMS don't like much over 5 amps on the charge side if it has separate channels .. depends on BMS ) .. . somehow .. could the 8 amps not be shared equally thru both BMS even if the packs are in parallel ?

 

[mrbill]

If one of the BMS quit or cut the charge for any reason, the other BMS will get 8 amps and likely shut down as well ( most BMS don't like much over 5 amps on the charge side if it has separate channels .. depends on BMS ) .. . somehow .. could the 8 amps not be shared equally thru both BMS even if the packs are in parallel ?

Thanks, Ypedal, for replying.

The current should be split fairly equally. One battery is 4p and uses brand new Panasonic 18650 3.1Ah cells. The other uses slightly older (but still healthy) cells in 5p, Samsung 18650 2.9Ah.

I'm using the BMS's at the link below. They're cheap, and I don't believe they'll handle the advertised 60A, but they should be able to handle 15A "operating current" that they're rated for given the number FETs, and that's all I need for my application. These are 2-wire BMSs, so they have the same current rating in either charge or discharge.

http://www.ebay.com/itm/221214802862

Besides that, why would the Satiator cut off charge after exactly 4.00 Ah had been put into the battery, 4.00Ah exactly on two different sessions where the batteries were at a different SoC? That exactly 4.00 Ah were put in before the charger stopped makes me suspicious of a firmware bug and not some random temperature or connection problem.

Just to check again, I updated the charger to the latest firmware (0.905) and started another charge session at 8 Amps and observed the charger as it reached 4.00 Ah. The Satiator again cut off charge at exactly 4.00 Ah. The temperature of the charger was 52 degrees C, and the battery temperature figure on the screen was 84 degrees C, although I have no battery temperature sensor connected. The voltage was 39.8v when the charger stopped charging. While charging, one battery was receiving 4.5A, the other 3.5A.

 

[justin_le]

When I attempted to charge fully these two batteries (wired in parallel) from a state of half-charge, I should have put in about 12 Ah or more before the charger registered "charge completed". On my first attempt, the charger stopped charging ("charge completed") after exactly 4.00 Ah were put into the battery. I restarted the charge cycle (now from 66% SoC), and again the charger stopped after putting exactly another 4.00 Ah into the battery.

Hi Bill, this is most curious for sure. On the bug report, you indicated that you didn't experience this at all with a custom 4.0A profile, but that with the 8.0A profile it happened all the time? And did you try this just with the a custom made 42V 8A lithium profile or did you also use the preset 8A profile as well?

I've been trying to duplicate this behavior but so far with no success. Is there anyone else who has noticed a charge terminate at 4.00 Ah?

The other thing I found interesting is this:

The temperature of the charger was 52 degrees C, and the battery temperature figure on the screen was 84 degrees C, although I have no battery temperature sensor connected.

Battery temperature should only be showing on the NiCad/NiMH chemistry profiles, not on a lithium profile. Is it possible that you might have accidentally created a nickel based profile for these charges, and did you create them uploading from a computer or via the Satiator buttons?

It sounds like the satiator itself is running a Nickel profile with an 8A top-up current that has been left at a 0.5 hour top-up time. What else does it say on the top title bar?

 

[justin_le]

edit : It would also be great to have a " Check For Latest Firmware Version " button that would link to a web directory and report back the current version vs the last version flashed previously. Instead of customers going on the site and finding the link themselves. ( lazy.. i know.. but it would be a neat option ) ..

Wish granted! Have a download of the latest V0.96 satiator suite:
http://www.ebikes.ca/downloads/Satiator_Suite_Windows_v0.96.exe
(only windows version is complete now, expect to have MacOS and Linux early next week)

There in the file menu is new a "Check for new Firmware" link that queries our website and downloads the latest statiator bundle if there is a newer one than what is in the unpack directory of the software.

A simple check box on the Profile Editor would be great,
- Stay connected to Satiator
- Restart Satiator after uploading

Got this too. We deliberated over that implementation or simply giving the option after each read/write event on whether you wanted to reset or stay in bootloader, and opted for the latter.


So now you can read and write profiles and update firmwares all day long without the Satiator resetting after every event forcing you to put it back into bootloader mode.

 

[BVH]

Great useful improvements! Thanks!

 

[mrbill]

When I attempted to charge fully these two batteries (wired in parallel) from a state of half-charge, I should have put in about 12 Ah or more before the charger registered "charge completed". On my first attempt, the charger stopped charging ("charge completed") after exactly 4.00 Ah were put into the battery. I restarted the charge cycle (now from 66% SoC), and again the charger stopped after putting exactly another 4.00 Ah into the battery.

Hi Bill, this is most curious for sure. On the bug report, you indicated that you didn't experience this at all with a custom 4.0A profile, but that with the 8.0A profile it happened all the time? And did you try this just with the a custom made 42V 8A lithium profile or did you also use the preset 8A profile as well?

I've been trying to duplicate this behavior but so far with no success. Is there anyone else who has noticed a charge terminate at 4.00 Ah?

The other thing I found interesting is this:

The temperature of the charger was 52 degrees C, and the battery temperature figure on the screen was 84 degrees C, although I have no battery temperature sensor connected.

Battery temperature should only be showing on the NiCad/NiMH chemistry profiles, not on a lithium profile. Is it possible that you might have accidentally created a nickel based profile for these charges, and did you create them uploading from a computer or via the Satiator buttons?

It sounds like the satiator itself is running a Nickel profile with an 8A top-up current that has been left at a 0.5 hour top-up time. What else does it say on the top title bar?

Hi Justin:

Thanks for following up.

When I go to the Modify Profile screen for the LiPo 10s 8A profile, it reads across the title bar, "MODIFY LITHIUM 1/4". As I scroll through the setup screens the title bar reads, "MODIFY LITHIUM 2/4", "MODIFY LITHIUM 3/4", and "MODIFY LITHIUM 4/4".

I created this profile on the Satiator itself (unlike the other profiles that I created on my computer), then uploaded all profiles to an XML file before updating the firmware to 0.905. I then reloaded the profiles from the XML file after updating the firmware to 0.905.

Since you just released version 0.96 of the profile editor and 0.908 firmware, I updated the firmware to 0.908 and downloaded to the Satiator the same profile file containing the problem LiPo 10s/8A profile. I have run another charge session on my batteries using the 0.908 firmware, and I again observed the same outcome, although with different message on the display.

Battery Temperature was still reading 84.1C in the 29th minute of charge, although there was no thermistor connected. At exactly 30 minutes, the charger had put 3.99Ah into the batteries, and the voltage was 40.0v. But, instead of reading "CHARGE COMPLETE", I got a screen that read:

!ERROR: UN&REPLUG!
39.1v

Perhaps after a period of time displaying this message, the screen changes to "CHARGE COMPLETE". I did not wait to find out.

The relevant Satiator charge profile file can for the moment be downloaded from here:
http://mrbill.homeip.net/downloads/mrbillsProfiles.xml

The profile that causes this issue is profile id=12.

I did two trials by modifying that profile on the charger to a charge rate of 6.0A and later 4.0A. I discovered that the charger pauses charging at 30 minutes and enters a mode where the title of the display is "CALIBRATION". Voltage is shown, and current is 0.0A. After several seconds, the charger then resumes charging at the set rate, in this case 6.0A or 4.0A. With a rate of 8.0A, the charger presented the error above instead of the CALIBRATION screen.

Thinking a modifiable profile might have something to do with it, I changed profile 12 to be not modifiable and then downloaded the profiles again to the charger. I tested again using profile 12, and this time like the other trials, the charger entered a CALIBRATION mode after 30 minutes, then resumed charging at 8.0A! The early charge termination problem disappeared after I downloaded a new set of profiles where the charge profile used is not modifiable and is not custom. It also seemed to disappear when I reduced the current to 6.0A or lower on a profile modified on the Satiator.

It's possible my failures or success are merely coincident to my altering the profile and have another yet unknown cause, perhaps a glitch during the profile download process, profile modification process, or profile creation process that will never appear again.

When testing I used the middle SoC range for the two batteries, (wired in parallel) between 35% and 70% SoC, and using a power resistor network to discharge the batteries between trials. So, I doubt BMS misbehavior is to blame. I did test both batteries individually by fully cycling them several times to balance the cells and to assess the capacity. Both BMSs behaved as expected.

The battery temperature readout was present during all trials which is slightly disconcerting when there is no thermistor connected. Perhaps the information should not be displayed when no thermistor is connected, or an Advanced Settings option should be added so the user can turn off the battery temperature display.

Also, there is no mention in the manual (or earlier in this thread) of a "CALIBRATION" mode that is entered after a half-hour of charging. What is being calibrated?

You may have the greatest likelihood of duplicating this error by downloading my profile file (above) to a charger with firmware version 0.908. Then modify on the Satiator the last profile in some way (but leave the charge current at 8.0A). Then charge a 10s LiPo battery that can accept at least 4Ah of charge and observe the Satiator behavior at 29-31 minutes into the charge cycle.

I am interested to learn if you or anyone else following this thread can duplicate the fault or the early termination behavior.

 

[Alan B]

Really looking good Justin. Makes me sorry most of my packs are 18S Lipo, looking forward to a higher voltage unit some day.

 

[justin_le]

Battery Temperature was still reading 84.1C in the 29th minute of charge, although there was no thermistor connected. At exactly 30 minutes, the charger had put 3.99Ah into the batteries, and the voltage was 40.0v. But, instead of reading "CHARGE COMPLETE", I got a screen that read:
!ERROR: UN&REPLUG!
39.1v
Perhaps after a period of time displaying this message, the screen changes to "CHARGE COMPLETE". I did not wait to find out.

Hi Bill, so one of the main changes in the V0.908 build is that it is supposed to detect when the Satiator has a power stage fault and show an error message in that situation rather than a false "charge complete" message, so this detail is actually doing it's job.

The battery temperature readout was present during all trials which is slightly disconcerting when there is no thermistor connected. Perhaps the information should not be displayed when no thermistor is connected, or an Advanced Settings option should be added so the user can turn off the battery temperature display.

This is the bit that makes me think that somehow the profile structure got corrupted on the device in a way that wasn't detected, because even with a thermistor present the lithium profile is not supposed to show a battery temperature readout on the display. And that might also be we we can't reproduce this behavior since it's not the profile settings as such that caused it, rather than a glitch in the way that the profile was loaded into the satiator's memory.

Also, there is no mention in the manual (or earlier in this thread) of a "CALIBRATION" mode that is entered after a half-hour of charging. What is being calibrated?

That's a good question. Every 30 minute the Satiator briefly drops the current to zero amps in order to measure and cancel out any offset drift in the current sensor. In later firmware releases it will also be using this recalibration in order to estimate the internal resistance of the battery pack as well, by comparing the voltage both loaded and unloaded at the same SOC point. It seems that in your corrupted profile, after the first 30 minute offset recalibration the satiator's output wasn't turning on properly again, and hence you were seeing either "charge complete" (current < charge complete amps) or the power stage fault error in the latest build (since it could tell that even though current < charge complete, the voltage was not yet at the target full charge voltage).

We'll update more details as we delve into what might have been going on. Am I correct in understanding that at the moment even with the same nominal custom 8A profile you are no longer able to get this error after since reloading the profile list?

 

[mrbill]

Hi Justin:

Thanks for the explanation. It would be very cool to get an estimate for battery IR while charging and to compare it with what I see on the CAV3.

I cannot recreate the error condition with the profiles I have currently loaded. If you wish I can load the profile file at the URL in my prior message, modify the modifiable profile id=12, then see if the error reappears.

Btw, I still see battery temperature displayed when charging with any other active lithium profile.

A corrupted profile or profiles is a reasonable explanation. Which reminds me... I forgot to mention earlier that I have trouble getting a good data connection at the stereo 1/8" phone plug/jack. I need to pull the plug slightly out, then apply sideways force on it relative to the jack, otherwise the Profile Editor thinks there is no connection to the Satiator. I have to hold this firmly during any update process, and that can be a bit tedious when updating firmware as it takes a minute or two instead of a couple seconds for the profiles. I'm so in the habit of doing this now that it no longer feels extraordinary and worth mentioning, until you mentioned data corruption.

Thinking about this more I now recall that when I first updated firmware on the Satiator, I thought at one point I had "bricked" my Satiator. This was before I realized I wasn't getting good data contact at the jack. Is it possible that at that time or perhaps during some other update occasion when all seemed well, the update process (firmware or profile) wrote to a part of the Satiator's memory that isn't normally written, leading to the permanently-active "batt temp" display and possibly other irregular behavior?

Is there a process I can follow that guarantees a repair of the Satiator's memory like re-formatting a hard disk, that all writeable memory is erased and then rewritten with the correct data?

 

[cal3thousand]

I have a hardware defect issue to report. I tried using the Bug Reporter on Ebikes.ca, but have not received any feedback, so I thought I could get your attention here.

My XLR cable's #1 is burnt out, causing the charger to not recognize the battery. The male side is brown and the female side has been warped. Any clues as to why this would happen now? I have not used the charger in wet conditions yet, but the last fully operational charge was on a road trip, if that matters:


(Is this covered by a warranty?)

 

[justin_le]

My XLR cable's #1 is burnt out, causing the charger to not recognize the battery. The male side is brown and the female side has been warped. Any clues as to why this would happen now?

Ah a first but one that's not super surprising. The XLR plug on the Satiator is a proper Neutrik brand nominally rated to max 16A per pin, but the female XLR is one we've been making the adapters with is whatever we could pick up in short notice from the local electronics supply shops here in Vancouver. We did a bunch of tests with them at 8A continuous with no signs of failures, but I've seen this kind of melted pin and contact overheating with generic XLR plugs even at just 4A charge currents in the past so a lot of variation is expected. Probably what happened is that this one female pin wasn't making great contact and started arcing a little bit, and that would in turn oxidize the surface and cause things to heat even further.

Anyways if none of your packs use XLR plugs on the charging port, then I would actually recommend getting rid of the XLR altogether and just crimp andersons directly to the end of the charge cable, including a 3rd anderson for the signal pin. You could then easily rework the adapter cable to be anderon->TRS for reprogramming. Then there is no XLR in the mix at all.

Otherwise if you want a replacement cable just send us an email and we'll sort that out.

Have any others noticed signs of pin overheating on their XLR plug? At some point we'll be doing a proper manufacturing run of XLR -> ??? adapters from a cable harness company to accommodate some of the other common battery plugs, and will use matching Neutrik XLR jacks on these. So I don't expect it to be an issue in the production product in any case.